Classifying Retinal Degeneration in Histological Sections Using Deep Learning

Abstract

Purpose Artificial intelligence (AI) techniques are increasingly being used to classify retinal diseases. In this study we investigated the ability of a convolutional neural network (CNN) in categorizing histological images into different classes of retinal degeneration. Methods Images were obtained from a chemically induced feline model of monocular retinal dystrophy and split into training and testing sets. The training set was graded for the level of retinal degeneration and used to train various CNN architectures. The testing set was evaluated through the best architecture and graded by six observers. Comparisons between model and observer classifications, and interobserver variability were measured. Finally, the effects of using less training images or images containing half the presentable context were investigated. Results The best model gave weighted-F1 scores in the range 85% to 90%. Cohen kappa scores reached up to 0.86, indicating high agreement between the model and observers. Interobserver variability was consistent with the model-observer variability in the model's ability to match predictions with the observers. Image context restriction resulted in model performance reduction by up to 6% and at least one training set size resulted in a model performance reduction of 10% compared to the original size. Conclusions Detecting the presence and severity of up to three classes of retinal degeneration in histological data can be reliably achieved with a deep learning classifier. Translational Relevance This work lays the foundations for future AI models which could aid in the evaluation of more intricate changes occurring in retinal degeneration, particularly in other types of clinically derived image data.

Development, Validation, and Comparison of 2 Ultrasound Feature–Guided Machine Learning Models to Distinguish Cervical Lymphadenopathy

“…To measure the efficiency of the 2 proposed models, 5 performance measures were used: AUC, accuracy value, precision value, recall value, and the F1 score. 19 With our decision tree, the importance analysis of variables revealed that grid_like echo, echogenic_hilum, vascular_pattern, age, and short axis were the top 5 variables, indicating that clinicians should primarily focus on these lymph nodes features. These variables are consistent with the results of previous studies.…”

“…In consideration that elasticity imaging and contrast-enhanced ultrasonography are not always available in clinical settings, we decided to avoid using these methods in the present study. To measure the efficiency of the 2 proposed models, 5 performance measures were used: AUC, accuracy value, precision value, recall value, and the F1 score 19 …”

Artificial Intelligence: A Review of Progress and Prospects in Medicine and Healthcare

“…The most path breaking development from the researchers at Google health [88] is the detection of metastatic breast cancer by analyzing the images from the lymph node tissue performing better than human pathologists. Mouiee et al, [89] explore the usage of CNN for categorizing retinal degeneration into three different classes using retinal histological images. Pantanowitz et al, [90] utilize deep learning methods to analyze prostate core needle biopsies to aid in prostate cancer diagnosis.…”