Abstract
In knee MRI analysis, classical radiomic features are designed to quantify image appearance and intensity patterns. Compared with end-to-end deep learning (DL) models trained for disease classification, radiomics pipelines with low-dimensional parametric classifiers offer enhanced transparency and interpretability, yet often underperform because of the reliance on population-level predefined feature sets. Recent work on adaptive radiomics uses DL to predict feature weights over a large radiomic pool, then thresholds these weights to retain the “top-k” features from F (often F = |F| ~ 10^3). However, such marginal ranking can over-admit redundant descriptors and overlook complementary feature interactions. We propose a patient-specific feature-set selection framework that predicts a single compact feature set per subject, targeting complementary and diverse evidence rather than marginal top-k features. To overcome the intractable combinatorial search space of (F choose k) features, our method utilizes a two-stage retrieval strategy (i) randomly sample diverse candidate feature sets, then (ii) rank these sets with a learned scoring function to select a high-performing “top-1” feature set for the specific patient. The system consists of a feature-set scoring function, and a downstream classifier that performs the final diagnosis using the selected set. We empirically show that the proposed two-stage retrieval approximates the original exhaustive (F choose k) selection, with a 95th-percentile error of 0.0055. Validating on tasks including ACL tear detection and Kellgren-Lawrence grading for osteoarthritis, the experimental results achieve diagnostic performance, outperforming the top-k approach with the same k values, and competitive with end-to-end DL models while maintaining high transparency. The model generates [ABSTRACT CONTINUES ON NEXT PAGE - PLEASE COMPLETE FROM THE PDF]
Links to Paper and Supplementary Materials
Main Paper (Open Access Version): https://papers.miccai.org/miccai-2026-sat/paper/iMIMIC_006.pdf
SharedIt Link: Not yet available
SpringerLink (DOI): Not yet available
Supplementary Material: Not Submitted
Link to Open Review
Open Review Page: Not Available
BibTex
@InProceedings{CheYax_Retrieving_MICCAISAT2026,
author = { Chen, Yaxi AND Ni, Simin AND Zhang, Jingjing AND Saeed, Shaheer U. AND Wang, Yipei AND Ivanova, Aleksandra AND Hargunani, Rikin AND Liu, Chaozong AND Huang, Jie AND Hu, Yipeng},
title = { { Retrieving Patient-Specific Radiomic Feature Sets for Transparent Knee MRI Assessment } },
booktitle = {Medical Image Computing and Computer Assisted Intervention -- MICCAI 2026 Workshops and Challenges},
year = {2026},
publisher = {Springer Nature Switzerland},
volume = {LNCS 17279},
month = {pending},
page = {pending}
}