EUVIMEDEUVIMED entdecken
ÜbersichtsarbeitOpen Access · CC BY-NC-ND 4.0

Practical guide to SHAP analysis: Explaining supervised machine learning model predictions in drug development

Ana Victoria Ponce-Bobadilla iD, Vanessa Schmitt iD, Corinna S Maier, Sven Mensing iD, Sven Stodtmann iD

Clinical and Translational Science · 2024-11 · 17 (11)

DOI: 10.1111/cts.70056 ↗

Kurz gefasst

Übersichtsarbeit in der Zeitschrift Clinical and Translational Science, online erschienen am 28.10.2024. Verfasst von Ana Victoria Ponce-Bobadilla und 4 weiteren Autor:innen. Beteiligte Einrichtung: AbbVie Deutschland GmbH & Co. KG, Ludwigshafen, Germany. Hauptthemen laut MEDLINE: Drug Development, Supervised Machine Learning. Die Arbeit wurde bisher 963-mal zitiert (Crossref) und nennt 50 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Despite increasing interest in using Artificial Intelligence (AI) and Machine Learning (ML) models for drug development, effectively interpreting their predictions remains a challenge, which limits their impact on clinical decisions. We address this issue by providing a practical guide to SHapley Additive exPlanations (SHAP), a popular feature‐based interpretability method, which can be seamlessly integrated into supervised ML models to gain a deeper understanding of their predictions, thereby enhancing their transparency and trustworthiness. This tutorial focuses on the application of SHAP analysis to standard ML black‐box models for regression and classification problems. We provide an overview of various visualization plots and their interpretation, available software for implementing SHAP, and highlight best practices, as well as special considerations, when dealing with binary endpoints and time‐series models. To enhance the reader's understanding for the method, we also apply it to inherently explainable regression models. Finally, we discuss the limitations and ongoing advancements aimed at tackling the current drawbacks of the method.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Drug DevelopmentSupervised Machine LearningHumansSoftware

Autor:innen und Institutionen

  • AbbVie Deutschland GmbH & Co. KG, Ludwigshafen, Germany.

ORCID-Profile: Ana Victoria Ponce-Bobadilla, Vanessa Schmitt, Sven Mensing, Sven Stodtmann

Zitieren

APA

Ponce-Bobadilla, A. V., Schmitt, V., Maier, C. S., Mensing, S., & Stodtmann, S. (2024). Practical guide to SHAP analysis: Explaining supervised machine learning model predictions in drug development. Clinical and Translational Science, 17(11). https://doi.org/10.1111/cts.70056

Vancouver

Ponce-Bobadilla AV, Schmitt V, Maier CS, Mensing S, Stodtmann S. Practical guide to SHAP analysis: Explaining supervised machine learning model predictions in drug development. Clinical and Translational Science. 2024;17(11). doi:10.1111/cts.70056

BibTeX
@article{PonceBobadilla2024,
  author = {Ponce-Bobadilla, Ana Victoria and Schmitt, Vanessa and Maier, Corinna S and Mensing, Sven and Stodtmann, Sven},
  title = {Practical guide to SHAP analysis: Explaining supervised machine learning model predictions in drug development},
  journal = {Clinical and Translational Science},
  year = {2024},
  volume = {17},
  number = {11},
  doi = {10.1111/cts.70056}
}
Kostenlos starten →