Graduate School and Research Center in Digital Sciences

UBS: A dimension-agnostic metric for concept vector interpretability applied to radiomics

Yeche, Hugo; Harrison, Justin; Berthier, Tess

iMIMIC 2019, 2nd International Workshop on Interpretability of Machine Intelligence in Medical Image Computing, and ML-CDS 2019, 9th International Workshop on Multimodal Learning for Clinical Decision Support, Held in Conjunction with MICCAI 2019, Shenzhen, China, October 17, 2019 / Also published in LNCS, Vol. 11797

Understanding predictions in Deep Learning (DL) models is crucial for domain experts without any DL expertise in order to justify resultant decision-making process. As of today, medical models are often based on hand-crafted features such as radiomics, though their link with neural network features remains unclear. To address the lack of interpretability, approaches based on human-understandable concepts such as TCAV have been introduced. These methods have shown promising results, though they are unsuited for continuous value concepts and their introduced metrics do not adapt well to high-dimensional spaces. To bridge the gap with radiomics-based models, we implement a regression concept vector showing the impact of radiomic features on the predictions of deep networks. In addition, we introduce a new metric with improved scaling to high-dimensional spaces, allowing comparison across multiple layers.

Doi Bibtex

Title:UBS: A dimension-agnostic metric for concept vector interpretability applied to radiomics
Keywords:Radiomics Concept vector Dimensionality Interpretability
Type:Conference
Language:English
City:Shenzhen
Country:CHINA
Date:
Department:Data Science
Eurecom ref:6075
Copyright: © Springer. Personal use of this material is permitted. The definitive version of this paper was published in iMIMIC 2019, 2nd International Workshop on Interpretability of Machine Intelligence in Medical Image Computing, and ML-CDS 2019, 9th International Workshop on Multimodal Learning for Clinical Decision Support, Held in Conjunction with MICCAI 2019, Shenzhen, China, October 17, 2019 / Also published in LNCS, Vol. 11797 and is available at : https://doi.org/10.1007/978-3-030-33850-3_2
Bibtex: @inproceedings{EURECOM+6075, doi = {https://doi.org/10.1007/978-3-030-33850-3_2}, year = {2019}, title = {{UBS}: {A} dimension-agnostic metric for concept vector interpretability applied to radiomics}, author = {{Y}eche, {H}ugo and {H}arrison, {J}ustin and {B}erthier, {T}ess}, booktitle = {i{MIMIC} 2019, 2nd {I}nternational {W}orkshop on {I}nterpretability of {M}achine {I}ntelligence in {M}edical {I}mage {C}omputing, and {ML}-{CDS} 2019, 9th {I}nternational {W}orkshop on {M}ultimodal {L}earning for {C}linical {D}ecision {S}upport, {H}eld in {C}onjunction with {MICCAI} 2019, {S}henzhen, {C}hina, {O}ctober 17, 2019 / {A}lso published in {LNCS}, {V}ol. 11797}, address = {{S}henzhen, {CHINA}}, month = {10}, url = {http://www.eurecom.fr/publication/6075} }