Automated Renal Tumor Segmentation in Computed Tomography Images Using a Global Attention–Based DeepLabV3+ Model: Algorithm Development and Validation
JMIR Medical Informatics · 15.09.2026 · 14 · e78523-e78523
DOI: 10.2196/78523 ↗Kurz gefasst
Fachartikel in der Zeitschrift JMIR Medical Informatics, online erschienen am 15.09.2026. Verfasst von Yueyan Zhao und 10 weiteren Autor:innen. Beteiligte Einrichtung: School of Medical Information and Engineering, Xuzhou Medical University, KJL_E202 Room, 209 Tongshan Road, Yunlong District, Xuzhou, Jiangsu, 221004…. Hauptthemen laut MEDLINE: Kidney Neoplasms, Tomography, X-Ray Computed, Algorithms, Image Processing, Computer-Assisted. Die Arbeit nennt 51 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar.
Abstract (Originalsprache)
Background
The rising global incidence of renal tumors necessitates precise diagnostic interventions. Accurate segmentation of computed tomography (CT) scans is essential for nephron-sparing surgery and radiotherapy. However, conventional manual delineation is labor-intensive and prone to significant interobserver variability due to tumor morphological heterogeneity. There is an urgent clinical demand for robust, automated segmentation solutions.
Objective
This study aims to develop and validate GAM-DeepLabV3+, an automated framework designed to address boundary ambiguity and high false-positive rates in complex renal imaging scenarios.
Methods
We propose an optimized encoder-decoder architecture specifically tailored for renal mass detection. The framework incorporates three key innovations: (1) a lightweight MobileNetV2 backbone to minimize computational overhead for clinical deployment; (2) an Atrous Spatial Pyramid Pooling (ASPP) module to capture multiscale contextual information; and (3) a Global Attention Mechanism (GAM) in the decoder to enhance channel-spatial interactions, thereby refining boundary delineation by suppressing background noise. The model was rigorously evaluated on a private clinical dataset (n=218) and the KiTS19 benchmark (n=210).
Results
GAM-DeepLabV3+ consistently outperformed state-of-the-art baselines. On the private dataset, the model achieved a mean Dice similarity coefficient (DSC) of 0.939 (SD 0.008), significantly surpassing feature pyramid network (FPN; mean 0.893, SD 0.008; P
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- School of Medical Information and Engineering, Xuzhou Medical University, KJL_E202 Room, 209 Tongshan Road, Yunlong District, Xuzhou, Jiangsu, 221004, China, 86 15852036109.
- Department of Radiology, The Second Hospital & Clinical Medical School, Lanzhou University, Lanzhou, Gansu, China.
Zitieren
APA
Zhao, Y., Liu, J., Shao, L., Li, L., Liu, Z., Liu, Y., Wen, J., Hao, X., Li, S., Zhao, J., & Song, B. (2026). Automated Renal Tumor Segmentation in Computed Tomography Images Using a Global Attention–Based DeepLabV3+ Model: Algorithm Development and Validation. JMIR Medical Informatics, 14, e78523-e78523. https://doi.org/10.2196/78523
Vancouver
Zhao Y, Liu J, Shao L, Li L, Liu Z, Liu Y, et al. Automated Renal Tumor Segmentation in Computed Tomography Images Using a Global Attention–Based DeepLabV3+ Model: Algorithm Development and Validation. JMIR Medical Informatics. 2026;14:e78523-e78523. doi:10.2196/78523
BibTeX
@article{Zhao2026,
author = {Zhao, Yueyan and Liu, Jianqiang and Shao, Lingyu and Li, Lin and Liu, Zhaoqing and Liu, Yujie and Wen, Jiaxin and Hao, Xinyao and Li, Shuyan and Zhao, Jianhong and Song, Boming},
title = {Automated Renal Tumor Segmentation in Computed Tomography Images Using a Global Attention–Based DeepLabV3+ Model: Algorithm Development and Validation},
journal = {JMIR Medical Informatics},
year = {2026},
volume = {14},
pages = {e78523--e78523},
doi = {10.2196/78523}
}
EUVIMED