Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Intelligible Models for HealthCare: Predicting the Probability of 6-Month Unfavorable Outcome in Patients with Ischemic Stroke

Intelligible Models for HealthCare: Predicting the Probability of 6-Month Unfavorable Outcome in... Early prediction of unfavorable outcome after ischemic stroke is significant for clinical management. Machine learning as a novel computational modeling technique could help clinicians to address the challenge. We aim to investigate the applicability of machine learning models for individualized prediction in ischemic stroke patients and demonstrate the utility of various model-agnostic explanation techniques for machine learning predictions. A total of 499 consecutive patients with Unfavorable [modified Rankin Scale (mRS) score 3–6, n = 140] and favorable (mRS score 0–2, n = 359) outcome after 6-month from ischemic stroke were enrolled in this study. Four machine learning models, including Random Forest [RF], eXtreme Gradient Boosting [XGBoost], Adaptive Boosting [Adaboost] and Support Vector Machine [SVM] were performed with the area-under-the-curve (AUC): (90.20 ± 0.22)%, (86.91 ± 1.05)%, (86.49 ± 2.35)%, (81.89 ± 2.40)%, respectively. Three global interpretability techniques (Feature Importance shows the contribution of selected features, Partial Dependence Plot aims to visualize the average effect of a feature on the predicted probability of unfavorable outcome, Feature Interaction detects the change in the prediction that occurs by varying the features after considering the individual feature effects) and one local interpretability technique (Shapley Value indicates the probability of unfavorable outcome of different instances) have been applied to present the interpretability techniques via visualization. Thereby, the current study is important for better understanding intelligible healthcare analytics via explanations for the prediction of local and global levels, and potentially reduction of the mortality of patients with ischemic stroke by assisting clinicians in the decision-making process. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Neuroinformatics Springer Journals

Intelligible Models for HealthCare: Predicting the Probability of 6-Month Unfavorable Outcome in Patients with Ischemic Stroke

Download PDF
11 pages

Loading next page...
 
/lp/springer-journals/intelligible-models-for-healthcare-predicting-the-probability-of-6-weBk0Pa0Kb
Publisher
Springer Journals
Copyright
Copyright © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2021
ISSN
1539-2791
eISSN
1559-0089
DOI
10.1007/s12021-021-09535-6
Publisher site
See Article on Publisher Site

Abstract

Early prediction of unfavorable outcome after ischemic stroke is significant for clinical management. Machine learning as a novel computational modeling technique could help clinicians to address the challenge. We aim to investigate the applicability of machine learning models for individualized prediction in ischemic stroke patients and demonstrate the utility of various model-agnostic explanation techniques for machine learning predictions. A total of 499 consecutive patients with Unfavorable [modified Rankin Scale (mRS) score 3–6, n = 140] and favorable (mRS score 0–2, n = 359) outcome after 6-month from ischemic stroke were enrolled in this study. Four machine learning models, including Random Forest [RF], eXtreme Gradient Boosting [XGBoost], Adaptive Boosting [Adaboost] and Support Vector Machine [SVM] were performed with the area-under-the-curve (AUC): (90.20 ± 0.22)%, (86.91 ± 1.05)%, (86.49 ± 2.35)%, (81.89 ± 2.40)%, respectively. Three global interpretability techniques (Feature Importance shows the contribution of selected features, Partial Dependence Plot aims to visualize the average effect of a feature on the predicted probability of unfavorable outcome, Feature Interaction detects the change in the prediction that occurs by varying the features after considering the individual feature effects) and one local interpretability technique (Shapley Value indicates the probability of unfavorable outcome of different instances) have been applied to present the interpretability techniques via visualization. Thereby, the current study is important for better understanding intelligible healthcare analytics via explanations for the prediction of local and global levels, and potentially reduction of the mortality of patients with ischemic stroke by assisting clinicians in the decision-making process.

Journal

Neuroinformatics – Springer Journals

Published: Jul 1, 2022

Keywords: Ischemic stroke; Unfavorable outcome; Machine learning; Interpretability; Visualization

References

  • Precision diagnosis: A view of the clinical decision support systems (CDSS) landscape through the lens of critical care
    Belard, A; Buchman, T; Forsberg, J; Potter, BK; Dente, CJ; Kirk, A; Elster, E
  • Selection of relevant features and examples in machine learning
    Blum, AL; Langley, P
  • Random forests
    Breiman, L
  • Introduction of direct oral anticoagulant within 7 days of stroke onset: A nomogram to predict the probability of 3-month modified Rankin scale score > 2
    Cappellari, M; Turcato, G; Forlivesi, S; Micheletti, N; Tomelleri, G; Bonetti, B; Merlino, G; Eleopra, R; Russo, M; L’Erario, R; Adami, A; Gentile, C; Gaudenzi, A; Bruno, S; Bovi, P
  • SMOTE: Synthetic minority over-sampling technique
    Chawla, NV; Bowyer, KW; Hall, LO; Kegelmeyer, WP
  • External validation of the ASTRAL and DRAGON scores for prediction of functional outcome in stroke
    Cooray, C; Mazya, M; Bottai, M; Dorado, L; Skoda, O; Toni, D
  • Cerebral small vessel disease: A review focusing on pathophysiology, biomarkers, and machine learning strategies
    Cuadrado-Godia, E; Dwivedi, P; Sharma, S; Ois Santiago, A; Roquer Gonzalez, J; Balcells, M; Laird, J; Turk, M; Suri, HS; Nicolaides, A; Saba, L; Khanna, NN; Suri, JS
  • Effect of treatment delay, age, and stroke severity on the effects of intravenous thrombolysis with alteplase for acute ischaemic stroke: A meta-analysis of individual patient data from randomised trials
    Emberson, J; Lees, KR; Lyden, P; Blackwell, L; Albers, G; Bluhmki, E; Brott, T; Cohen, G; Davis, S; Donnan, G; Grotta, J; Howard, G; Kaste, M; Koga, M; von Kummer, R; Lansberg, M; Lindley, RI; Murray, G; Olivot, JM; Parsons, M; Tilley, B; Toni, D; Toyoda, K; Wahlgren, N; Wardlaw, J; Whiteley, W; del Zoppo, G; Baigent, C; Sandercock, P; Hacke, W
  • Update on the global burden of ischemic and hemorrhagic stroke in 1990-2013: The GBD 2013 study
    Feigin, VL; Krishnamurthi, RV; Parmar, P; Norrving, B; Mensah, GA; Bennett, DA; Barker-Collo, S; Moran, AE; Sacco, RL; Truelsen, T; Davis, S; Pandian, JD; Naghavi, M; Forouzanfar, MH; Nguyen, G; Johnson, CO; Vos, T; Meretoja, A; Murray, CJ; Roth, GA
  • THRIVE score predicts ischemic stroke outcomes and thrombolytic hemorrhage risk in VISTA
    Flint, AC; Faigeles, BS; Cullen, SP; Kamel, H; Rao, VA; Gupta, R; Smith, WS; Bath, PM; Donnan, GA; Lees, KR; Alexandrov, A; Bath, PM; Bluhmki, E; Bornstein, N; Claesson, L; Davis, SM; Donnan, G; Diener, HC; Fisher, M; Gregson, B; Grotta, J; Hacke, W; Hennerici, MG; Hommel, M; Kaste, M; Lyden, P; Marler, J; Muir, K; Sacco, R; Shuaib, A; Teal, P; Wahlgren, NG; Warach, S; Weimar, C
  • Greedy function approximation: A gradient boosting machine
    Friedman, J
  • Predictive learning via rule ensembles
    Friedman, JH; Popescu, BE
  • A comparison of models for predicting early hospital readmissions
    Futoma, J; Morris, J; Lucas, J
  • Genetics of ischemic stroke
    Gschwendtner, A; Dichgans, M
  • Mining high-dimensional administrative claims data to predict early hospital readmissions
    He, D; Mathews, SC; Kalloo, AN; Hutfless, S
  • Support vector machines
    Hearst, MA; Dumais, ST; Osman, E; Platt, J; Scholkopf, B
  • Machine learning-based model for prediction of outcomes in acute stroke
    Heo, J; Yoon, JG; Park, H; Kim, YD; Nam, HS; Heo, JH
  • A structural approach to selection bias
    Hernán, MA; Hernández-Díaz, S; Robins, JM
  • Limitation of inverse probability-of-censoring weights in estimating survival in the presence of strong selection bias
    Howe, CJ; Cole, SR; Chmiel, JS; Muñoz, A
  • Age dependency of safety and outcome of endovascular therapy for acute stroke
    Luedi, R; Hsieh, K; Slezak, A; El-Koussy, M; Fischer, U; Heldner, MR
  • A comparison of random forest and its Gini importance with standard chemometric methods for the feature selection and classification of spectral data
    Menze, BH; Kelm, BM; Masuch, R; Himmelreich, U; Bachert, P; Petrich, W; Hamprecht, FA
  • Telehealth stroke dysphagia evaluation is safe and effective
    Morrell, K; Hyers, M; Stuchiner, T; Lucas, L; Schwartz, K; Mako, J; Spinelli, KJ; Yanase, L
  • An integer-based score to predict functional outcome in acute ischemic stroke: The ASTRAL score
    Ntaios, G; Faouzi, M; Ferrari, J; Lang, W; Vemmos, K; Michel, P
  • Current depressive symptoms but not history of depression predict hospital readmission or death after discharge from medical wards: A multisite prospective cohort study
    Pederson, JL; Majumdar, SR; Forhan, M; Johnson, JA; McAlister, FA
  • Evaluation: From precision, recall and fmeasure to roc, informedness, markedness and correlation
    Powers, D
  • Design and methodology of a pilot randomized controlled trial of transcranial direct current stimulation in acute middle cerebral artery stroke (STICA)
    Pruvost-Robieux, E; Calvet, D; Ben Hassen, W; Turc, G; Marchi, A; Mélé, N; Seners, P; Oppenheim, C; Baron, JC; Mas, JL; Gavaret, M
  • Machine learning and big data in psychiatry: Toward clinical applications
    Rutledge, RB; Chekroud, AM; Huys, QJ
  • An automated model using electronic medical record data identifies patients with cirrhosis at high risk for readmission
    Singal, AG; Rahimi, RS; Clark, C; Ma, Y; Cuthbert, JA; Rockey, DC; Amarasingham, R
  • Explaining prediction models and individual predictions with feature contributions
    Štrumbelj, E; Kononenko, I
  • Explanation and reliability of prediction models: The case of breast cancer recurrence
    Štrumbelj, E; Bosnić, Z; Kononenko, I; Zakotnik, B; Kuhar, CG
  • A NADE nomogram to predict the probability of 6-month unfavorable outcome in Chinese patients with ischemic stroke
    Sun, C; Li, X; Song, B; Chen, X; Nyame, L; Liu, Y; Tang, D; Ibrahim, M; Zhao, Z; Liu, C; Yan, M; Pan, X; Yang, J; Zhou, J; Zou, J
  • Early function decline after ischemic stroke can be predicted by a nomogram based on age, use of thrombolysis, RDW and NIHSS score at admission
    Turcato, G; Cervellin, G; Cappellari, M; Bonora, A; Zannoni, M; Bovi, P; Ricci, G; Lippi, G
  • Predicting outcome of endovascular treatment for acute ischemic stroke: Potential value of machine learning algorithms
    van Os, HJA; Ramos, LA; Hilbert, A; van Leeuwen, M; van Walderveen, MAA; Kruyt, ND
  • Heart disease and stroke Statistics-2020 update: A report from the American Heart Association
    Virani, SS; Alonso, A; Benjamin, EJ; Bittencourt, MS; Callaway, CW; Carson, AP; Chamberlain, AM; Chang, AR; Cheng, S; Delling, FN; Djousse, L; Elkind, MSV; Ferguson, JF; Fornage, M; Khan, SS; Kissela, BM; Knutson, KL; Kwan, TW; Lackland, DT; Lewis, TT; Lichtman, JH; Longenecker, CT; Loop, MS; Lutsey, PL; Martin, SS; Matsushita, K; Moran, AE; Mussolino, ME; Perak, AM; Rosamond, WD; Roth, GA; Sampson, UKA; Satou, GM; Schroeder, EB; Shah, SH; Shay, CM; Spartano, NL; Stokes, A; Tirschwell, DL; VanWagner, L; Tsao, CW
  • Risk factors of outcomes in elderly patients with acute ischemic stroke in China
    Wu, Q; Zou, C; Wu, C; Zhang, S; Huang, Z
  • Relationship of obesity to adverse events among patients with mean 10-year history of type 2 diabetes mellitus: Results of the ACCORD study
    Xing, Z; Pei, J; Huang, J; Peng, X; Chen, P; Hu, X
  • Extreme gradient boosting model has a better performance in predicting the risk of 90-day readmissions in patients with Ischaemic stroke
    Xu, Y; Yang, X; Huang, H; Peng, C; Ge, Y; Wu, H; Wang, J; Xiong, G; Yi, Y
  • 1409.1556
    Simonyan, K., & Zisserman, A.
  • An automated model using electronic medical record data identifies patients with cirrhosis at high risk for readmission
    AG Singal, RS Rahimi, C Clark, Y Ma, JA Cuthbert, DC Rockey, R Amarasingham
  • Explaining prediction models and individual predictions with feature contributions
    E Štrumbelj, I Kononenko
  • Explanation and reliability of prediction models: The case of breast cancer recurrence
    E Štrumbelj, Z Bosnić, I Kononenko, B Zakotnik, CG Kuhar
  • A NADE nomogram to predict the probability of 6-month unfavorable outcome in Chinese patients with ischemic stroke
    C Sun, X Li, B Song, X Chen, L Nyame, Y Liu, D Tang, M Ibrahim, Z Zhao, C Liu, M Yan, X Pan, J Yang, J Zhou, J Zou
  • Early function decline after ischemic stroke can be predicted by a nomogram based on age, use of thrombolysis, RDW and NIHSS score at admission
    G Turcato, G Cervellin, M Cappellari, A Bonora, M Zannoni, P Bovi, G Ricci, G Lippi
  • Predicting outcome of endovascular treatment for acute ischemic stroke: Potential value of machine learning algorithms
    HJA Os, LA Ramos, A Hilbert, M Leeuwen, MAA Walderveen, ND Kruyt
  • Heart disease and stroke Statistics-2020 update: A report from the American Heart Association
    SS Virani, A Alonso, EJ Benjamin, MS Bittencourt, CW Callaway, AP Carson, AM Chamberlain, AR Chang, S Cheng, FN Delling, L Djousse, MSV Elkind, JF Ferguson, M Fornage, SS Khan, BM Kissela, KL Knutson, TW Kwan, DT Lackland, TT Lewis, JH Lichtman, CT Longenecker, MS Loop, PL Lutsey, SS Martin, K Matsushita, AE Moran, ME Mussolino, AM Perak, WD Rosamond, GA Roth, UKA Sampson, GM Satou, EB Schroeder, SH Shah, CM Shay, NL Spartano, A Stokes, DL Tirschwell, L VanWagner, CW Tsao
  • Risk factors of outcomes in elderly patients with acute ischemic stroke in China
    Q Wu, C Zou, C Wu, S Zhang, Z Huang
  • Relationship of obesity to adverse events among patients with mean 10-year history of type 2 diabetes mellitus: Results of the ACCORD study
    Z Xing, J Pei, J Huang, X Peng, P Chen, X Hu
  • Extreme gradient boosting model has a better performance in predicting the risk of 90-day readmissions in patients with Ischaemic stroke
    Y Xu, X Yang, H Huang, C Peng, Y Ge, H Wu, J Wang, G Xiong, Y Yi
  • 1506.06579
    Yosinski, J., Clune, J., Nguyen, A., Fuchs, T., & Lipson, H.