Tingting Zhou1, Haozhe Zhuang1, Shiju Yan1, Erze Xie1, Yibo Ma2, Tao Zhang1, Tianxiang Yu1, Shuang Deng1
1School of Health Science and Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China. 2Department of Ultrasound, the Third Affiliated Hospital of Soochow University, Changzhou 213000, Jiangsu, China.
Address correspondence to: Shiju Yan, School of Health Science and Engineering, University of Shanghai for Science and Technology, No.516 Jungong Road, Shanghai 200093, China. Email: yanshiju@usst.edu.cn; Yibo Ma, Department of Ultrasound, the Third Affiliated Hospital of Soochow University, No.185 Juqian Street, Changzhou 213000, Jiangsu, China. Email: mayibo@czfph.com.
DOI: https://doi.org/10.61189/594641kmfbkw
Highlights
● Automated pleural line identification: A method was introduced to automatically identify pleural lines in lung ultrasound images, ensuring diagnosis speed and accuracy.
● High reliability: An average of 90.45% identification rate of pleural lines was achieved in a comprehensive experiment on 890 ultrasound videos, highlighting its broad applicability and reliability.
● Efficient integration: The algorithm's rapid processing (1.36 seconds for a 5-second video) makes it suitable for seamless integration into ultrasound instrument software, aiding clinicians in diagnosing conditions like pneumo-thorax more efficiently.