Explaining Machine Learning Predictions of Decision Support Systems in Healthcare
Abstract: Artificial Intelligence (AI) methods, which are often based on Machine Learning (ML) algorithms, are also applied in the healthcare domain to provide predictions to physicians and patients based on electronic health records (EHRs), such as history of laboratory values, applied procedures and diagnoses. The question about these predictions “Why Should I Trust You?” encapsulates the issue with ML black boxes. Therefore, explaining the reasons for these ML predictions to physicians and patients is crucial to allow them to decide whether the prediction is applicable or not. In this paper, we explained and evaluated two prediction explanation methods for healthcare professionals (physicians and nurses). We compared two model-agnostic explanation methods based on global feature importance and local feature importance. We evaluated the user trust and reliance (UTR) for the explanation results of each method in a user study based on real patients’ electronic health records (EHR) and the feedback of healthcare professionals. Based on the user study, we observed that both methods have strengths and weaknesses according to the patients’ data, especially based on the data size of the patient. When the amount of data is small, global feature importance is enough to use. However, when the patient’s data size is big, using a local feature importance method makes more sense. As future work, we will develop a hybrid explanation method (by combining these methods automatically with a smart setting) to obtain higher and more stable performance results in terms of user trust and reliance.
- Standort
-
Deutsche Nationalbibliothek Frankfurt am Main
- Umfang
-
Online-Ressource
- Sprache
-
Englisch
- Erschienen in
-
Explaining Machine Learning Predictions of Decision Support Systems in Healthcare ; volume:8 ; number:2 ; year:2022 ; pages:117-120 ; extent:4
Current directions in biomedical engineering ; 8, Heft 2 (2022), 117-120 (gesamt 4)
- Urheber
-
Polat Erdeniz, Seda
Veeranki, Sai
Schrempf, Michael
Jauk, Stefanie
Ngoc Trang Tran, Thi
Felfernig, Alexander
Kramer, Diether
Leodolter, Werner
- DOI
-
10.1515/cdbme-2022-1031
- URN
-
urn:nbn:de:101:1-2022090315264050322378
- Rechteinformation
-
Open Access; Der Zugriff auf das Objekt ist unbeschränkt möglich.
- Letzte Aktualisierung
-
15.08.2025, 07:28 MESZ
Datenpartner
Deutsche Nationalbibliothek. Bei Fragen zum Objekt wenden Sie sich bitte an den Datenpartner.
Beteiligte
- Polat Erdeniz, Seda
- Veeranki, Sai
- Schrempf, Michael
- Jauk, Stefanie
- Ngoc Trang Tran, Thi
- Felfernig, Alexander
- Kramer, Diether
- Leodolter, Werner
Ähnliche Objekte (12)
Contextualizing Support Vector Machine Predictions
Explaining human mobility predictions through a pattern matching algorithm
Empowering advanced medical decision-making through machine learning in healthcare
Explaining predictions and attacks in federated learning via random forests
PROVERB - A System Explaining Machine-Found Proofs
Macroeconomic predictions using payments data and machine learning
Machine Learning Methodologies for Interpretable Compound Activity Predictions
Direct Machine Learning Predictions of C 3 Pathways
Biases Model Machine Learning Predictions in Protein Biology
A comparative analysis of machine learning methods for classification type decision problems in healthcare
Explaining compound activity predictions with a substructure-aware loss for graph neural networks
Privacy-preserving dataset combination and Lasso regression for healthcare predictions
Contextualizing Support Vector Machine Predictions
Explaining human mobility predictions through a pattern matching algorithm
Empowering advanced medical decision-making through machine learning in healthcare
Explaining predictions and attacks in federated learning via random forests
PROVERB - A System Explaining Machine-Found Proofs
Macroeconomic predictions using payments data and machine learning
Machine Learning Methodologies for Interpretable Compound Activity Predictions
Direct Machine Learning Predictions of C 3 Pathways
Biases Model Machine Learning Predictions in Protein Biology
A comparative analysis of machine learning methods for classification type decision problems in healthcare
Explaining compound activity predictions with a substructure-aware loss for graph neural networks
Privacy-preserving dataset combination and Lasso regression for healthcare predictions
Contextualizing Support Vector Machine Predictions
Explaining human mobility predictions through a pattern matching algorithm
Empowering advanced medical decision-making through machine learning in healthcare
Explaining predictions and attacks in federated learning via random forests
PROVERB - A System Explaining Machine-Found Proofs
Macroeconomic predictions using payments data and machine learning
Machine Learning Methodologies for Interpretable Compound Activity Predictions
Direct Machine Learning Predictions of C 3 Pathways
Biases Model Machine Learning Predictions in Protein Biology
A comparative analysis of machine learning methods for classification type decision problems in healthcare
Explaining compound activity predictions with a substructure-aware loss for graph neural networks