Ante- and Post-Hoc Explanations for Prediction Models of Cisplatin-Induced Acute Kidney Injury: A Comparative Study

Tatsuya Nishizawa, Shogo Hanabusa, Yoshitaka Kameya, Kazuo Takahashi, Naotake Tsuboi, Tomohiro Mizuno

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Whereas cisplatin is considered as an effective anticancer drug, it can cause cisplatin-induced acute kidney injury (Cis-AKI) as a side-effect. We trained a prediction model of Cis-AKI to predict the incidence of Cis-AKI using Generalized Additive Models (GAMs) and GAMs plus Interactions (GA2Ms), which can provide us with ante-hoc (or model-based) explanations for their decisions. Furthermore, we trained XGBoost and used SHapley Additive exPlanations (SHAP) in order to obtain post-hoc explanations. We then compared these ante-hoc and post-hoc explanation methods in terms of consistency with current medical insights on Cis-AKI.

Original languageEnglish
Title of host publicationICMHI 2023 - 2023 the 7th International Conference on Medical and Health Informatics
PublisherAssociation for Computing Machinery
Pages66-71
Number of pages6
ISBN (Electronic)9798400700712
DOIs
Publication statusPublished - 12-05-2023
Event7th International Conference on Medical and Health Informatics, ICMHI 2023 - Kyoto, Japan
Duration: 12-05-202314-05-2023

Publication series

NameACM International Conference Proceeding Series

Conference

Conference7th International Conference on Medical and Health Informatics, ICMHI 2023
Country/TerritoryJapan
CityKyoto
Period12-05-2314-05-23

All Science Journal Classification (ASJC) codes

  • Human-Computer Interaction
  • Computer Networks and Communications
  • Computer Vision and Pattern Recognition
  • Software

Fingerprint

Dive into the research topics of 'Ante- and Post-Hoc Explanations for Prediction Models of Cisplatin-Induced Acute Kidney Injury: A Comparative Study'. Together they form a unique fingerprint.

Cite this