Decentralised machine learning in healthcare and life sciences: Applying the technology acceptance model

Machine learning (ML) has significant potential for the healthcare sector. To implement novel technologies such as decentralised machine learning (DML), platform providers must overcome low acceptance levels and implementation hurdles. We used a conceptualised model for DML for healthcare and life s...

Full description

Saved in:
Bibliographic Details
Main Authors: Katrin Förster, Tobias Strauss
Format: Article
Language:English
Published: Elsevier 2025-06-01
Series:Telematics and Informatics Reports
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2772503025000131
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Machine learning (ML) has significant potential for the healthcare sector. To implement novel technologies such as decentralised machine learning (DML), platform providers must overcome low acceptance levels and implementation hurdles. We used a conceptualised model for DML for healthcare and life sciences, based on the technology acceptance model and the unified theory of the acceptance and use of technology, to determine what drives DML acceptance among users. Data were generated using a quantitative approach and analysed using partial least squares structural equation modelling, following a structured analysis approach. The results show that the DML platform providers have a great opportunity to increase user acceptance for collaborative projects when addressing institutional issues. The analysis of the model reveals a well-conceptualised, adjusted acceptance model for DML. This study provides valuable initial insights for DML platform providers on how to increase acceptance among healthcare users, improve usage, and create a better network for the decentralised collaborative training of an ML model. It contributes to the literature on acceptance research based on TAMs, especially in healthcare and life sciences, and adds a new dimension by considering DML as a technology-based service and not just an application.
ISSN:2772-5030