Explainable domain transfer of distant supervised cancer subtyping model via imaging-based rules extraction
dc.contributor.author
Cavinato, Lara
dc.contributor.author
Gozzi, Noemi
dc.contributor.author
Sollini, Martina
dc.contributor.author
Kirienko, Margarita
dc.contributor.author
Carlo-Stella, Carmelo
dc.contributor.author
Rusconi, Chiara
dc.contributor.author
Chiti, Arturo
dc.contributor.author
Ieva, Francesca
dc.date.accessioned
2023-05-30T07:47:55Z
dc.date.available
2023-05-30T03:03:24Z
dc.date.available
2023-05-30T07:47:55Z
dc.date.issued
2023-04
dc.identifier.issn
0933-3657
dc.identifier.other
10.1016/j.artmed.2023.102522
en_US
dc.identifier.uri
http://hdl.handle.net/20.500.11850/614138
dc.description.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.
en_US
dc.language.iso
en
en_US
dc.publisher
Elsevier
en_US
dc.subject
Cancer subtyping
en_US
dc.subject
Explainability
en_US
dc.subject
Image Clustering
en_US
dc.subject
Radiomics
en_US
dc.subject
Rule extraction
en_US
dc.subject
Domain transfer
en_US
dc.title
Explainable domain transfer of distant supervised cancer subtyping model via imaging-based rules extraction
en_US
dc.type
Journal Article
dc.date.published
2023-03-04
ethz.journal.title
Artificial intelligence in medicine
ethz.journal.volume
138
en_US
ethz.journal.abbreviated
Artif Intell Med
ethz.pages.start
102522
en_US
ethz.size
13 p.
en_US
ethz.identifier.wos
ethz.publication.place
Amsterdam
en_US
ethz.publication.status
published
en_US
ethz.date.deposited
2023-05-30T03:03:32Z
ethz.source
WOS
ethz.eth
yes
en_US
ethz.availability
Metadata only
en_US
ethz.rosetta.installDate
2024-02-02T23:48:32Z
ethz.rosetta.lastUpdated
2024-02-02T23:48:32Z
ethz.rosetta.versionExported
true
ethz.COinS
ctx_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.atitle=Explainable%20domain%20transfer%20of%20distant%20supervised%20cancer%20subtyping%20model%20via%20imaging-based%20rules%20extraction&rft.jtitle=Artificial%20intelligence%20in%20medicine&rft.date=2023-04&rft.volume=138&rft.spage=102522&rft.issn=0933-3657&rft.au=Cavinato,%20Lara&Gozzi,%20Noemi&Sollini,%20Martina&Kirienko,%20Margarita&Carlo-Stella,%20Carmelo&rft.genre=article&rft_id=info:doi/10.1016/j.artmed.2023.102522&
Files in this item
Files | Size | Format | Open in viewer |
---|---|---|---|
There are no files associated with this item. |
Publication type
-
Journal Article [132204]