Multi-modality Contrastive Learning for Sarcopenia Screening from Hip X-rays and Clinical Information

Qiangguo Jin, Changjiang Zou, Hui Cui, Changming Sun, Shu Wei Huang, Yi Jie Kuo, Ping Xuan, Leilei Cao, Ran Su, Leyi Wei, Henry B.L. Duh, Yu Pin Chen

研究成果: 書貢獻/報告類型會議貢獻

1 引文 斯高帕斯(Scopus)

摘要

Sarcopenia is a condition of age-associated muscle degeneration that shortens the life expectancy in those it affects, compared to individuals with normal muscle strength. Accurate screening for sarcopenia is a key process of clinical diagnosis and therapy. In this work, we propose a novel multi-modality contrastive learning (MM-CL) based method that combines hip X-ray images and clinical parameters for sarcopenia screening. Our method captures the long-range information with Non-local CAM Enhancement, explores the correlations in visual-text features via Visual-text Feature Fusion, and improves the model’s feature representation ability through Auxiliary contrastive representation. Furthermore, we establish a large in-house dataset with 1,176 patients to validate the effectiveness of multi-modality based methods. Significant performances with an AUC of 84.64%, ACC of 79.93%, F1 of 74.88%, SEN of 72.06%, SPC of 86.06%, and PRE of 78.44%, show that our method outperforms other single-modality and multi-modality based methods.
原文英語
主出版物標題Medical Image Computing and Computer Assisted Intervention – MICCAI 2023 - 26th International Conference, Proceedings
編輯Hayit Greenspan, Hayit Greenspan, Anant Madabhushi, Parvin Mousavi, Septimiu Salcudean, James Duncan, Tanveer Syeda-Mahmood, Russell Taylor
發行者Springer Science and Business Media Deutschland GmbH
頁面85-94
頁數10
ISBN(列印)9783031439865
DOIs
出版狀態已發佈 - 2023
事件26th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2023 - Vancouver, 加拿大
持續時間: 10月 8 202310月 12 2023

出版系列

名字Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
14225 LNCS
ISSN(列印)0302-9743
ISSN(電子)1611-3349

會議

會議26th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2023
國家/地區加拿大
城市Vancouver
期間10/8/2310/12/23

ASJC Scopus subject areas

  • 理論電腦科學
  • 一般電腦科學

指紋

深入研究「Multi-modality Contrastive Learning for Sarcopenia Screening from Hip X-rays and Clinical Information」主題。共同形成了獨特的指紋。

引用此