University of Twente Student Theses


Towards explainable machine learning for prediction of disease progression

Berendse, S.E. (2023) Towards explainable machine learning for prediction of disease progression.

[img] PDF
Abstract:This research addresses the current problems surrounding interpretability of machine learning techniques in the field of prediction of disease progression. The use of machine learning in the diagnosis of diseases and the prediction of disease progression is a recent and promising development. Such predictions have potential to help physicians to make better informed decisions based on patient data, ultimately improving the patient's quality of life or altering the outcome of treatment. However, the interpretability and transparency of machine learning models aimed at this is lagging behind. To achieve improved interpretability and transparency, we first perform a systematic literature review to identify the state-of-the-art in machine learning for disease progression modelling and challenges related to this context. Based on the review, we design and develop a pipeline consisting of data preparation, prediction and explanation. Furthermore, we provide a number of concrete recommendations and directions for future research, such as improving input flexibility of the prediction model and improving the visualisation of generated explanations. Based on the results of this research, we conclude that the proposed pipeline achieves the goal of integrating a state-of-the-art prediction model and the LIME framework to make the model more transparent and interpretable.
Item Type:Essay (Master)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:54 computer science
Programme:Business Information Technology MSc (60025)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page