Mixture of Multicenter Experts in Multimodal AI for Debiased Radiotherapy Target Delineation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Oh, Yujin, Park, Sangjoon, Li, Xiang, Jin, Pengfei, Wang, Yi, Paly, Jonathan, Efstathiou, Jason, Chan, Annie, Kim, Jun Won, Byun, Hwa Kyung, Lee, Ik Jae, Cho, Jaeho, Wee, Chan Woo, Shu, Peng, Wang, Peilong, Yu, Nathan, Holmes, Jason, Ye, Jong Chul, Li, Quanzheng, Liu, Wei, Koom, Woong Sub, Kim, Jin Sung, Kim, Kyungsang
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909795398713344
author Oh, Yujin
Park, Sangjoon
Li, Xiang
Jin, Pengfei
Wang, Yi
Paly, Jonathan
Efstathiou, Jason
Chan, Annie
Kim, Jun Won
Byun, Hwa Kyung
Lee, Ik Jae
Cho, Jaeho
Wee, Chan Woo
Shu, Peng
Wang, Peilong
Yu, Nathan
Holmes, Jason
Ye, Jong Chul
Li, Quanzheng
Liu, Wei
Koom, Woong Sub
Kim, Jin Sung
Kim, Kyungsang
author_facet Oh, Yujin
Park, Sangjoon
Li, Xiang
Jin, Pengfei
Wang, Yi
Paly, Jonathan
Efstathiou, Jason
Chan, Annie
Kim, Jun Won
Byun, Hwa Kyung
Lee, Ik Jae
Cho, Jaeho
Wee, Chan Woo
Shu, Peng
Wang, Peilong
Yu, Nathan
Holmes, Jason
Ye, Jong Chul
Li, Quanzheng
Liu, Wei
Koom, Woong Sub
Kim, Jin Sung
Kim, Kyungsang
contents Clinical decision-making reflects diverse strategies shaped by regional patient populations and institutional protocols. However, most existing medical artificial intelligence (AI) models are trained on highly prevalent data patterns, which reinforces biases and fails to capture the breadth of clinical expertise. Inspired by the recent advances in Mixture of Experts (MoE), we propose a Mixture of Multicenter Experts (MoME) framework to address AI bias in the medical domain without requiring data sharing across institutions. MoME integrates specialized expertise from diverse clinical strategies to enhance model generalizability and adaptability across medical centers. We validate this framework using a multimodal target volume delineation model for prostate cancer radiotherapy. With few-shot training that combines imaging and clinical notes from each center, the model outperformed baselines, particularly in settings with high inter-center variability or limited data availability. Furthermore, MoME enables model customization to local clinical preferences without cross-institutional data exchange, making it especially suitable for resource-constrained settings while promoting broadly generalizable medical AI.
format Preprint
id arxiv_https___arxiv_org_abs_2410_00046
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Mixture of Multicenter Experts in Multimodal AI for Debiased Radiotherapy Target Delineation
Oh, Yujin
Park, Sangjoon
Li, Xiang
Jin, Pengfei
Wang, Yi
Paly, Jonathan
Efstathiou, Jason
Chan, Annie
Kim, Jun Won
Byun, Hwa Kyung
Lee, Ik Jae
Cho, Jaeho
Wee, Chan Woo
Shu, Peng
Wang, Peilong
Yu, Nathan
Holmes, Jason
Ye, Jong Chul
Li, Quanzheng
Liu, Wei
Koom, Woong Sub
Kim, Jin Sung
Kim, Kyungsang
Image and Video Processing
Computer Vision and Pattern Recognition
Machine Learning
Clinical decision-making reflects diverse strategies shaped by regional patient populations and institutional protocols. However, most existing medical artificial intelligence (AI) models are trained on highly prevalent data patterns, which reinforces biases and fails to capture the breadth of clinical expertise. Inspired by the recent advances in Mixture of Experts (MoE), we propose a Mixture of Multicenter Experts (MoME) framework to address AI bias in the medical domain without requiring data sharing across institutions. MoME integrates specialized expertise from diverse clinical strategies to enhance model generalizability and adaptability across medical centers. We validate this framework using a multimodal target volume delineation model for prostate cancer radiotherapy. With few-shot training that combines imaging and clinical notes from each center, the model outperformed baselines, particularly in settings with high inter-center variability or limited data availability. Furthermore, MoME enables model customization to local clinical preferences without cross-institutional data exchange, making it especially suitable for resource-constrained settings while promoting broadly generalizable medical AI.
title Mixture of Multicenter Experts in Multimodal AI for Debiased Radiotherapy Target Delineation
topic Image and Video Processing
Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2410.00046