Automated classification of angle-closure mechanisms based on anterior segment optical coherence tomography images via deep learning

被引：0

作者：

Zhang, Ye ^{[1
]}

Zhang, Xiaoyue ^{[2
]}

Zhang, Qing ^{[3
]}

Lv, Bin ^{[2
]}

Hu, Man ^{[4
]}

Lv, Chuanfeng ^{[2
]}

Ni, Yuan ^{[2
]}

Xie, Guotong ^{[2
,5
,6
]}

Li, Shuning ^{[1
]}

Zebardast, Nazlee ^{[7
]}

Shweikh, Yusrah ^{[8
]}

Wang, Ningli ^{[1
,3
]}

机构：

[1] Capital Med Univ, Beijing Tongren Hosp, Beijing Tongren Eye Ctr, Beijing Key Lab Ophthalmol & Visual Sci, 1 Dong Jiao Min Xiang St, Beijing 100730, Peoples R China

[2] Ping An Healthcare Technol, Beijing 100027, Peoples R China

[3] Beijing Inst Ophthalmol, Beijing, Peoples R China

[4] Capital Med Univ, Natl Key Discipline Pediat, Minist Educ, Dept Endocrinol,Beijing Childrens Hosp, Beijing 100045, Peoples R China

[5] Ping Hlth Cloud Co Ltd, Shenzhen, Peoples R China

[6] Ping Int Smart City Technol Co Ltd, Shenzhen, Peoples R China

[7] Harvard Med Sch, Massachusetts Eye & Ear Infirm, Dept Ophthalmol, Boston, MA USA

[8] Univ Hosp Sussex NHS Fdn Trust, Sussex Eye Hosp, Brighton, Sussex, England

来源：

HELIYON | 2024年 / 10卷 / 15期

关键词：

Deep learning; Angle-closure mechanisms; Anterior segment optical coherence; tomography; Automated classification; ULTRASOUND BIOMICROSCOPY; QUANTITATIVE-ANALYSIS; IRIS CHANGES; GLAUCOMA; PREVALENCE; CHINESE; MYDRIASIS; PEOPLE; LENS;

D O I：

10.1016/j.heliyon.2024.e35236

中图分类号：

O [数理科学和化学]; P [天文学、地球科学]; Q [生物科学]; N [自然科学总论];

学科分类号：

07 ; 0710 ; 09 ;

摘要：

Purpose: To develop and validate deep learning algorithms that can identify and classify angleclosure (AC) mechanisms using anterior segment optical coherence tomography (AS-OCT) images. Methods: This cross-sectional study included participants of the Handan Eye Study aged >= 35 years with AC detected via gonioscopy or on the AS-OCT images. These images were classified by human experts into the following to indicate the predominant AC mechanism (ground truth): pupillary block, plateau iris configuration, or thick peripheral iris roll. A deep learning architecture, known as comprehensive mechanism decision net (CMD-Net), was developed to simulate the identification of image-level AC mechanisms by human experts. Cross-validation was performed to optimize and evaluate the model. Human-machine comparisons were conducted using a held-out and separate test sets to establish generalizability. Results: In total, 11,035 AS-OCT images of 1455 participants (2833 eyes) were included. Among these, 8828 and 2.207 images were included in the cross-validation and held-out test sets, respectively. A separate test was formed comprising 228 images of 35 consecutive patients with AC detected via gonioscopy at our eye center. In the classification of AC mechanisms, CMD-Net achieved a mean area under the receiver operating characteristic curve (AUC) of 0.980, 0.977, and 0.988 in the cross-validation, held-out, and separate test sets, respectively. The bestperforming ophthalmologist achieved an AUC of 0.903 and 0.891 in the held-out and separate test sets, respectively. And CMD-Net outperformed glaucoma specialists, achieving an accuracy of 89.9 % and 93.0 % compared to 87.0 % and 86.8 % for the best-performing ophthalmologist in the held-out and separate test sets, respectively. Conclusions: Our study suggests that CMD-Net has the potential to classify AC mechanisms using AS-OCT images, though further validation is needed.

引用

页数：13

共 44 条

[1]

[Anonymous], 2010, Preferred practice pattern guidelines: primary open- angle glaucoma

[2] Effectiveness of early lens extraction for the treatment of primary angle-closure glaucoma (EAGLE): a randomised controlled trial [J].