深層学習を用いた超音波画像からの離断性骨軟骨炎の検出における最適な画像認識モデルの検討

佐々木 研太; 藤田 大輔; 高辻 謙太; 琴浦 義浩; 南 昌孝; 小林 雄輔; 祐成 毅; 木田 圭重; 髙橋 謙治; 小橋 昌司

doi:10.24466/pacbfsa.36.0_9

DOI https://doi.org/10.24466/pacbfsa.36.0_9

Conference information

Name : [in Japanese]

Number : 36

Location : [in Japanese]

Date : December 16, 2023 - December 17, 2023

Optimal Image Recognition Model for Detecting Osteochondritis Dissecans from Ultrasound Images Using Deep Learning

*Kenta SASAKI, *Daisuke FUJITA, *Kenta TAKATSUJI, *Yoshihiro KOTOURA, *Masataka MINAMI, *Yusuke KOBAYASHI, *Tsuyoshi SUKENARI, *Yoshikazu KIDA, *Kenji TAKAHASHI, *Syoji KOBASHI

Author information

*Kenta SASAKI
Graduate School of Engineering, University of Hyogo
*Daisuke FUJITA
Graduate School of Engineering, University of Hyogo
*Kenta TAKATSUJI
Kyoto Prefectural University of Medicine
*Yoshihiro KOTOURA
Kyoto Chubu Medical Center
*Masataka MINAMI
Kyoto Kuramaguchi Medical Center
*Yusuke KOBAYASHI
Saiseikai Shiga Hospital
*Tsuyoshi SUKENARI
Kyoto Prefectural University of Medicine
*Yoshikazu KIDA
Kyoto Prefectural University of Medicine
*Kenji TAKAHASHI
Kyoto Prefectural University of Medicine
*Syoji KOBASHI
Graduate School of Engineering, University of Hyogo

Keywords: Deep Learning, Osteochondritis Dissecans, Computer-aided diagnosis, Elbow, Ultrasound image

CONFERENCE PROCEEDINGS FREE ACCESS

Pages 9-12

Details

Abstract

Osteochondritis dissecans (OCD) is one of the most common sports disorders in elementary and junior high school baseball players. OCD is often detected after it has progressed to a painful condition. Therefore, early detection in the early stages is considered to be important. In this study, we propose an OCD detection model using deep learning from ultrasound images to assist diagnosis. As a result, the OCD classification model using VGG16 achieved 0.845 in accuracy and 0.983 in AUC.

Corresponding author

Conference information

Register with J-STAGE for free!