文献类型：期刊文献

英文题名：LoRA-PT: Low-rank adapting UNETR for hippocampus segmentation using principal tensor singular values and vectors

作者：He, Guanghua[1,2];Cheng, Wangang[2];Zhu, Hancan[2];Yu, Gaohang[1]

机构：[1]Hangzhou Dianzi Univ, Dept Math, Hangzhou 310018, Peoples R China;[2]Shaoxing Univ, Sch Math Phys & Informat, Shaoxing 312000, Peoples R China

年份：2025

卷号：170

外文期刊名：ARTIFICIAL INTELLIGENCE IN MEDICINE

收录：SCI-EXPANDED(收录号：WOS:001566535900001)、、EI(收录号：20253619108008)、Scopus(收录号：2-s2.0-105014826077)、WOS

基金：Gaohang Yu's work was supported in part by the National Natural Science Foundation of China (No. 12071104) ; Hancan Zhu's work was supported by the Humanities and Social Science Fund of the Ministry of Education of China (No. 23YJAZH232) ; and Guanghua He's work was supported by the Scientific Research Project of Shaoxing University (No. 20210038) .

语种：英文

外文关键词：Parameter-efficient fine-tuning; Tensor singular value decomposition; Hippocampus segmentation; Deep learning

外文摘要：The hippocampus is an important brain structure involved in various psychiatric disorders, and its automatic and accurate segmentation is vital for studying these diseases. Recently, deep learning-based methods have made significant progress in hippocampus segmentation. However, training deep neural network models requires substantial computational resources, time, and a large amount of labeled training data, which is frequently scarce in medical image segmentation. To address these issues, we propose LoRA-PT, a novel parameter-efficient finetuning (PEFT) method that transfers the pre-trained UNETR model from the BraTS2021 dataset to the hippo-campus segmentation task. Specifically, LoRA-PT divides the parameter matrix of the transformer structure into three distinct sizes, yielding three third-order tensors. These tensors are decomposed using tensor singular value decomposition to generate low-rank tensors consisting of the principal singular values and vectors, with the remaining singular values and vectors forming the residual tensor. During fine-tuning, only the low-rank tensors (i.e., the principal tensor singular values and vectors) are updated, while the residual tensors remain unchanged. We validated the proposed method on three public hippocampus datasets, and the experimental results show that LoRA-PT outperformed state-of-the-art PEFT methods in segmentation accuracy while significantly reducing the number of parameter updates. Our source code is available at https://github.com/WangangCheng/LoRA-PT/tree /LoRA-PT.