Deep learning‐based classification of acute scrotum using single ultrasound images
BJU International · 2026
DOI: 10.1111/bju.70091 ↗Kurz gefasst
Multizentrische Studie in der Zeitschrift BJU International, 2026. Verfasst von Sahyun Pak und 8 weiteren Autor:innen. Beteiligte Einrichtung: Department of Urology, Kangnam Sacred Heart Hospital, Hallym University College of Medicine, Seoul, Korea. Hauptthemen laut MEDLINE: Deep Learning, Scrotum, Spermatic Cord Torsion. Die Arbeit nennt 30 Literaturangaben. Gefördert von: Korean Society for Urological Ultrasonography, Hallym University Research Fund.
Abstract (Originalsprache)
Objective
To develop a deep learning model for differential diagnosis of acute scrotum using single ultrasound (US) images. Patients and Methods We analysed 1172 patients with acute scrotal pain evaluated by Doppler US at four hospitals. From each case, we selected a representative axial colour Doppler US image. We trained a binary classification model to distinguish torsion from non‐torsion using an EfficientNet architecture. The dataset was split 70% for training and 30% for validation. We addressed class imbalance with data augmentation and class weighting. Class Activation Mapping was used to interpret model decisions.
Results
The model achieved robust performance: accuracy 97%, precision 98%, sensitivity 97%, and F1 score 97%. Class activation mapping heatmaps localised decision‐making to pathologically critical regions, including absent testicular blood flow and whirlpool signs. In a 20‐patient prospective pilot study, the system correctly identified both surgically confirmed torsion cases, with one non‐torsion case misclassified as torsion.
Conclusions
A deep learning model demonstrated promising diagnostic performance in differentiating acute scrotal emergencies using single US images. Its feasibility was preliminarily assessed in a small pilot study. These findings support further investigation, with larger and more balanced multicentre studies required to establish clinical utility and effective workflow integration.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Urology, Kangnam Sacred Heart Hospital, Hallym University College of Medicine, Seoul, Korea.
- Department of Urology, Dongtan Sacred Heart Hospital, Hallym University College of Medicine, Hwaseong, Korea.
- Department of Urology, Chuncheon Sacred Heart Hospital, Hallym University College of Medicine, Chuncheon, Korea.
- Department of Urology, Hallym University Sacred Heart Hospital, Hallym University College of Medicine, Anyang, Korea.
- Department of Radiology, Kangnam Sacred Heart Hospital, Hallym University College of Medicine, Seoul, Korea.
ORCID-Profile: Sahyun Pak, Jong Keun Kim
Förderung
- Korean Society for Urological Ultrasonography
- Hallym University Research Fund
Zitieren
APA
Pak, S., Park, S. G., Kim, J. K., Lee, S. H., Lee, W. K., Oh, C. Y., Woo, J. Y., Cho, S. T., & Lee, Y. G. (2026). Deep learning‐based classification of acute scrotum using single ultrasound images. BJU International. https://doi.org/10.1111/bju.70091
Vancouver
Pak S, Park SG, Kim JK, Lee SH, Lee WK, Oh CY, et al. Deep learning‐based classification of acute scrotum using single ultrasound images. BJU International. 2026. doi:10.1111/bju.70091
BibTeX
@article{Pak2026,
author = {Pak, Sahyun and Park, Sung Gon and Kim, Jong Keun and Lee, Seong Ho and Lee, Won Ki and Oh, Cheol Young and Woo, Ji Young and Cho, Sung Tae and Lee, Young Goo},
title = {Deep learning‐based classification of acute scrotum using single ultrasound images},
journal = {BJU International},
year = {2026},
doi = {10.1111/bju.70091}
}
EUVIMED