Explainable domain transfer of distant supervised cancer subtyping model via imaging-based rules extraction
OPEN ACCESS
Loading...
Author / Producer
Date
2023-04
Publication Type
Journal Article
ETH Bibliography
yes
Citations
Altmetric
OPEN ACCESS
Data
Rights / License
Abstract
Image texture analysis has for decades represented a promising opportunity for cancer assessment and disease progression evaluation, evolving in a discipline, i.e., radiomics. However, the road to a complete translation into clinical practice is still hampered by intrinsic limitations. As purely supervised classification models fail in devising robust imaging-based biomarkers for prognosis, cancer subtyping approaches would benefit from the employment of distant supervision, for instance exploiting survival/recurrence information. In this work, we assessed, tested, and validated the domain-generality of our previously proposed Distant Supervised Cancer Subtyping model on Hodgkin Lymphoma. We evaluate the model performance on two independent datasets coming from two hospitals, comparing and analyzing the results. Although successful and consistent, the com-parison confirmed the instability of radiomics due to an across-center lack of reproducibility, leading to explainable results in one center and poor interpretability in the other. We thus propose a Random Forest-based Explainable Transfer Model for testing the domain-invariance of imaging biomarkers extracted from retrospec-tive cancer subtyping. In doing so, we tested the predictive ability of cancer subtyping in a validation and perspective setting, which led to successful results and supported the domain-generality of the proposed approach. On the other hand, the extraction of decision rules enables to draw of risk factors and robust bio-markers to inform clinical decisions. This work shows the potentialities of the Distant Supervised Cancer Sub-typing model to be further evaluated in larger multi-center datasets, to reliably translate radiomics into medical practice. The code is available at this GitHub repository.
Permanent link
Publication status
published
External links
Editor
Book title
Journal / series
Volume
138
Pages / Article No.
102522
Publisher
Elsevier
Event
Edition / version
Methods
Software
Geographic location
Date collected
Date created
Subject
Cancer subtyping; Explainability; Image Clustering; Radiomics; Rule extraction; Domain transfer