Access the full text.
Sign up today, get DeepDyve free for 14 days.
R. Hebbel, R. Osarogiagbon, D. Kaul (2004)
The Endothelial Biology of Sickle Cell Disease: Inflammation and a Chronic VasculopathyMicrocirculation, 11
V Cĕrný, Z Turek, R Pařízková (2007)
Orthogonal polarization spectral imaging: a reviewPhysiol Res, 56
B. Zitová, J. Flusser (2003)
Image registration methods: a surveyImage Vis. Comput., 21
(CreagerMLuscherTCosentinoFBechkmanJDiabetes and vascular disease: pathophysiology, clinical consequences, and medical therapy: part ICirculation20031081527153210.1161/01.CIR.0000091257.27563.3214504252)
CreagerMLuscherTCosentinoFBechkmanJDiabetes and vascular disease: pathophysiology, clinical consequences, and medical therapy: part ICirculation20031081527153210.1161/01.CIR.0000091257.27563.3214504252CreagerMLuscherTCosentinoFBechkmanJDiabetes and vascular disease: pathophysiology, clinical consequences, and medical therapy: part ICirculation20031081527153210.1161/01.CIR.0000091257.27563.3214504252, CreagerMLuscherTCosentinoFBechkmanJDiabetes and vascular disease: pathophysiology, clinical consequences, and medical therapy: part ICirculation20031081527153210.1161/01.CIR.0000091257.27563.3214504252
(ChenWSmithRJiSYWardKNajarianKAutomated ventricular systems segmentation in brain CT images by combining low-level segmentation and high level template matchingBMC Med Inform Decis Mak20099Suppl 1S410.1186/1472-6947-9-S1-S419891798)
ChenWSmithRJiSYWardKNajarianKAutomated ventricular systems segmentation in brain CT images by combining low-level segmentation and high level template matchingBMC Med Inform Decis Mak20099Suppl 1S410.1186/1472-6947-9-S1-S419891798ChenWSmithRJiSYWardKNajarianKAutomated ventricular systems segmentation in brain CT images by combining low-level segmentation and high level template matchingBMC Med Inform Decis Mak20099Suppl 1S410.1186/1472-6947-9-S1-S419891798, ChenWSmithRJiSYWardKNajarianKAutomated ventricular systems segmentation in brain CT images by combining low-level segmentation and high level template matchingBMC Med Inform Decis Mak20099Suppl 1S410.1186/1472-6947-9-S1-S419891798
A Bauer, S Kofler, M Thiel, S Elfert, F Christ (2007)
Monitoring of the sublingual microcirculation in cardiac surgery using orthogonal polarization spectral imagingThe American Society of Anesthesiologists, 107
(BatemanRSharpeMEllisCBench-to-bedside review: microvascular dysfunction in sepsis: hemodynamics, oxygen transport and nitric oxideCrit Care Med2003735937310.1186/cc2353)
BatemanRSharpeMEllisCBench-to-bedside review: microvascular dysfunction in sepsis: hemodynamics, oxygen transport and nitric oxideCrit Care Med2003735937310.1186/cc2353BatemanRSharpeMEllisCBench-to-bedside review: microvascular dysfunction in sepsis: hemodynamics, oxygen transport and nitric oxideCrit Care Med2003735937310.1186/cc2353, BatemanRSharpeMEllisCBench-to-bedside review: microvascular dysfunction in sepsis: hemodynamics, oxygen transport and nitric oxideCrit Care Med2003735937310.1186/cc2353
(NathKKatusicZGladwinMThe perfusion paradox and instability in sickle cell diseaseMicrocirculation20041117919315280091)
NathKKatusicZGladwinMThe perfusion paradox and instability in sickle cell diseaseMicrocirculation20041117919315280091NathKKatusicZGladwinMThe perfusion paradox and instability in sickle cell diseaseMicrocirculation20041117919315280091, NathKKatusicZGladwinMThe perfusion paradox and instability in sickle cell diseaseMicrocirculation20041117919315280091
(LevyBAmbrosioGPriesAStruijker-BoudierHMicrocirculation in hypertension: a new target for treatment?Circulation200110473574010.1161/hc3101.09115811489784)
LevyBAmbrosioGPriesAStruijker-BoudierHMicrocirculation in hypertension: a new target for treatment?Circulation200110473574010.1161/hc3101.09115811489784LevyBAmbrosioGPriesAStruijker-BoudierHMicrocirculation in hypertension: a new target for treatment?Circulation200110473574010.1161/hc3101.09115811489784, LevyBAmbrosioGPriesAStruijker-BoudierHMicrocirculation in hypertension: a new target for treatment?Circulation200110473574010.1161/hc3101.09115811489784
(MuralikrishnanBNajarianKRajaJSurface wavelength based clustering using neural networks for manufacturing process mappingInt J Mach Tool Manuf200343436947710.1016/S0890-6955(02)00272-9)
MuralikrishnanBNajarianKRajaJSurface wavelength based clustering using neural networks for manufacturing process mappingInt J Mach Tool Manuf200343436947710.1016/S0890-6955(02)00272-9MuralikrishnanBNajarianKRajaJSurface wavelength based clustering using neural networks for manufacturing process mappingInt J Mach Tool Manuf200343436947710.1016/S0890-6955(02)00272-9, MuralikrishnanBNajarianKRajaJSurface wavelength based clustering using neural networks for manufacturing process mappingInt J Mach Tool Manuf200343436947710.1016/S0890-6955(02)00272-9
A. Bauer, S. Kofler, M. Thiel, S. Eifert, F. Christ (2007)
Monitoring of the Sublingual Microcirculation in Cardiac Surgery Using Orthogonal Polarization Spectral Imaging: Preliminary ResultsAnesthesiology, 107
(Zitov_aBFlusserJImage registration methods: a surveyImage and Vision Computing20032111977100010.1016/S0262-8856(03)00137-9)
Zitov_aBFlusserJImage registration methods: a surveyImage and Vision Computing20032111977100010.1016/S0262-8856(03)00137-9Zitov_aBFlusserJImage registration methods: a surveyImage and Vision Computing20032111977100010.1016/S0262-8856(03)00137-9, Zitov_aBFlusserJImage registration methods: a surveyImage and Vision Computing20032111977100010.1016/S0262-8856(03)00137-9
(2001)
Validation of an analysis software for OPS-imaging used in humans
(VerdantCBackerDDMonitoring of the microcirculation may help us at the bedsideCurr OpinCrit Care200511324024410.1097/01.ccx.0000158849.94225.11)
VerdantCBackerDDMonitoring of the microcirculation may help us at the bedsideCurr OpinCrit Care200511324024410.1097/01.ccx.0000158849.94225.11VerdantCBackerDDMonitoring of the microcirculation may help us at the bedsideCurr OpinCrit Care200511324024410.1097/01.ccx.0000158849.94225.11, VerdantCBackerDDMonitoring of the microcirculation may help us at the bedsideCurr OpinCrit Care200511324024410.1097/01.ccx.0000158849.94225.11
(AtaseverBBoerCvan der KuilMQuantitative imaging of microcirculatory response during nitroglycerin-induced hypotensionJ Cardiothorac Vasc Anesth201125114014410.1053/j.jvca.2009.11.00620096602)
AtaseverBBoerCvan der KuilMQuantitative imaging of microcirculatory response during nitroglycerin-induced hypotensionJ Cardiothorac Vasc Anesth201125114014410.1053/j.jvca.2009.11.00620096602AtaseverBBoerCvan der KuilMQuantitative imaging of microcirculatory response during nitroglycerin-induced hypotensionJ Cardiothorac Vasc Anesth201125114014410.1053/j.jvca.2009.11.00620096602, AtaseverBBoerCvan der KuilMQuantitative imaging of microcirculatory response during nitroglycerin-induced hypotensionJ Cardiothorac Vasc Anesth201125114014410.1053/j.jvca.2009.11.00620096602
R. Hebbel, G. Vercellotti (1997)
The endothelial biology of sickle cell disease.The Journal of laboratory and clinical medicine, 129 3
(JiSYSmithRHuynhTNajarianKA comparative analysis of multi-level computer-assisted decision making systems for traumatic brain injuriesBMC Med Inform Decis Mak20099210.1186/1472-6947-9-219144188)
JiSYSmithRHuynhTNajarianKA comparative analysis of multi-level computer-assisted decision making systems for traumatic brain injuriesBMC Med Inform Decis Mak20099210.1186/1472-6947-9-219144188JiSYSmithRHuynhTNajarianKA comparative analysis of multi-level computer-assisted decision making systems for traumatic brain injuriesBMC Med Inform Decis Mak20099210.1186/1472-6947-9-219144188, JiSYSmithRHuynhTNajarianKA comparative analysis of multi-level computer-assisted decision making systems for traumatic brain injuriesBMC Med Inform Decis Mak20099210.1186/1472-6947-9-219144188
C. Verdant, D. Backer (2005)
How monitoring of the microcirculation may help us at the bedsideCurrent Opinion in Critical Care, 11
Xiaoyi Jiang, D. Mojon (2003)
Adaptive Local Thresholding by Verification-Based Multithreshold Probing with Application to Vessel Detection in Retinal ImagesIEEE Trans. Pattern Anal. Mach. Intell., 25
W. Groner, J. Winkelman, A. Harris, C. Ince, G. Bouma, K. Messmer, R. Nadeau (1999)
Orthogonal polarization spectral imaging: A new method for study of the microcirculationNature Medicine, 5
(DobbeJStreekstraGAtaseverBvan ZijderveldRInceCThe measurement of functional microcirculatory density and velocity distributions using automated image analysisMed BiolEngComput2008467659670)
DobbeJStreekstraGAtaseverBvan ZijderveldRInceCThe measurement of functional microcirculatory density and velocity distributions using automated image analysisMed BiolEngComput2008467659670DobbeJStreekstraGAtaseverBvan ZijderveldRInceCThe measurement of functional microcirculatory density and velocity distributions using automated image analysisMed BiolEngComput2008467659670, DobbeJStreekstraGAtaseverBvan ZijderveldRInceCThe measurement of functional microcirculatory density and velocity distributions using automated image analysisMed BiolEngComput2008467659670
B. Levy, G. Ambrosio, A. Pries, H. Struijker-Boudier (2001)
Microcirculation in Hypertension: A New Target for Treatment?Circulation: Journal of the American Heart Association, 104
Jicun Zhong, Claes Asker, E. Salerud (2000)
Imaging, image processing and pattern analysis of skin capillary ensemblesSkin Research and Technology, 6
R. Bateman, M. Sharpe, C. Ellis (2003)
Bench-to-bedside review: Microvascular dysfunction in sepsis –hemodynamics, oxygen transport, and nitric oxideCritical Care, 7
T. Klyscz, M. Jünger, F. Jung, H. Zeintl (1997)
[Cap image--a new kind of computer-assisted video image analysis system for dynamic capillary microscopy].Biomedizinische Technik. Biomedical engineering, 42 6
Z. Awan, T. Wester, Knut Kvernebo (2010)
Human microvascular imaging: a review of skin and tongue videomicroscopy techniques and analysing variablesClinical Physiology and Functional Imaging, 30
G. Balestra, Rick Bezemer, Rick Bezemer, E. Boerma, Ze-Yie Yong, K. Sjauw, Annemarie Engström, M. Koopmans, C. Ince, C. Ince (2010)
Improvement of Sidestream Dark Field Imaging with an Image Acquisition StabilizerBMC Medical Imaging, 10
K. Zuiderveld (1994)
Contrast Limited Adaptive Histogram Equalization
(BalestraGBezemerRBoermaEYongZSjauwKEngstromAKoopmansMInceCImprovement of sidestream dark field imaging with an image acquisition stabilizerBMC Med Imaging2010101510.1186/1471-2342-10-1520626888)
BalestraGBezemerRBoermaEYongZSjauwKEngstromAKoopmansMInceCImprovement of sidestream dark field imaging with an image acquisition stabilizerBMC Med Imaging2010101510.1186/1471-2342-10-1520626888BalestraGBezemerRBoermaEYongZSjauwKEngstromAKoopmansMInceCImprovement of sidestream dark field imaging with an image acquisition stabilizerBMC Med Imaging2010101510.1186/1471-2342-10-1520626888, BalestraGBezemerRBoermaEYongZSjauwKEngstromAKoopmansMInceCImprovement of sidestream dark field imaging with an image acquisition stabilizerBMC Med Imaging2010101510.1186/1471-2342-10-1520626888
Sumeyra Demir, N. Mirshahi, M. Tiba, Gerald Draucker, Kevin Ward, R. Hobson, K. Najarian (2009)
Image processing and machine learning for diagnostic analysis of microcirculation2009 ICME International Conference on Complex Medical Engineering
K. Nath, Z. Katusic, M. Gladwin (2004)
The Perfusion Paradox and Vascular Instability in Sickle Cell DiseaseMicrocirculation, 11
R Bezemer, J Dobbe, S Bartels, C Boerma, P Elbers, M Heger, C Ince (2011)
Rapid automatic assessment of microvascular density in sidestream dark field imagesMed Biol Eng Comput, 49
C Verdant, DD Backer (2005)
Monitoring of the microcirculation may help us at the bedsideCurr OpinCrit Care, 11
K Nath, Z Katusic, M Gladwin (2004)
The perfusion paradox and instability in sickle cell diseaseMicrocirculation, 11
(GoedhartPTKhalilzadaMBezemerRMerzaJInceCSidestream dark field (SDF) imaging: a novel stroboscopic LED ring-based imaging modality for clinical assessment of the microcirculationOpt Express200715151011511410.1364/OE.15.01510119550794)
GoedhartPTKhalilzadaMBezemerRMerzaJInceCSidestream dark field (SDF) imaging: a novel stroboscopic LED ring-based imaging modality for clinical assessment of the microcirculationOpt Express200715151011511410.1364/OE.15.01510119550794GoedhartPTKhalilzadaMBezemerRMerzaJInceCSidestream dark field (SDF) imaging: a novel stroboscopic LED ring-based imaging modality for clinical assessment of the microcirculationOpt Express200715151011511410.1364/OE.15.01510119550794, GoedhartPTKhalilzadaMBezemerRMerzaJInceCSidestream dark field (SDF) imaging: a novel stroboscopic LED ring-based imaging modality for clinical assessment of the microcirculationOpt Express200715151011511410.1364/OE.15.01510119550794
(NolteDZeintlHSteinbauerMPickelmannSMessmerKFunctional capillary density: an indicator of tissue perfusion?Int J MicrocircClinExp19951524424910.1159/000179025)
NolteDZeintlHSteinbauerMPickelmannSMessmerKFunctional capillary density: an indicator of tissue perfusion?Int J MicrocircClinExp19951524424910.1159/000179025NolteDZeintlHSteinbauerMPickelmannSMessmerKFunctional capillary density: an indicator of tissue perfusion?Int J MicrocircClinExp19951524424910.1159/000179025, NolteDZeintlHSteinbauerMPickelmannSMessmerKFunctional capillary density: an indicator of tissue perfusion?Int J MicrocircClinExp19951524424910.1159/000179025
(KaulDFarbyMIn vivo studies of sickle red blood cellsMicrocirculation20041115315615280089)
KaulDFarbyMIn vivo studies of sickle red blood cellsMicrocirculation20041115315615280089KaulDFarbyMIn vivo studies of sickle red blood cellsMicrocirculation20041115315615280089, KaulDFarbyMIn vivo studies of sickle red blood cellsMicrocirculation20041115315615280089
S Schaudig, B Dadasch, K Kellam, F Christ (2001)
Proceedings of the 7th World Congress for Microcirculation
(SerneEGansRter MaatenJTangelderGDonkerAStehouwerCImpaired skin capillary recruitment in essential hypertension is caused by both functional and structural capillary rarefactionHypertension20013823824210.1161/01.HYP.38.2.23811509483)
SerneEGansRter MaatenJTangelderGDonkerAStehouwerCImpaired skin capillary recruitment in essential hypertension is caused by both functional and structural capillary rarefactionHypertension20013823824210.1161/01.HYP.38.2.23811509483SerneEGansRter MaatenJTangelderGDonkerAStehouwerCImpaired skin capillary recruitment in essential hypertension is caused by both functional and structural capillary rarefactionHypertension20013823824210.1161/01.HYP.38.2.23811509483, SerneEGansRter MaatenJTangelderGDonkerAStehouwerCImpaired skin capillary recruitment in essential hypertension is caused by both functional and structural capillary rarefactionHypertension20013823824210.1161/01.HYP.38.2.23811509483
P. Lane (1996)
Sickle cell disease.Pediatric clinics of North America, 43 3
D. Kaul, M. Fabry (2004)
In Vivo Studies of Sickle Red Blood CellsMicrocirculation, 11
S Demir, N Mirshahi, M Tiba, G Draucker, K Ward, R Hobson, K Najarian (2009)
ICME International Conference on Complex Medical Engineering (CME 2009): 9-11 April 2009; Tempe
M. Stuart, R. Nagel (2004)
Sickle-cell diseaseThe Lancet, 364
(KlysczTJungerMJungFZeintlHCap image: a new kind of computer-assisted video image analysis system for dynamic capillary microscopyBiomed Tech (Berl)1997421681759312307)
KlysczTJungerMJungFZeintlHCap image: a new kind of computer-assisted video image analysis system for dynamic capillary microscopyBiomed Tech (Berl)1997421681759312307KlysczTJungerMJungFZeintlHCap image: a new kind of computer-assisted video image analysis system for dynamic capillary microscopyBiomed Tech (Berl)1997421681759312307, KlysczTJungerMJungFZeintlHCap image: a new kind of computer-assisted video image analysis system for dynamic capillary microscopyBiomed Tech (Berl)1997421681759312307
D. Backer, S. Hollenberg, Christiaan Boerma, P. Goedhart, G. Büchele, G. Ospina-Tascón, Iwan Dobbe, C. Ince (2007)
How to evaluate the microcirculation: report of a round table conferenceCritical Care, 11
(ZuiderveldKHeckbert PSContrast limited adaptive histogram equalizationGraphics Gems IV1994San Francisco: Morgan Kaufmann474485)
ZuiderveldKHeckbert PSContrast limited adaptive histogram equalizationGraphics Gems IV1994San Francisco: Morgan Kaufmann474485ZuiderveldKHeckbert PSContrast limited adaptive histogram equalizationGraphics Gems IV1994San Francisco: Morgan Kaufmann474485, ZuiderveldKHeckbert PSContrast limited adaptive histogram equalizationGraphics Gems IV1994San Francisco: Morgan Kaufmann474485
B. Atasever, C. Boer, Marjolein Kuil, Eric Lust, A. Beishuizen, R. Speekenbrink, J. Seyffert, B. Mol, C. Ince (2011)
Quantitative imaging of microcirculatory response during nitroglycerin-induced hypotension.Journal of cardiothoracic and vascular anesthesia, 25 1
Paul Heckbert (1994)
Graphics gems IV
O. Genzel‐Boroviczény, J. Strötgen, A. Harris, K. Messmer, F. Christ (2002)
Orthogonal Polarization Spectral Imaging (OPS): A Novel Method to Measure the Microcirculation in Term and Preterm Infants TranscutaneouslyPediatric Research, 51
(SainthillierJGharbiTMuretPHumbertPSkin capillary network recognition and analysis by means of neural algorithmsSkin Res Technol20051191610.1111/j.1600-0846.2005.00087.x15691254)
SainthillierJGharbiTMuretPHumbertPSkin capillary network recognition and analysis by means of neural algorithmsSkin Res Technol20051191610.1111/j.1600-0846.2005.00087.x15691254SainthillierJGharbiTMuretPHumbertPSkin capillary network recognition and analysis by means of neural algorithmsSkin Res Technol20051191610.1111/j.1600-0846.2005.00087.x15691254, SainthillierJGharbiTMuretPHumbertPSkin capillary network recognition and analysis by means of neural algorithmsSkin Res Technol20051191610.1111/j.1600-0846.2005.00087.x15691254
(SchaudigSDadaschBKellamKChristFValidation of an analysis software for OPS-imaging used in humansProceedings of the 7th World Congress for Microcirculation2001Sydney: Australian and New Zealand Microcirculation Society259)
SchaudigSDadaschBKellamKChristFValidation of an analysis software for OPS-imaging used in humansProceedings of the 7th World Congress for Microcirculation2001Sydney: Australian and New Zealand Microcirculation Society259SchaudigSDadaschBKellamKChristFValidation of an analysis software for OPS-imaging used in humansProceedings of the 7th World Congress for Microcirculation2001Sydney: Australian and New Zealand Microcirculation Society259, SchaudigSDadaschBKellamKChristFValidation of an analysis software for OPS-imaging used in humansProceedings of the 7th World Congress for Microcirculation2001Sydney: Australian and New Zealand Microcirculation Society259
(AwanZAWesterTKverneboKHuman microvascular imaging: a review of skin and tongue videomicroscopy techniques and analysing variablesClin Physiol Funct Imaging201030798810.1111/j.1475-097X.2009.00913.x20041906)
AwanZAWesterTKverneboKHuman microvascular imaging: a review of skin and tongue videomicroscopy techniques and analysing variablesClin Physiol Funct Imaging201030798810.1111/j.1475-097X.2009.00913.x20041906AwanZAWesterTKverneboKHuman microvascular imaging: a review of skin and tongue videomicroscopy techniques and analysing variablesClin Physiol Funct Imaging201030798810.1111/j.1475-097X.2009.00913.x20041906, AwanZAWesterTKverneboKHuman microvascular imaging: a review of skin and tongue videomicroscopy techniques and analysing variablesClin Physiol Funct Imaging201030798810.1111/j.1475-097X.2009.00913.x20041906
(NajarianKSplinterRBiomedical Signal and Image Processing2005Boca Raton: The CRC Press)
NajarianKSplinterRBiomedical Signal and Image Processing2005Boca Raton: The CRC PressNajarianKSplinterRBiomedical Signal and Image Processing2005Boca Raton: The CRC Press, NajarianKSplinterRBiomedical Signal and Image Processing2005Boca Raton: The CRC Press
(JiangXMojonDAdaptive local thresholding by verification-based multithreshold probing with application to vessel detection in retinal imagesIEEE Trans Pattern Anal Mach Intell20032513113710.1109/TPAMI.2003.1159954)
JiangXMojonDAdaptive local thresholding by verification-based multithreshold probing with application to vessel detection in retinal imagesIEEE Trans Pattern Anal Mach Intell20032513113710.1109/TPAMI.2003.1159954JiangXMojonDAdaptive local thresholding by verification-based multithreshold probing with application to vessel detection in retinal imagesIEEE Trans Pattern Anal Mach Intell20032513113710.1109/TPAMI.2003.1159954, JiangXMojonDAdaptive local thresholding by verification-based multithreshold probing with application to vessel detection in retinal imagesIEEE Trans Pattern Anal Mach Intell20032513113710.1109/TPAMI.2003.1159954
(BackerDDHollenbergSBoermaCGoedhartPBücheleGOspina-TasconGDobbeIInceCHow to evaluate the microcirculation: report of a round table conferenceCrit Care2007115R10110.1186/cc611817845716)
BackerDDHollenbergSBoermaCGoedhartPBücheleGOspina-TasconGDobbeIInceCHow to evaluate the microcirculation: report of a round table conferenceCrit Care2007115R10110.1186/cc611817845716BackerDDHollenbergSBoermaCGoedhartPBücheleGOspina-TasconGDobbeIInceCHow to evaluate the microcirculation: report of a round table conferenceCrit Care2007115R10110.1186/cc611817845716, BackerDDHollenbergSBoermaCGoedhartPBücheleGOspina-TasconGDobbeIInceCHow to evaluate the microcirculation: report of a round table conferenceCrit Care2007115R10110.1186/cc611817845716
(HebbelROsarogiagbonRKaulDThe endothelial biology of sickle cell disease; inflammation and chronic vasculopathyMicrocirculation20041112915115280088)
HebbelROsarogiagbonRKaulDThe endothelial biology of sickle cell disease; inflammation and chronic vasculopathyMicrocirculation20041112915115280088HebbelROsarogiagbonRKaulDThe endothelial biology of sickle cell disease; inflammation and chronic vasculopathyMicrocirculation20041112915115280088, HebbelROsarogiagbonRKaulDThe endothelial biology of sickle cell disease; inflammation and chronic vasculopathyMicrocirculation20041112915115280088
Virginia Commonwealth University Reanimation Engineering Science (VCURES)
J. Sainthillier, T. Gharbi, P. Muret, P. Humbert (2005)
Skin capillary network recognition and analysis by means of neural algorithmsSkin Research and Technology, 11
Med Biol Eng Comput (2011) 49:1269–1278 DOI 10.1007/s11517-011-0824-1 ORIGINAL ARTICLE
E. Serné, R. Gans, J. Maaten, G. Tangelder, A. Donker, C. Stehouwer (2001)
Impaired Skin Capillary Recruitment in Essential Hypertension Is Caused by Both Functional and Structural Capillary RarefactionHypertension: Journal of the American Heart Association, 38
(ZhongJAskerCSalerudEImaging, image processing and pattern analysis of skin capillary ensemblesSkin Res Technol20006455710.1034/j.1600-0846.2000.006002045.x11428942)
ZhongJAskerCSalerudEImaging, image processing and pattern analysis of skin capillary ensemblesSkin Res Technol20006455710.1034/j.1600-0846.2000.006002045.x11428942ZhongJAskerCSalerudEImaging, image processing and pattern analysis of skin capillary ensemblesSkin Res Technol20006455710.1034/j.1600-0846.2000.006002045.x11428942, ZhongJAskerCSalerudEImaging, image processing and pattern analysis of skin capillary ensemblesSkin Res Technol20006455710.1034/j.1600-0846.2000.006002045.x11428942
V. Černý, Z. Turek, R. Pařízková (2007)
Orthogonal polarization spectral imaging.Physiological research, 56 2
(BauerAKoflerSThielMElfertSChristFMonitoring of the sublingual microcirculation in cardiac surgery using orthogonal polarization spectral imagingThe American Society of Anesthesiologists20071076939945)
BauerAKoflerSThielMElfertSChristFMonitoring of the sublingual microcirculation in cardiac surgery using orthogonal polarization spectral imagingThe American Society of Anesthesiologists20071076939945BauerAKoflerSThielMElfertSChristFMonitoring of the sublingual microcirculation in cardiac surgery using orthogonal polarization spectral imagingThe American Society of Anesthesiologists20071076939945, BauerAKoflerSThielMElfertSChristFMonitoring of the sublingual microcirculation in cardiac surgery using orthogonal polarization spectral imagingThe American Society of Anesthesiologists20071076939945
W. Rgn, Rm (1996)
Sickle cell diseasePractice Nursing, 7
S. Cerutti, G. Baselli, A. Bianchi, E. Caiani, D. Contini, R. Cubeddu, F. Dercole, L. Rienzo, D. Liberati, L. Mainardi, P. Ravazzani, Sergio Rinaldi, M. Signorini, A. Torricelli (2011)
Biomedical Signal and Image ProcessingIEEE Pulse, 2
P. Goedhart, M. Khalilzada, R. Bezemer, J. Merza, C. Ince (2007)
Sidestream Dark Field (SDF) imaging: a novel stroboscopic LED ring-based imaging modality for clinical assessment of the microcirculation.Optics express, 15 23
R. Touyz (2005)
Intracellular mechanisms involved in vascular remodelling of resistance arteries in hypertension: role of angiotensin IIExperimental Physiology, 90
B. Muralikrishnan, J. Raja, K. Najarian (2003)
Surface wavelength content based clustering using neural networks for manufacturing process mappingInternational Journal of Machine Tools & Manufacture, 43
(Genzel-BoroviczenyOStrotgenJHarrisAGMessmerKChristFOrthogonal polarization spectral imaging (OPS): a novel method to measure the microcirculation in term and preterm infants transcutaneouslyPediatr Res20025138639110.1203/00006450-200203000-0001911861946)
Genzel-BoroviczenyOStrotgenJHarrisAGMessmerKChristFOrthogonal polarization spectral imaging (OPS): a novel method to measure the microcirculation in term and preterm infants transcutaneouslyPediatr Res20025138639110.1203/00006450-200203000-0001911861946Genzel-BoroviczenyOStrotgenJHarrisAGMessmerKChristFOrthogonal polarization spectral imaging (OPS): a novel method to measure the microcirculation in term and preterm infants transcutaneouslyPediatr Res20025138639110.1203/00006450-200203000-0001911861946, Genzel-BoroviczenyOStrotgenJHarrisAGMessmerKChristFOrthogonal polarization spectral imaging (OPS): a novel method to measure the microcirculation in term and preterm infants transcutaneouslyPediatr Res20025138639110.1203/00006450-200203000-0001911861946
Iwan Dobbe (2012)
' s response to reviews Title : An automated Method for Analysis of Microcirculation Videos for Accurate Assessment of Tissue Perfusion
D. Nolte, H. Zeintl, M. Steinbauer, S. Pickelmann, K. Messmer (1995)
Functional capillary density: an indicator of tissue perfusion?International journal of microcirculation, clinical and experimental, 15 5
(GronerWWinkelmanJWHarrisAGInceCBoumaGJMessmerKNadeauRGOrthogonal polarization spectral imaging: a new method for study of the microcirculationNatmed1999512091212)
GronerWWinkelmanJWHarrisAGInceCBoumaGJMessmerKNadeauRGOrthogonal polarization spectral imaging: a new method for study of the microcirculationNatmed1999512091212GronerWWinkelmanJWHarrisAGInceCBoumaGJMessmerKNadeauRGOrthogonal polarization spectral imaging: a new method for study of the microcirculationNatmed1999512091212, GronerWWinkelmanJWHarrisAGInceCBoumaGJMessmerKNadeauRGOrthogonal polarization spectral imaging: a new method for study of the microcirculationNatmed1999512091212
(2007)
Received: 16 November 2011 Accepted: 17 December 2012 Published: 21 December 2012 References 1. Cĕrný V, Turek Z, Pařízková R: Orthogonal polarization spectral imaging: a review
(TouyzRIntracellular mechanisms involved in vascular remodeling of resistance arteries in hypertension: role of angiotensin IIExp Physiol20059044945510.1113/expphysiol.2005.03008015890798)
TouyzRIntracellular mechanisms involved in vascular remodeling of resistance arteries in hypertension: role of angiotensin IIExp Physiol20059044945510.1113/expphysiol.2005.03008015890798TouyzRIntracellular mechanisms involved in vascular remodeling of resistance arteries in hypertension: role of angiotensin IIExp Physiol20059044945510.1113/expphysiol.2005.03008015890798, TouyzRIntracellular mechanisms involved in vascular remodeling of resistance arteries in hypertension: role of angiotensin IIExp Physiol20059044945510.1113/expphysiol.2005.03008015890798
(BezemerRDobbeJBartelsSBoermaCElbersPHegerMInceCRapid automatic assessment of microvascular density in sidestream dark field imagesMed Biol Eng Comput2011491269127810.1007/s11517-011-0824-121879345)
BezemerRDobbeJBartelsSBoermaCElbersPHegerMInceCRapid automatic assessment of microvascular density in sidestream dark field imagesMed Biol Eng Comput2011491269127810.1007/s11517-011-0824-121879345BezemerRDobbeJBartelsSBoermaCElbersPHegerMInceCRapid automatic assessment of microvascular density in sidestream dark field imagesMed Biol Eng Comput2011491269127810.1007/s11517-011-0824-121879345, BezemerRDobbeJBartelsSBoermaCElbersPHegerMInceCRapid automatic assessment of microvascular density in sidestream dark field imagesMed Biol Eng Comput2011491269127810.1007/s11517-011-0824-121879345
(2005)
Sidestream dark field imaging: an improved technique to observe sublingual microcirculation
Soo-Yeon Ji, Rebecca Smith, T. Huynh, K. Najarian (2009)
A comparative analysis of multi-level computer-assisted decision making systems for traumatic injuriesBMC Medical Informatics and Decision Making, 9
M. Creager, T. Lüscher, F. Cosentino, J. Beckman (2003)
Diabetes and Vascular Disease: Pathophysiology, Clinical Consequences, and Medical Therapy: Part ICirculation: Journal of the American Heart Association, 108
(2008)
The measurement of functional microcirculatory density and velocity distributions using automated image analysis
(StuartMNagelRSickle cell diseaseLancet20043641343136010.1016/S0140-6736(04)17192-415474138)
StuartMNagelRSickle cell diseaseLancet20043641343136010.1016/S0140-6736(04)17192-415474138StuartMNagelRSickle cell diseaseLancet20043641343136010.1016/S0140-6736(04)17192-415474138, StuartMNagelRSickle cell diseaseLancet20043641343136010.1016/S0140-6736(04)17192-415474138
(InceCSidestream dark field imaging: an improved technique to observe sublingual microcirculationCrit Care2005972)
InceCSidestream dark field imaging: an improved technique to observe sublingual microcirculationCrit Care2005972InceCSidestream dark field imaging: an improved technique to observe sublingual microcirculationCrit Care2005972, InceCSidestream dark field imaging: an improved technique to observe sublingual microcirculationCrit Care2005972
(DemirSMirshahiNTibaMDrauckerGWardKHobsonRNajarianKImage processing and machine learning for diagnostic analysis of microcirculationICME International Conference on Complex Medical Engineering (CME 2009): 9-11 April 2009; Tempe2009IEEE15)
DemirSMirshahiNTibaMDrauckerGWardKHobsonRNajarianKImage processing and machine learning for diagnostic analysis of microcirculationICME International Conference on Complex Medical Engineering (CME 2009): 9-11 April 2009; Tempe2009IEEE15DemirSMirshahiNTibaMDrauckerGWardKHobsonRNajarianKImage processing and machine learning for diagnostic analysis of microcirculationICME International Conference on Complex Medical Engineering (CME 2009): 9-11 April 2009; Tempe2009IEEE15, DemirSMirshahiNTibaMDrauckerGWardKHobsonRNajarianKImage processing and machine learning for diagnostic analysis of microcirculationICME International Conference on Complex Medical Engineering (CME 2009): 9-11 April 2009; Tempe2009IEEE15
Wenan Chen, Rebecca Smith, Soo-Yeon Ji, Kevin Ward, K. Najarian (2009)
Automated ventricular systems segmentation in brain CT images by combining low-level segmentation and high-level template matchingBMC Medical Informatics and Decision Making, 9
(CĕrnýVTurekZPařízkováROrthogonal polarization spectral imaging: a reviewPhysiol Res20075614114716555953)
CĕrnýVTurekZPařízkováROrthogonal polarization spectral imaging: a reviewPhysiol Res20075614114716555953CĕrnýVTurekZPařízkováROrthogonal polarization spectral imaging: a reviewPhysiol Res20075614114716555953, CĕrnýVTurekZPařízkováROrthogonal polarization spectral imaging: a reviewPhysiol Res20075614114716555953
Nitin Mukerji, Dorothy Wallace, Dipayan Mitra (2006)
Audit of the change in the on-call practices in neuroradiology and factors affecting itBMC Medical Imaging, 6
Background: Imaging of the human microcirculation in real-time has the potential to detect injuries and illnesses that disturb the microcirculation at earlier stages and may improve the efficacy of resuscitation. Despite advanced imaging techniques to monitor the microcirculation, there are currently no tools for the near real-time analysis of the videos produced by these imaging systems. An automated system tool that can extract microvasculature information and monitor changes in tissue perfusion quantitatively might be invaluable as a diagnostic and therapeutic endpoint for resuscitation. Methods: The experimental algorithm automatically extracts microvascular network and quantitatively measures changes in the microcirculation. There are two main parts in the algorithm: video processing and vessel segmentation. Microcirculatory videos are first stabilized in a video processing step to remove motion artifacts. In the vessel segmentation process, the microvascular network is extracted using multiple level thresholding and pixel verification techniques. Threshold levels are selected using histogram information of a set of training video recordings. Pixel-by-pixel differences are calculated throughout the frames to identify active blood vessels and capillaries with flow. Results: Sublingual microcirculatory videos are recorded from anesthetized swine at baseline and during hemorrhage using a hand-held Side-stream Dark Field (SDF) imaging device to track changes in the microvasculature during hemorrhage. Automatically segmented vessels in the recordings are analyzed visually and the functional capillary density (FCD) values calculated by the algorithm are compared for both health baseline and hemorrhagic conditions. These results were compared to independently made FCD measurements using a well-known semi-automated method. Results of the fully automated algorithm demonstrated a significant decrease of FCD values. Similar, but more variable FCD values were calculated using a commercially available software program requiring manual editing. Conclusions: An entirely automated system for analyzing microcirculation videos to reduce human interaction and computation time is developed. The algorithm successfully stabilizes video recordings, segments blood vessels, identifies vessels without flow and calculates FCD in a fully automated process. The automated process provides an equal or better separation between healthy and hemorrhagic FCD values compared to currently available semi-automatic techniques. The proposed method shows promise for the quantitative measurement of changes occurring in microcirculation during injury. * Correspondence: knajarian@vcu.edu Signal Processing Technologies LLC, Richmond, VA, USA Department of Emergency Medicine, Virginia Commonwealth University, Richmond, VA, USA Full list of author information is available at the end of the article © 2012 Demir et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Demir et al. BMC Medical Imaging 2012, 12:37 Page 2 of 13 http://www.biomedcentral.com/1471-2342/12/37 Background JavaCap and Capilap Toolbox are two other available Understanding the distribution and circulation of blood software tools which use triangulation methods to calcu- in capillaries has been considered a key aspect for as- late intercapillary distance [22,23]. sessment of tissue perfusion [1]. Visualization and quan- Automated Vascular Analysis (AVA)—also known as tification of changes in microcirculation has been proposed MAS (Microvascular Analysis Software, Microvision as a potential tool in diagnosis and treatment of illnesses Medical BV)—is the most current commercial software and diseases such as sepsis [2], sickle cell disease [3,4], tool developed by Dobbe et al [24] for analysis of micro- chronic ulcers, diabetes mellitus and hypertension [5,6]. In circulation videos. The method is the most accurate each of these diseases, several characteristics of the micro- among the existing systems and performs a semi- circulation such as the structure of capillaries and quality of automated process based on image stabilization, center- blood flow in the capillaries change over time [7-11]. A line detection and space time diagram. Despite all the recent study suggests there is value in monitoring the capabilities provided by AVA, it does not provide full microcirculation for titrating vasodilators in perioperative automation which leaves the burden of selecting the use [12]. Monitoring microcirculation during resuscitation areas of interest, configuration, initialization, filtering of could also be envisioned as a tool to prevent over and many false positives, dealing with many false negatives, under resuscitation of victims of hemorrhage and other and addressing of connectivity to the user. In addition, critical illness and injuries such as sepsis. Therefore, quanti- according to the developers of AVA, the software pro- tative and accurate analysis of the microcirculation is likely vides neither automatic vessel detection nor vessel diam- to be essential if microcirculatory imaging is to be adopted eter and blood flow calculation [24]. The system, while as useful tool in clinical monitoring [13]. an improvement, requires manual editing which can take Orthogonal polarization spectral (OPS) imaging [1,2,5] over 20 minutes for a typical video sequence. and side-stream dark field (SDF) imaging [13,14] have The aim of this study is to automate the analysis of been extensively employed in the field of clinical micro- microcirculatory video recordings and the derivation of circulatory research. OPS and SDF imaging are both Functional Capillary Density (FCD). FCD is defined as the non-invasive imaging modalities and have been used to ratio of the area of functionally active capillaries to the en- track changes in the microcirculation on mucosal sur- tire area of the image. FCD has been considered an im- faces. Most studies have used the sublingual surfaces of portant measurement of the microcirculation to indicate the oral cavity. These imaging techniques use green the quality of tissue perfusion [25]. Image processing algo- polarized light with wavelength of 550 nm which is rithms are designed for this study to automatically detect absorbed by hemoglobin and makes red blood cells capillaries and small blood vessels in order to derive diag- visible [15,16]. nostically useful information that may assist clinicians and To analyze microcirculatory images and videos, sev- medical researchers in the future. eral software tools have been developed. However cur- rently available software is unable to perform real or Methods near real time analysis of the videos and require manual The methodology behind the proposed algorithm to intervention to ensure accurate results. Researchers in quantify the assessment of the microcirculation is sum- the field have stated the need for improvements of marized in Figure 1. The process—which is not shown current software to expedite clinical bedside use [17]. in the schematic diagram—starts with the stabilization The computer-assisted image analysis system CapImage of the video frames. The weighted mean of consecutive (Zeintl, Heidelberg, Germany) was originally developed frames from the stabilized video is calculated for each for traditional intravital microscopy [18], but is capable five-frame block. Pre-processing, multi-thresholding of analysis of SDF and OPS images [19]. It uses a Line and vessel segmentation are the following steps and are Shift Diagram Method for measurement of velocity and highlighted in the diagram. A more detailed diagram is real-time movement correlation. This software tool is provided in Figure 2 to describe pre-processing, multi- capable of measuring different properties of the micro- thresholding and segmentation algorithm. After per- circulation such as blood cell velocity and capillary forming morphological operations such as filling and density. However, it is only capable of detecting straight opening, thebinaryimagesresulting from segmentation blood vessels which limits its efficacy since the micro- are unioned together. If a pixel is segmented as vessel in vascular geometry is complex. Expert users of CapImage more than one frame, it is assigned the label of “vessel”. claim that analysis of microcirculation with CapImage is Post processing includes additional morphological time consuming and may only be performed off-line [20]. operations (e.g., bridge, spur, fill) and region growing to CapiScope, a system for the measurement of capillary eliminate any possible discontinuities. Vessels with morphology and capillary blood cell velocity, requires blood flow—perfused vessels—are identified in the last stable images, but lacks a stabilization function [21]. step and FCD is calculated accordingly. Demir et al. BMC Medical Imaging 2012, 12:37 Page 3 of 13 http://www.biomedcentral.com/1471-2342/12/37 Figure 1 Schematic of the proposed methodology. Proposed methodology is summarized in Figure 1. Stabilization of the videos is not included in the schematics. After stabilization, the weighted mean of five consecutive frames is calculated and the preprocessing and segmentation algorithms are applied on the mean frame. Averaging more than five frames results in “over-averaging” phenomena that would eliminate some of the important features important for segmentation. Experimentally, our results indicate that the 5-frame approach provides the best results. After the segmentation, the segmented frames are combined together to generate one single binary image and calculate Functional Capillary Density from this binary image. Stabilization the images, which are typically the branching points of The microcirculatory videos are captured by a commer- blood vessels, are selected and assigned as control points cially available hand-held SDF device (Microscan, Micro- to overcome remarkableness issues. Control points are vision Medical, BV). Because it is hand-held and is selected which are known to belong to capillaries to calcu- highly susceptible to motion artifacts due to movement late the transformation between two consecutive frames. of subject and/or device, we developed a stabilization al- For this purpose, a 3-by-3 Laplacian filter which calculates gorithm to eliminate these motion artifacts. A block second order derivatives is applied to the output of the matching algorithm is developed that calculates cross- Gaussian Gradient. correlation coefficients to measure the similarity of the The maximum values from seven areas of the frame blocks. Block matching algorithms use a predefined size are selected as control points. Then, around the control of windows-blocks or even entire images to estimate points, 25 × 25 pixel windows are selected as sub- motion vectors. One of the disadvantages of block regions. The cross-correlation is calculated between matching methods is defined as 'remarkableness' of the these sub-regions in the current frame and a 40 × 40 window content [26]. If a window does not contain dis- pixel window around these sub-regions in the following tinctive details, there is a high probability of mismatch. frame. The size of the windows discussed above was To avoid errors caused by ‘remarkableness’ issues, i.e. optimized based on a previous visual empirical assess- using points for matching that have no significant image- ment of the algorithm over a set of microcirculation processing values (e.g. regular pixels inside the back- videos. The frames are registered according to the ground) the processed blocks are checked to ensure they results of correlation calculations. Since the control include blood vessels using Laplacian of Gaussian filtering. points are defined for the first frame and they are The stabilization process is described in detail in a previ- tracked through adjacent frames, Laplacian filtering is ously published study [27]. Gradients of the frames are not repeated throughout the algorithm. If any of the calculated using the first order derivative of the Gaussian defined control points leaves the current frame due to function. Gradient of the Gaussian enhances images and excessive motion, new control points are defined using improves visibility of blood vessels. Distinctive features of thesamemethod. Demir et al. BMC Medical Imaging 2012, 12:37 Page 4 of 13 http://www.biomedcentral.com/1471-2342/12/37 Figure 2 Detailed diagram of pre-processing and vessel segmentation. Figure 2 provides a detailed diagram of preprocessing and segmentation steps for different threshold levels. It starts with the averaged frame. For 10 different threshold levels, the parameters of CLAHE (Contrast Limited Adaptive Histogram Equalization) and median filter vary throughout the process. For the first threshold level, the window size of CLAHE is kept small. Median filter is applied right after histogram equalization with a small filter size such as 3 × 3. Median filtering is followed by image adjustment. The preprocessed image is converted to binary image using the first threshold level. Euclidean Distance Transform (EDT) is calculated for the binary image. Diameter and angle parameters are extracted from EDT and with the addition of contrast ratio; three parameters are used to determine if a pixel belongs to a vessel. Preprocessing, multi-thresholding, and segmentation the local contrast, Contrast Limited Adaptive Histogram After the stabilization of the video recordings, the Equalization (CLAHE) is performed on microcirculation weighted mean of five consecutive frames is calculated images. CLAHE partitions the image into small regions, to improve connectivity of the blood vessels. The frame called ‘tiles’, and applies histogram equalization to each in the middle has the highest weight in this process. tile in order to even out the overall gray level distribu- Averaging the frames is followed by preprocessing algo- tion of the image [28]. Histogram equalization is fol- rithms. The vessel extraction algorithm is based on mul- lowed by median filtering. To remove noise, median tiple level thresholding. Preprocessing is repeated at filtering is a widely preferred method in the literature each threshold with different parameters. [29]. In this application, the purpose of applying median Video contrast and clarity vary widely from source to filtering is smoothing. source and necessitate preprocessing in order to gener- The window sizes of adaptive histogram equalization ate images that will yield accurate results. To enhance and median filtering are subject to change at each Demir et al. BMC Medical Imaging 2012, 12:37 Page 5 of 13 http://www.biomedcentral.com/1471-2342/12/37 threshold level. At low threshold levels, in order to in- clude only wide and clear vessels, histogram equalization is applied in smaller windows. The median filter size is kept large at these low threshold levels. As the threshold level increases, the result of the process is a darker binary image with almost all vessels and background included. Window size of adaptive histogram equalization is increased and median filter size is reduced to enhance the thinner vessels. Without these steps results will suffer as vessels may not be fully segmented and flow correctly Figure 3 Method of validating vessel pixels. A vessel candidate detected. This preprocessing is followed by thresholding. pixel is labeled as p in Figure 3. The output of EDT is used to find the Vessel segmentation is based on verifying each pixel at nearest background pixel to p, b . For each of the 24 neighboring multiple threshold levels as vessel. The method is modi- pixels in the 5 × 5 neighborhood around p, n − n , the nearest 1 24 fied from a pixel verification method proposed by Jiang background pixel is found, b , and used to calculate the diameter, et al. using retinal images [30]. The pre-processed angle and contrast ratio values. The b having the greatest n1-24 distance from b is considered the opposite background pixel, b , p max images are converted to binary images using multiple and the distance is the diameter, d, of the vessel. If d is less than P threshold levels resulting in multiple binary images. Eu- then the angle, θ, is calculated between b , p and b and is used to p max clidean Distance Transform (EDT) is created for each validate the distance by ensuring that b and b are on opposite p max binary image. EDT calculates the distance of nearest sides of the vessel (θ must be greater than P ). Finally, the contrast background pixel for each object pixel. The coordinates ratio between p and b is calculated and if greater than P , the max c candidate pixel is considered a valid vessel pixel. Since the found of the nearest background pixel is the second output of vessels lie along the center of the actual vessel, the vessel must be the transform. For each pixel of each binary image, three reconstructed using the found diameter and pixel locations. different features are calculated using the outputs of the EDT and the gray-level intensity values of the pixels. Two of these parameters are the diameter P and the the angle, θ, is calculated between b , p and b . The d p max angle P of the vessel, which are considered as geomet- maximum angle derived from the 24 neighbor pixels is rical features. The gray level intensity values of pre- used as ‘θ’: processed images are used to calculate the third feature, 2 2 p; b þ p; b d P , which is the contrast ratio identifying the ratio of in- p j θ ¼ max cos bj∈b b n1 n24 tensity across background and blood vessel pixels. These 2 p; b p; b p j three features serve to determine if a pixel in the image ð2Þ is indeed a vessel pixel. Pd limits the size of the vessel to ensure it is a capillary. P ensures curvilinear structure to the pixel, and excludes any anomalous pixels because Finally the third feature is the ratio of gray level values of physiological improbabilities. Pc, ensures contrast be- of nearest background pixels and the vessel candidate ’p’: tween background and vessel pixels. Figure 3 provides a visual depiction of these para- GL b C ¼ max ð3Þ bj∈b b n1 n24 meters. 5*5 neighborhood of a candidate pixel is GLðÞ p included in the calculations. The current pixel is referred to as ‘p’, which is a vessel candidate. Its 24 neighbors are where GL(p) is the gray-level intensity value of current defined as N − N . The nearest background pixel to ‘p’ pixel ’p’. To calculate contrast ratio, images with enhanced 1 24 is ‘b ’ and the nearest background pixels to each of the contrast are used. Therefore, GL in Equation 3 stands for 24 neighbors are ‘b − b ’. The b having the great- the gray level of the output of the CLAHE. n1 n24 n1-24 est distance from b is considered the opposite back- To verify the vessel candidate pixel, ’d’ needs to be less ground pixel, b . The maximum Euclidean distance than pre-defined P to avoid large vessels, ’θ’ needs to be max d from ‘b ’ to ‘b -b ’ is decided to be the diameter of larger than P to assure curvilinear structure and the p n1 n24 θ the candidate pixel: calculated contrast ratio needs to be higher than P to remove background noise. If the pixel in the binary d ¼ max b ; b ð1Þ bj∈b b p j image is black and it meets the criteria defined by three n1 n24 parameters P , P and P , it is verified to be a vessel d θ C where b ; b is the distance between b and b.The p j p j pixel. After repeating the same procedure for all thresh- parameter θ is the angle between the background pixels old levels, the segmented images of each threshold level ‘b ’ and ‘b -b ’ according to Figure 3. The angle is cal- are combined resulting in one segmented binary image p n1 n24 culated using the cosine rule. If d is less than P then for each frame. d Demir et al. BMC Medical Imaging 2012, 12:37 Page 6 of 13 http://www.biomedcentral.com/1471-2342/12/37 The parameters, P , P and P were selected from hand, since it incorporates the thickness of the capillar- d θ C multiple experiments using a set of training videos. P ies into the calculation, is not susceptible to this issue. controls the maximum diameter of the blood vessels to We have also included the length-based FCD calculation accept. It is determined based on the diameter of blood in this paper for comparison with the output from AVA. vessels and capillaries to be included in the microcircu- lation. P is another geometric parameter to ensure Results and discussion curvilinear structure of the vessels and is empirically Results derived. The proposed experimental algorithm and the software product Microcirculation Analyzer (MCA) were applied Region growing to videos acquired from a library of microcirculatory Since the segmentation algorithm is based on pixel veri- videos of a previous animal study. The protocol was fication, it is possible to have isolated pixels in the re- approved by the Virginia Commonwealth University In- sult. To prevent that, binary morphological operators stitutional Animal Care and Use Committee in accord- are used before region growing. The morphological ance with the National Institutes of Health Guide for the operators include filling the isolated interior pixels and Care and Use of Laboratory Animals (National Institutes opening. of Health Publication 86-23, revised 1996). In this ani- The region growing algorithm is developed to over- mal study, sublingual microcirculatory videos were taken come disconnectivity of blood vessels. First, the final from nine healthy juvenile swine at baseline as well as segmented image is divided into 35∗35 windows to de- after 40% of the animal’s blood was removed. All animals termine the orientation of segmented vessel within the were under a state of general anesthesia. Twenty frames window. The vessel is allowed to grow in the computed from each video were used for the assessment of the direction if the gray level is within the range of average video recordings. The parameters at multi-thresholding gray level of vessel pixels in the window ±0.5∗standard stage are defined empirically. Specifically, a series of vid- deviation. eos were used as the training set in which we change the values of these parameters over a reasonable range and FCD calculation choose the values that give the parameters providing the Segmentation processes described up until this stage de- best segmentation results. In this study, “the best tect all blood vessels at each frame of the video record- segmentation result” was visually evaluated. According ing. To provide quantitative information on blood flow, to this process the image features are obtained; P =13, the vessels through which blood is flowing must be iden- P =130 and P = 1.17. The angle is calculated in θ C tified. To that end, the difference of consecutive segmen- degrees. In order to capture desired vessels and avoid ted frames is calculated pixel by pixel. If the summation segmenting larger structures like venules, P =13 corre- of difference for twenty segmented frames is higher than sponds to a vessel diameter of about 20 μm at the reso- a threshold value, the pixel is assigned as an active blood lution captured by the test camera. An original frame vessel. from sublingual microcirculatory video of a healthy sub- FCD is currently one of the main parameters used to ject is displayed in Figure 4. Active capillaries segmented evaluate the microcirculation. FCD can be calculated using MCA are highlighted in Figure 5. using two different approaches: one is completely man- FCD parameters are calculated from SDF videos for ual by gridding the frame and counting the number of nine subjects in both baseline (PPV = 1) and hemorrhage vessels crossing the grid lines; the second approach cal- (PPV < 1) conditions. As expected, a significant decrease culates the ratio of perfused vessels to the total surface in FCD is noticed during hemorrhage recordings with re- using a software tool [31]. FCD is calculated automatic- spect to baseline videos. For MCA, a paired-samples t-test ally in this study by dividing the area of active vessels to was conducted to compare FCD values of healthy baseline the total area of interest. It is much easier to form the and hemorrhage video recordings. There was a significant skeleton of the network of active capillaries and calculate difference in the scores for healthy baseline (μ =12.68, σ = -5 -6 the length of this skeleton to form the density measure. 1.479, area-based; μ =3.26 × 10 , σ =5.93 × 10 ,length- However, since the width/thickness of capillaries along based) and hemorrhage (μ =7.35, σ = 2.139, area-based; -5 -6 this network would be inconsistent (on the actual sub- μ =1.99 × 10 , σ =5.53 × 10 , length-based) conditions; lingual surface, the captured video, and in the processed with t =6.50 and p-value = .000189. These results sug- image), the density measure calculated on this length gest that the proposed algorithm, MCA, can successfully would be the least reliable measure, as it does not in- derive quantitative information from microcirculation vid- corporate the changes in the thickness of the capillary eos. Specifically, the results suggest that microcirculatory and therefore the true extent of circulation inside the ca- alterations caused by hemorrhage can be identified by pillary. The area-based density measure, on the other analyzing sublingual microcirculatory video recordings. Demir et al. BMC Medical Imaging 2012, 12:37 Page 7 of 13 http://www.biomedcentral.com/1471-2342/12/37 agreement). The Kappa coefficient is a statistical measure of inter-rater agreement. The FCD results of this analysis are provided in Table 2. A significant difference is also noticed in these scores for healthy baseline conditions (μ = 12.26, -5 -6 σ = 1.759, area-based; μ =1.65 ×10 , σ =5.485 ×10 , length-based) versus hemorrhage conditions (μ = 8.56, -5 -6 σ = 1.432, area-based; μ =1.20 ×10 , σ =4.124 ×10 , length-based); with t =4.19and p-value = 0.003. It should be pointed out that even though AVA allows manual interaction, the version we had access to does not allow the user to draw the capillaries to be included. If the user sees that a capillary has not been identified, the user must define an area that the vessel resides in Figure 4 An example frame from a healthy subject. An example and then have the program outline the vessel. However, frame from a sublingual microcirculatory video captured from a in some instances, the program will still not outline the healthy baseline subject is presented. vessel for inclusion in the FCD calculations. The user cannot manually force the outlining of vessels. FCD percentages for nine subjects for healthy baseline To clearly understand the results generated by the and hemorrhage conditions are shown in Table 1. proposed experimental algorithm (MCA) and heavily edited The same videos were analyzed using the currently AVA, a chart is generated showing the FCD values for available semi-automated tool, AVA, with manual edit- healthybaselineand hemorrhage conditions. Figure 7 shows ing. Analysis was performed using the method described the results from the heavily edited AVA software. The FCD in the manufacture’s tutorial of the product. All videos values from subjects in the healthy baseline condition were analyzed in an identical format. The microcircula- (PPV = 1) are labeled as baseline. Results of MCA are dis- tion videos are first automatically analyzed using the played in Figure 8. The decrease in FCD values for software tools and then manual interaction with the soft- hemorrhage (PPV < 1) is clearly visible for each subject for ware for manipulation of the segmentation results must the FCD values calculated by the experimental MCA auto- be performed to remove the large vessels, false positives, mated algorithm. Figures 9 and 10 provide an example of and vessels without flow. Figure 6 provides and example the overlay between MCA and the edited semi-automated of the microcirculation after hemorrhage and can be AVA method. However, comparing the compilation of data compared to the healthy (baseline) microcirculation in shown in Figures 7 and 8 and Tables 1 and 2, we conclude Figure 4. that MCA provides a better separation of FCD values be- Videos were analyzed by two experts previously trained tween healthy baseline and hemorrhage. Furthermore, a with AVA and the results for each video averaged in the Bland-Altman comparison of heavily edited AVA vs. fully- analysis of FCD. These individuals were blinded to FCD automated MCA (Figure 11) shows that MCA is capable of values derived from the fully automated method. The ana- producing results in line with those achieved from edited lysis resulted in a Kappa coefficient of 0.9 (very good AVA. Discussion This study presents an entirely automated method, MCA, to derive quantitative information from microcirculatory videos in near real-time. Currently available techniques for analysis of the microcirculation, in their present state, do not appear to be practical in the clinical setting due to the need for significant manual interaction with the software in order to process the image and determine FCD. The signifi- cant difference in calculated FCD values across healthy baseline and hemorrhage shows that MCA has the poten- tial for analyzing microcirculation videos in the clinical set- tings. Although the sample size of this study is relatively small, the algorithm demonstrates promise in its ability to Figure 5 Segmented active capillaries of the frame in Figure 4. rapidly provide quantitative information. Future studies The result of the proposed algorithm highlights all active capillaries will test the MCA algorithm on larger datasets—including from the original frame in Figure 4. human microcirculation videos—and improved accordingly. Demir et al. BMC Medical Imaging 2012, 12:37 Page 8 of 13 http://www.biomedcentral.com/1471-2342/12/37 Table 1 Calculated FCD values using the automated MCA algorithm, including both area and length based results Baseline (PPV = 1) Hemorrhagic (PPV < 1) Difference FCD (Area) % FCD (Length) FCD (Area) % FCD (Length) FCD (Area) % FCD (Length) 2 2 2 mm/mm mm/mm mm/mm -5 -5 -6 Subject 1 12.16 2.35 × 10 9.72 1.42 × 10 2.44 9.28 × 10 -5 -5 -5 Subject 2 14.99 3.35 × 10 7.32 1.48 × 10 7.67 1.87 × 10 -5 -5 -6 Subject 3 13.97 3.10 × 10 11.38 3.26 × 10 2.59 -1.54 × 10 -5 -5 -6 Subject 4 11.95 2.56 × 10 5.92 1.97 × 10 6.03 5.91 × 10 -5 -5 -5 Subject 5 10.21 3.10 × 10 7.58 2.02 × 10 2.63 1.08 × 10 -5 -5 -5 Subject 6 13.81 3.75 × 10 4.41 2.12 × 10 9.40 1.63 × 10 -5 -5 -5 Subject 7 11.90 4.01 × 10 7.38 1.65 × 10 4.52 2.36 × 10 -5 -5 -5 Subject 8 13.49 4.04 × 10 7.05 1.77 × 10 6.44 2.28 × 10 -5 -5 -6 Subject 9 11.65 3.11 × 10 5.37 2.26 × 10 6.28 8.52 × 10 -5 -5 -5 Mean 12.68 3.26 × 10 7.35 1.99 × 10 5.33 1.27 × 10 To overcome the variance among different video recordings, demonstrates very close performance between the auto- machine learning techniques, in particular neural networks mated MCA algorithm and that of the heavily edited AVA and support vector machines that show superior perform- method. FCD values during hemorrhage generated by ance in detection of elongated vessel-like objects in both bio- MCA were consistently lower than FCD values from the medical image processing applications [32,33] as well as healthy baseline state (Figure 8). Visual inspection of the other image processing applications [34], will be applied and videos confirmed the ability of MCA to identify more ves- algorithm parameters adjusted accordingly. The use of ma- sels without flow and thus not include them in the deter- chine learning techniques provide for a means to compen- mination of FCD. A potential limitation to this finding is sate for variations due to differences in factors such as new version of AVA that reportedly allows users to add lighting, pressure, video quality and specific machine/camera missed vessels by manual drawing. We did not have access used for imaging. Notwithstanding the small sample size, the to this version. While use of this newer version may have results show promise for an automated system that derives reduced the differences between semi-automated AVA diagnostically important information from microcirculation and the fully automated MCA, this improved version of videos. AVA still requires editing and interaction of the user with MCA and semi-automated AVA both show a signifi- the software. cant decrease in FCD values for the hemorrhagic sub- While the approach taken with MCA cannot be con- jects. Even though FCD values calculated using semi- sidered actual real-time, the 20 second wait for results automated AVA are statistically different for the healthy versus the 20-40 minutes of manual interaction required and hemorrhagic cases (μ=3.78, σ=2.267, t8=5.0, p-value = with semi-automated AVA makes the use of MCA near 1.1 × 10−3), the difference is not consistent (Figure 7). As real-time and may thus be appropriate in the future for noted in Figure 9, the overlay of analyzed results bedside point-of-care decision making. Additionally, the use of MCA would negate the considerable training that must be provided to the AVA user in order for them to be able to properly identify the active vessels in the results. This is in contrast to the proposed automated system, which incorporates that knowledge in the algorithm. In the future, the automated method could be easily integrated into existing SDF or OPSI hardware systems which would allow real-time bedside determinations of FCD and potentially other microcirculatory parameters such as flow quantification. It is likely that in order to use SDF or OPSI derived FCD measures to affect care, an automated and reproducible software approach to analysis will be required for regulatory approval of such an approach. Figure 6 Example of hemorrhage subject video source. An Recent automated capillary detection methods such as original frame from sublingual microcirculatory video of a Bezemer et al. [35] demonstrated impressive speed in its hemorrhage subject. analysis. The authors of this method, however, indicate that Demir et al. BMC Medical Imaging 2012, 12:37 Page 9 of 13 http://www.biomedcentral.com/1471-2342/12/37 Table 2 Calculated FCD values using semi-automated software, AVA Baseline (PPV = 1) Hemorrhagic (PPV < 1) Difference FCD (Area) % FCD (Length) FCD (Area) % FCD (Length) FCD (Area) % FCD (Length) 2 2 2 mm/mm mm/mm mm/mm -5 -6 -6 Subject 1 13.77 1.63 × 10 9.72 9.52 × 10 4.05 6.73 × 10 -5 -5 -6 Subject 2 14.50 2.15 × 10 7.32 1.47 × 10 7.18 6.80 × 10 -5 -5 -7 Subject 3 10.81 1.72 × 10 11.39 1.62 × 10 -0.58 9.75 × 10 -5 -5 -6 Subject 4 11.95 2.41 × 10 6.68 1.91 × 10 5.27 5.00 × 10 -5 -6 -5 Subject 5 9.44 2.18 × 10 9.21 9.65 × 10 0.23 1.22 × 10 -5 -6 -6 Subject 6 13.38 1.00 × 10 7.47 8.29 × 10 5.91 1.72 × 10 -5 -5 -6 Subject 7 10.15 1.75 × 10 8.36 1.43 × 10 1.79 3.15 × 10 -5 -6 -6 Subject 8 13.23 1.23 × 10 8.06 8.22 × 10 5.17 4.08 × 10 -6 -6 -7 Subject 9 13.07 8.18 × 10 8.82 8.04 × 10 4.25 1.35 × 10 -5 -5 -6 Mean 12.26 1.65 × 10 8.56 1.20 × 10 3.70 4.53 × 10 performance is limited by high cell densities and velocities, another (just as in AVA). Again, these issues form the basis which severely impede the applicability of this method in for the MCA approach as a means of reaching true automa- real SDF images. We believe this is due to the fact that many tion. While we used a binary assessment of flow (flow or factors and thresholds in this method were set to fixed num- no-flow) in identifying functional capillaries, improvements bers and that they require adjustment from one video to in assessing flow beyond this simple method may be helpful. Figure 7 FCD results calculated from heavily edited AVA. The FCD values calculated from heavily edited AVA for both healthy baseline and hemorrhage conditions are displayed. The healthy condition FCD values are labeled as baseline. The change in FCD values during hemorrhage is not consistent. a: FCD results (area based) from heavily edited AVA show inconsistent separation between the healthy (baseline, PPV = 1) and hemorrhagic (PPV < 1) cases. b: FCD results (length based) from heavily edited AVA. Demir et al. BMC Medical Imaging 2012, 12:37 Page 10 of 13 http://www.biomedcentral.com/1471-2342/12/37 Figure 8 FCD results calculated using the proposed algorithm (MCA). The FCD values calculated using the proposed algorithm for both healthy baseline and hemorrhage conditions are displayed. The healthy condition FCD values are labeled as baseline. The decrease in FCD values for each subject during hemorrhagic is obvious in the provided figure. a: FCD results (area based) from the proposed automated system show better and consistent separation between healthy (baseline, PPV = 1) and hemorrhagic (PPV < 1) cases. b: FCD results (length based) from the proposed automated system show good separation, but demonstrate the problem with length based FCD calculation where vessel width is not taken into consideration as it is with area based FCD. Figure 9 Overlay of proposed automated method onto heavily edited AVA results showing a high degree of similarity. Results from proposed automated method (green) superimposed over results from heavily edited AVA (red/black). Proposed method Figure 10 Frame of video used for analysis in Figure 9. An returns results 60 to 120 times faster than manual editing in AVA example frame of the video used to generate Figure 9 in both MCA (20 seconds vs. 20-40 minutes). and heavily edited AVA. Demir et al. BMC Medical Imaging 2012, 12:37 Page 11 of 13 http://www.biomedcentral.com/1471-2342/12/37 Bland-Altman plot 1.96 s 1.96 s mean mean -1 -2 -1.96 s -1.96 s -3 -4 0 2 4 6 8 10 12 14 Average by two assays Bland-Altman plot 1.96 s mean -1 -2 -1.96 s -3 0 2 4 6 8 10 Average by two assays Figure 11 Bland-Altman plots showing validity of fully-automated MCA as a measurement tool vs. current “gold standard” of heavily edited AVA (baseline (a) and hemorrhagic (b). a: Bland-Altman plot showing correlation between heavily edited AVA and fully-automated MCA for baseline (PPV = 1) subjects. b: Bland-Altman plot showing correlation between heavily edited AVA and fully-automated MCA for hemorrhagic (PPV < 1) subjects. This might result in improved accuracy and compensation is capable of detecting significant changes in FCD pro- for some of the mechanical shortcomings of image acquisi- duced by hemorrhage and are comparable to a heavily tion using the camera technology, which, is capable of pro- manually edited commercially available software product. ducing pressure related flow artifacts. Future work will focus on adjusting the algorithm para- meters on larger datasets and improving accuracy as well Conclusions as developing improved methods of quantifying blood A suggested algorithm to analyze microcirculation video flow. It is hoped that these expanded methods and ana- recordings based on advanced machine learning is pro- lyses will lead to the ability to derive diagnostically import- posed which is capable of identifying active capillaries and ant decisions from the microcirculatory video recordings calculating FCD parameters automatically. The approach as well as to guide therapeutic interventions. Difference (Heavily Edited AVA - MCA) Difference (Heavily Edited AVA - MCA) Demir et al. BMC Medical Imaging 2012, 12:37 Page 12 of 13 http://www.biomedcentral.com/1471-2342/12/37 Competing interests References Dr. Demir has intellectual property including pending patents on the 1. Cĕrný V, Turek Z, Pařízková R: Orthogonal polarization spectral imaging: a technology discussed in this manuscript through Virginia Commonwealth review. Physiol Res 2007, 56:141–147. University. She also has served as an employee of Signal Processing 2. Bateman R, Sharpe M, Ellis C: Bench-to-bedside review: microvascular Technologies, LLC. dysfunction in sepsis: hemodynamics, oxygen transport and nitric oxide. Dr. Hakimzadeh has intellectual property including pending patents on the Crit Care Med 2003, 7:359–373. technology discussed in this manuscript. She also is a co-owner/share-holder 3. Hebbel R, Osarogiagbon R, Kaul D: The endothelial biology of sickle cell of Signal Processing Technologies, LLC which received an NSF SBIR grant for disease; inflammation and chronic vasculopathy. Microcirculation 2004, this research. She also serves as Chief Executive Officer at the company. 11:129–151. Dr. Hargraves has intellectual property including pending patents on the 4. Stuart M, Nagel R: Sickle cell disease. Lancet 2004, 364:1343–1360. technology discussed in this manuscript. She also is a co-owner/share-holder 5. Levy B, Ambrosio G, Pries A, Struijker-Boudier H: Microcirculation in of Signal Processing Technologies, LLC and serves as Chief Technology hypertension: a new target for treatment? Circulation 2001, 104:735–740. Officer at the company. 6. Verdant C, Backer DD: Monitoring of the microcirculation may help us at Dr. Ward has intellectual property including pending patents on the the bedside. Curr OpinCrit Care 2005, 11(3):240–244. technology discussed in this manuscript through Virginia Commonwealth 7. Nath K, Katusic Z, Gladwin M: The perfusion paradox and instability in University. He also has serves as an uncompensated consulted to Signal sickle cell disease. Microcirculation 2004, 11:179–193. Processing Technologies (SPT), LLC. Dr. Ward used to serve as Chief Medical 8. Kaul D, Farby M: In vivo studies of sickle red blood cells. Microcirculation Officer for SPT with ownership interest but currently he neither has any 2004, 11:153–156. ownership in SPT nor serves as an officer for this company. 9. Touyz R: Intracellular mechanisms involved in vascular remodeling of Mr. Myer has intellectual property including pending patents on the resistance arteries in hypertension: role of angiotensin II. Exp Physiol 2005, technology discussed in this manuscript through Virginia Commonwealth 90:449–455. University. He has also served as an employee of Signal Processing 10. Serne E, Gans R, ter Maaten J, Tangelder G, Donker A, Stehouwer C: Technologies, LLC. Impaired skin capillary recruitment in essential hypertension is caused Dr. Najarian has intellectual property including pending patents on the by both functional and structural capillary rarefaction. Hypertension 2001, technology discussed in this manuscript through Virginia Commonwealth 38:238–242. University. His wife is a co-owner/share-holder of Signal Processing 11. Creager M, Luscher T, Cosentino F, Bechkman J: Diabetes and vascular Technologies, LLC, which received an NSF SBIR grant for this research. Dr. disease: pathophysiology, clinical consequences, and medical therapy: Najarian has also served as a technology advisor for Signal Processing part I. Circulation 2003, 108:1527–1532. Technology LLC. 12. Atasever B, Boer C, van der Kuil M: Quantitative imaging of microcirculatory response during nitroglycerin-induced hypotension. J Cardiothorac Vasc Anesth 2011, 25(1):140–144. Authors’ contributions 13. Balestra G, Bezemer R, Boerma E, Yong Z, Sjauw K, Engstrom A, Koopmans Dr. SUD participated in the design of the image reprocessing algorithms, M, Ince C: Improvement of sidestream dark field imaging with an image implemented the algorithms and processed the videos using the proposed acquisition stabilizer. BMC Med Imaging 2010, 10:15. algorithm. Dr.RH supervised the coordination of the entire study and the 14. Ince C: Sidestream dark field imaging: an improved technique to design and testing of the software. Dr. RHH co-supervised the design, observe sublingual microcirculation. Crit Care 2005, 9:72. implementation, testing and validation of the vessel detection algorithms 15. Groner W, Winkelman JW, Harris AG, Ince C, Bouma GJ, Messmer K, Nadeau and identification of pixels in vessels based on angle, intensity and width. Dr. RG: Orthogonal polarization spectral imaging: a new method for study of KRW formed the guidelines for data collection, provided the data, outlined the microcirculation. Natmed 1999, 5:1209–1212. the clinical objectives of the study, and assessed the results from 16. Goedhart PT, Khalilzada M, Bezemer R, Merza J, Ince C: Sidestream dark physiological and clinical standpoint and oversaw the general development field (SDF) imaging: a novel stroboscopic LED ring-based imaging and refinement of the image processing algorithm. Mr. EVM designed the modality for clinical assessment of the microcirculation. Opt Express 2007, improvements to the per-processing steps, participated in conducting the 15:15101–15114. comparison between the proposed method and other methods, and carries 17. Awan ZA, Wester T, Kvernebo K: Human microvascular imaging: a review out the statistical analysis. Dr. KN supervised the design of the entire image of skin and tongue videomicroscopy techniques and analysing variables. processing algorithm, in particular stabilization and region growing. All Clin Physiol Funct Imaging 2010, 30:79–88. authors’ read and approved the final manuscript. 18. Klyscz T, Junger M, Jung F, Zeintl H: Cap image: a new kind of computer- assisted video image analysis system for dynamic capillary microscopy. Biomed Tech (Berl) 1997, 42:168–175. Acknowledgments 19. Genzel-Boroviczeny O, Strotgen J, Harris AG, Messmer K, Christ F: This material is based upon work supported by the National Science Orthogonal polarization spectral imaging (OPS): a novel method to Foundation under Grant No. 1046521. All authors equally contributed to this measure the microcirculation in term and preterm infants work. The dataset used in this research was provided by Virginia transcutaneously. Pediatr Res 2002, 51:386–391. Commonwealth University, Department of Emergency Medicine. In addition, 20. Bauer A, Kofler S, Thiel M, Elfert S, Christ F: Monitoring of the sublingual the authors would like to thank Dr. Hakam Tiba and Gerard Draucker from microcirculation in cardiac surgery using orthogonal polarization spectral the Virginia Commonwealth University Reanimation Engineering Science imaging. The American Society of Anesthesiologists 2007, 107(6):939–945. Center (VCURES) for their participation in sample preparation, data collection 21. Schaudig S, Dadasch B, Kellam K, Christ F: Validation of an analysis and verification of the results. software for OPS-imaging used in humans.In Proceedings of the 7th World Congress for Microcirculation. Sydney: Australian and New Zealand Author details 1 2 Microcirculation Society; 2001:2–59. Signal Processing Technologies LLC, Richmond, VA, USA. Department of 22. Zhong J, Asker C, Salerud E: Imaging, image processing and pattern Electrical Engineering, Virginia Commonwealth University, Richmond, VA, analysis of skin capillary ensembles. Skin Res Technol 2000, 6:45–57. USA. Department of Emergency Medicine, Virginia Commonwealth 23. Sainthillier J, Gharbi T, Muret P, Humbert P: Skin capillary network recognition University, Richmond, VA, USA. Virginia Commonwealth University and analysis by means of neural algorithms. Skin Res Technol 2005, 11:9–16. Reanimation Engineering Science (VCURES), Richmond, VA, USA. 24. Dobbe J, Streekstra G, Atasever B, van Zijderveld R, Ince C: The Department of Computer Science, Virginia Commonwealth University, measurement of functional microcirculatory density and velocity Richmond, VA, USA. Department of Emergency Medicine, Michigan Center distributions using automated image analysis. Med BiolEngComput 2008, for Integrative Research in Critical Care, University of Michigan, Ann Arbor, 46(7):659–670. MI, USA. 25. Nolte D, Zeintl H, Steinbauer M, Pickelmann S, Messmer K: Functional Received: 16 November 2011 Accepted: 17 December 2012 capillary density: an indicator of tissue perfusion? Int J MicrocircClinExp Published: 21 December 2012 1995, 15:244–249. Demir et al. BMC Medical Imaging 2012, 12:37 Page 13 of 13 http://www.biomedcentral.com/1471-2342/12/37 26. Zitov_a B, Flusser J: Image registration methods: a survey. Image and Vision Computing 2003, 21(11):977–1000. 27. Demir S, Mirshahi N, Tiba M, Draucker G, Ward K, Hobson R, Najarian K: Image processing and machine learning for diagnostic analysis of microcirculation.In ICME International Conference on Complex Medical Engineering (CME 2009): 9-11 April 2009; Tempe: IEEE; 2009:1–5. 28. Zuiderveld K: Contrast limited adaptive histogram equalization.In Graphics Gems IV. Edited by Heckbert PS. San Francisco: Morgan Kaufmann; 1994:474–485. 29. Najarian K, Splinter R: Biomedical Signal and Image Processing. Boca Raton: The CRC Press; 2005. 30. Jiang X, Mojon D: Adaptive local thresholding by verification-based multithreshold probing with application to vessel detection in retinal images. IEEE Trans Pattern Anal Mach Intell 2003, 25:131–137. 31. Backer DD, Hollenberg S, Boerma C, Goedhart P, Büchele G, Ospina-Tascon G, Dobbe I, Ince C: How to evaluate the microcirculation: report of a round table conference. Crit Care 2007, 11(5):R101. 32. Chen W, Smith R, Ji SY, Ward K, Najarian K: Automated ventricular systems segmentation in brain CT images by combining low-level segmentation and high level template matching. BMC Med Inform Decis Mak 2009, 9(Suppl 1):S4. 33. Ji SY, Smith R, Huynh T, Najarian K: A comparative analysis of multi-level computer-assisted decision making systems for traumatic brain injuries. BMC Med Inform Decis Mak 2009, 9:2. 34. Muralikrishnan B, Najarian K, Raja J: Surface wavelength based clustering using neural networks for manufacturing process mapping. Int J Mach Tool Manuf 2003, 43(4):369–477. 35. Bezemer R, Dobbe J, Bartels S, Boerma C, Elbers P, Heger M, Ince C: Rapid automatic assessment of microvascular density in sidestream dark field images. Med Biol Eng Comput 2011, 49:1269–1278. doi:10.1186/1471-2342-12-37 Cite this article as: Demir et al.: An automated method for analysis of microcirculation videos for accurate assessment of tissue perfusion. BMC Medical Imaging 2012 12:37. Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color figure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit
BMC Medical Imaging – Springer Journals
Published: Dec 21, 2012
You can share this free article with as many people as you like with the url below! We hope you enjoy this feature!
Read and print from thousands of top scholarly journals.
Already have an account? Log in
Bookmark this article. You can see your Bookmarks on your DeepDyve Library.
To save an article, log in first, or sign up for a DeepDyve account if you don’t already have one.
Copy and paste the desired citation format or use the link below to download a file formatted for EndNote
Access the full text.
Sign up today, get DeepDyve free for 14 days.
All DeepDyve websites use cookies to improve your online experience. They were placed on your computer when you launched this website. You can change your cookie settings through your browser.