CoChat

Opening the black box of artificial intelligence for clinical decision support: A study predicting stroke outcome

Open access

Published version of a preprint

The DOI matches its Crossref record; 51 of the 51 references it cites were found in a registry or an index.

Abstract

State-of-the-art machine learning (ML) artificial intelligence methods are increasingly leveraged in clinical predictive modeling to provide clinical decision support systems to physicians. Modern ML approaches such as artificial neural networks (ANNs) and tree boosting often perform better than more traditional methods like logistic regression. On the other hand, these modern methods yield a limited understanding of the resulting predictions. However, in the medical domain, understanding of applied models is essential, in particular, when informing clinical decision support. Thus, in recent years, interpretability methods for modern ML methods have emerged to potentially allow explainable predictions paired with high performance. To our knowledge, we present in this work the first explainability comparison of two modern ML methods, tree boosting and multilayer perceptrons (MLPs), to traditional logistic regression methods using a stroke outcome prediction paradigm. Here, we used clinical features to predict a dichotomized 90 days post-stroke modified Rankin Scale (mRS) score. For interpretability, we evaluated clinical features' importance with regard to predictions using deep Taylor decomposition for MLP, Shapley values for tree boosting and model coefficients for logistic regression. With regard to performance as measured by Area under the Curve (AUC) values on the test dataset, all models performed comparably: Logistic regression AUCs were 0.83, 0.83, 0.81 for three different regularization schemes; tree boosting AUC was 0.81; MLP AUC was 0.83. Importantly, the interpretability analysis demonstrated consistent results across models by rating age and stroke severity consecutively amongst the most important predictive features. For less important features, some differences were observed between the methods. Our analysis suggests that modern machine learning methods can provide explainability which is compatible with domain knowledge interpretation and traditional method rankings. Future work should focus on replication of these findings in other datasets and further testing of different explainability methods.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1371/journal.pone.0231166

Cites

51 of 51 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig 1 Graphical representation of the model performance results. The graph illustrates the performance of the different models evaluated on the training (blue) and test (orange) sets: generalized linear model (GLM), Lasso, Elastic net, Tree Boosting and multilayer perceptron (MLP). The markers show the median AUC over 50 shuffles and the error bars represent interquartile range (IQR). All models showed a similar median AUC around 0.82. The largest difference in performance between training and test set, indicating potential overfitting, was observed for the Catboost model.
Fig 1 Graphical representation of the model performance results. The graph illustrates the performance of the different models evaluated on the training (blue) and test (orange) sets: generalized linear model (GLM), Lasso, Elastic net, Tree Boosting and multilayer perceptron (MLP). The markers show the median AUC over 50 shuffles and the error bars represent interquartile range (IQR). All models showed a similar median AUC around 0.82. The largest difference in performance between training and test set, indicating potential overfitting, was observed for the Catboost model.CC BY · PubMed Central
Fig 2 Graphical representation of the feature importance. The figure illustrates the features rating derived from the model-tailored interpretability methods for generalized linear model (GLM), Lasso, Elastic net, Catboost and multilayer perceptron (MLP). All models rated age and initial NIHSS consistently amongst the most important features. For less important features, results were more varied. For logistic regression techniques the results are given in weights, for Catboost in Shap(ley) values and for MLP in deep Taylor values that were normalized to the range [0,1]. The bar heights represent means and error bars represent standard deviation over samples (shuffles).
Fig 2 Graphical representation of the feature importance. The figure illustrates the features rating derived from the model-tailored interpretability methods for generalized linear model (GLM), Lasso, Elastic net, Catboost and multilayer perceptron (MLP). All models rated age and initial NIHSS consistently amongst the most important features. For less important features, results were more varied. For logistic regression techniques the results are given in weights, for Catboost in Shap(ley) values and for MLP in deep Taylor values that were normalized to the range [0,1]. The bar heights represent means and error bars represent standard deviation over samples (shuffles).CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 51 references

Research works printed but not linked: 1

  1. BC Csáji, 2001, Approximation with Artificial Neural Networks, 45

Cited by

See all 188

Metadata

Authors
Zihni, E., Madai, V. I., Livne, M., Galinovic, I., Khalil, A. A., Fiebach, J. B., & Frey, D.
Affiliations
  • Charité – Universitätsmedizin Berlin (DE)
  • Center for Stroke Research Berlin (DE)
Corresponding author
Madai, Vince Istvan
Venue
PLOS ONE
Published
2020-04-06
Volume
15
Issue
4
Pages
e0231166
ISSN
1932-6203 (ISSN-L)
OpenAlex
W3014245704
Cited by
187
References
50
Citations by year
2026: 20 · 2025: 26 · 2024: 34 · 2023: 31 · 2022: 30 · 2021: 35 · 2020: 10
FWCI
10.41
Type
journal article
Language
en
Open access
Locations
  • PLoS ONE (publishedVersion, cc-by)
  • medRxiv (acceptedVersion, cc-by-nc-nd)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • Arrow - TU Dublin (Technological University Dublin) (submittedVersion, cc-by-nc-sa)
  • RePEc: Research Papers in Economics (submittedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
  • Refubium (Universitätsbibliothek der Freien Universität Berlin) (submittedVersion, cc-by)
  • Universitätsbibliothek der FU Berlin Hochschulschriftenstelle u. Dokumentenserver (cc-by)
Funders
  • Deutsche Forschungsgemeinschaft
  • Bundesministerium für Bildung und Forschung (PREDICTioN2020)
Topics
  • Physical Sciences › Computer Science › Artificial Intelligence › Explainable Artificial Intelligence (XAI)
  • Health Sciences › Health Professions › Health Information Management › Artificial Intelligence in Healthcare
  • Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning in Healthcare
Keywords
explainable AI, clinical decision support, stroke outcome prediction, modified Rankin Scale, gradient boosting decision tree, multilayer perceptron, logistic regression, feature importance, Shapley value, stroke severity, age
MeSH terms
Clinical Decision-Making, Supervised Machine Learning, Aged, Aged, 80 and over, Female, Forecasting, Humans, Male, Middle Aged, Prognosis, Retrospective Studies, Logistic Models, Neural Networks, Computer, Outcome Assessment, Health Care, Area Under Curve, Stroke

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Opening the black box of artificial intelligence for clinical decision support: A study predicting stroke outcome. https://app.cochat.ai/papers/10.1371/journal.pone.0231166
The paper
Zihni, E., Madai, V. I., Livne, M., Galinovic, I., Khalil, A. A., Fiebach, J. B., & Frey, D. (2020). Opening the black box of artificial intelligence for clinical decision support: A study predicting stroke outcome. PLOS ONE. https://doi.org/10.1371/journal.pone.0231166

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.