Abstract
Many diseases are diagnosed using a combination of imaging findings and patient-specific clinical information. While medical images provide valuable diagnostic information, clinical decision-making also relies on additional demographics or clinical details, such as age or sex. Integrating imaging and clinical data has the potential to improve classification beyond what either modality achieves alone.
We propose a novel contribution-based fusion (CBF) method to combine image-derived measurements with clinical tabular data for disease classification. Unlike many existing multimodal approaches that rely on entangled latent representations and focus primarily on predictive performance, CBF explicitly quantifies the contribution to the final decision of each modality and its specific inputs. We demonstrate our approach on the task of screening for Developmental Dysplasia of the Hip (DDH), where patients are classified as normal or abnormal.
Our proposed method achieves comparable or superior classification performance to established multimodal fusion approaches, while providing modality- and input-level contributions that are reliably associated with the model’s decision-making process. By reporting how much each modality (imaging versus clinical variables) influenced a decision, the proposed approach enables direct inspection of how image and clinical information influence the final classification, providing concrete information that may help clinicians understand the model’s predictions.
Links to Paper and Supplementary Materials
Main Paper (Open Access Version): https://papers.miccai.org/miccai-2026-sat/paper/ML-CDS2026_004.pdf
SharedIt Link: Not yet available
SpringerLink (DOI): Not yet available
Supplementary Material: Not Submitted
Link to Open Review
Open Review Page: https://openreview.net/profile?id=~Allison_Clement1
BibTex
@InProceedings{CleAll_Quantifying_MICCAISAT2026,
author = { Clement, Allison AND Voiculescu, Irina},
title = { { Quantifying Contributions within Multimodal Fusion for Clinical Decisions } },
booktitle = {Medical Image Computing and Computer Assisted Intervention -- MICCAI 2026 Workshops and Challenges},
year = {2026},
publisher = {Springer Nature Switzerland},
volume = {LNCS 17262},
month = {pending},
page = {pending}
}