Automatic Anterior Chamber Angle Measurement for Ultrasound Biomicroscopy using Deep Learning.

Researchers

Journal

Modalities

Models

Abstract

To develop a software package for automated measuring of the trabecular-iris angle (TIA) using ultrasound biomicroscopy (UBM).
UBM images were collected and the TIA was manually measured by specialists. Different models were used as the convolutional neural network (CNN) for the automatic TIA measurement. The root-mean-squared error (RMSE), explained variance (EVA), and mean absolute percentage error (MAPE) were used to evaluate the performance of these models. The interobserver reproducibility, coefficient of variation (CV), and intraclass correlation coefficient (ICC) were calculated to evaluate the consistency between the manual measured and the model predicted values.
ResNet-18 had the best performance in RMSE, EVA, and MAPE among all five models. The average difference between the angles measured manually and by the model is -0.46±3.97° for all eyes, -1.67±5.19° for open angles, and 0.75±1.43° for narrow angles. The CV, ICC, and reproducibility of the total TIA measurements are 6.8%, 0.95, and 6.1° for all angles, 6.4%, 0.99, and 7.7° for open angles, and 8.8%, 0.93, and 4° for narrow angles, respectively.
Preliminary results show that this fully automated anterior chamber angle measurement method can achieve high accuracy and have good consistency with the manual measurement results, this has great significance for future clinical practice.