Development and Validation of a Convolutional Neural Network Framework Based on Ultrasound Imaging for Multi-Classification of Superficial Soft Tissue Masses
Academic Radiology · 2026-09
DOI: 10.1016/j.acra.2026.08.092 ↗Kurz gefasst
Fachartikel in der Zeitschrift Academic Radiology, 2026-09. Verfasst von Mengjie Wu und 17 weiteren Autor:innen. Beteiligte Einrichtung: Department of Medical Ultrasound, The First Affiliated Hospital with Nanjing Medical University, Nanjing, Jiangsu, PR China (M.W., A.L.). Die Arbeit nennt 39 Literaturangaben. Gefördert von: Ministry of Science and Technology of the People's Republic of China, National Natural Science Foundation of China, National Key Research and Development Program of China.
Abstract (Originalsprache)
RATIONALE AND OBJECTIVES: Superficial soft tissue masses (STMs) represent a diagnostic dilemma in clinical practice, with ultrasound (US) being the front-line imaging modality available globally. However, the high complexity of STMs in imaging makes subjective evaluation highly dependent on experience, frequently causing inconsistent malignancy assessments. This inconsistency triggers unnecessary benign biopsies and delays treatment for malignant STMs. We developed ST-USNet, a multitask convolutional neural network framework to classify superficial STMs based on manually drawn regions of interest. MATERIALS AND METHODS: This retrospective study included US images of 3168 patients (median age, 58 years; IQR, 46-68 years) with STMs from four institutions between March 2015 and October 2024. The ST-USNet was developed and validated on multi-center data. Its performance was then evaluated on a separate, independent test cohort. The diagnostic performance of ST-USNet was compared with that of radiologists using McNemar tests. RESULTS: The ST-USNet was composed of four sub-models (SM-1, SM-2, SM-a, and SM-b). In the validation cohort, SM-1 was trained to distinguish malignant from benign STMs (AUC: 0.984); SM-2 was to classify the malignant STM subtypes including sarcoma, lymphoma, and metastatic carcinoma (AUC: 0.932, 0.909, 0.922); SM-a was to discriminate between aggressive and indolent lymphoma (AUC: 0.951). SM-b was designed to explore the identification of metastatic carcinoma origin (thyroid, breast, respiratory, digestive, and reproductive systems) as a preliminary analysis; however, due to limited sample sizes, these results should be interpreted as exploratory. ST-USNet achieved high AUCs on the validation cohort and remained effective, albeit with slightly lower performance, on an independent test cohort. In a preliminary reader study (4 radiologists, 85 cases), ST-USNet either significantly outperformed senior radiologists (p = 0.012) or performed comparably to them (p = 0.267, 0.092, 0.332) in all classification tasks and effectively enhanced diagnostic accuracy for both junior and senior radiologists when used as an assistive tool. CONCLUSION: ST-USNet serves as an effective and practical decision support system for superficial STMs classification in clinical oncology, though multi-center prospective validation and continuous model updating are required before clinical deployment.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Medical Ultrasound, The First Affiliated Hospital with Nanjing Medical University, Nanjing, Jiangsu, PR China (M.W., A.L.).
- Department of Medical Ultrasound, Shanghai Tenth People's Hospital, Shanghai Engineering Research Center of Ultrasound Diagnosis and Treatment, School of Medicine, Tongji University, Shanghai, China (Y.W., L.G.).
- MedAI Technology (Wuxi) Co., Ltd, Wuxi, Jiangsu, PR China (C.L.).
- Department of Ultrasound, Nanjing Drum Tower Hospital, Affiliated Hospital of Medical School, Nanjing University, Nanjing, Jiangsu, PR China (S.W.).
- Technical University of Munich, Munich, Bavaria, Germany (X.Z.).
- Department of Pathology, The First Affiliated Hospital with Nanjing Medical University, Nanjing, Jiangsu, PR China (H.H.).
Förderung
- Ministry of Science and Technology of the People's Republic of China
- National Natural Science Foundation of China
- National Key Research and Development Program of China
Zitieren
APA
Wu, M., Wu, Y., Li, A., Shi, Y., Guo, L., Li, C., Wei, S., Zhu, X., Hua, H., Zhang, Y., Sun, L., Hu, J., Mou, L., Yu, X., Wang, X., Ouyang, R., Fu, H., & Xu, H. (2026). Development and Validation of a Convolutional Neural Network Framework Based on Ultrasound Imaging for Multi-Classification of Superficial Soft Tissue Masses. Academic Radiology. https://doi.org/10.1016/j.acra.2026.08.092
Vancouver
Wu M, Wu Y, Li A, Shi Y, Guo L, Li C, et al. Development and Validation of a Convolutional Neural Network Framework Based on Ultrasound Imaging for Multi-Classification of Superficial Soft Tissue Masses. Academic Radiology. 2026. doi:10.1016/j.acra.2026.08.092
BibTeX
@article{Wu2026,
author = {Wu, Mengjie and Wu, Yulu and Li, Ao and Shi, Yilei and Guo, Lehang and Li, Chunlei and Wei, Shuping and Zhu, Xiaoxiang and Hua, Hongjin and Zhang, Yaqin and Sun, Liping and Hu, Jingliang and Mou, Lichao and Yu, Xiaojie and Wang, Xinyue and Ouyang, Ruilong and Fu, Huijun and Xu, Huixiong},
title = {Development and Validation of a Convolutional Neural Network Framework Based on Ultrasound Imaging for Multi-Classification of Superficial Soft Tissue Masses},
journal = {Academic Radiology},
year = {2026},
doi = {10.1016/j.acra.2026.08.092}
}
EUVIMED