ScholarDeck

AI / informaticsTranslational vision science & technology · 2026

A Deep Learning-Based Study on Automatic Measurement Method of Biological Parameters in Anterior Segment UBM Images of Angle-Closure Glaucoma.

Xinqi Yu, Zhiyuan Zhao, Chenxu Zhang … Sheng Zhou

OphthalmologyAI / informatics

Deep learning automatically measured eye biometric parameters accurately from ultrasound biomicroscopy images.

A Deep Learning-Based Study on Automatic Measurement Method of Biological Parameters in Anterior Segment UBM Images of Angle-Closure Glaucoma.

Xinqi Yu … Sheng Zhou

Translational vision science & technology · 2026

Purpose

This study aimed to develop a deep learning-based model for measuring biometric parameters from anterior segment ultrasound biomicroscopy (ultrasound biomicroscopy) images, assisting clinicians in the early screening and diagnosis of primary angle-closure glaucoma (primary angle-closure glaucoma).

Methods

Through comparative selection, the best-performing model was used to segment four regions in ultrasound biomicroscopy images of primary angle-closure glaucoma patients: cornea and sclera, iris, ciliary body, and lens anterior surface.

Results

0.94

eye measurements from scans matched expert measurements very closely

More results

The DeepLabv3+ model performed excellently in the segmentation tasks, with an mean intersection over union (mIoU) of 85.84%, precision of 92.67%, recall of 91.36%, and Dice coefficient of 92.01%, all representing optimal values.

More results

In the object detection task, the segmented dataset achieved a precision of 88.3%, recall of 89.9%, and mean average precision at 0.50 IoU (mAP50) of 92.9%, showing at least a 19% improvement over the original image dataset.

The average absolute error of the Euclidean distance for four-point localization was 50.41µm, with a root mean square error of 77.85 µm.

“

Conclusion

This study showed that the proposed deep learning-based method for automatic measurement of closure mechanism-related biometric parameters is accurate and effective.

Post

A Deep Learning-Based Study on Automatic Measurement Method of Biological Parameters in Anterior Segment UBM Images of Angle-Closure Glaucoma.

Deep learning automatically measured eye biometric parameters accurately from ultrasound biomicroscopy images.

Related papers