Use of explainable AI on slit-lamp images of anterior surface of eyes to diagnose allergic conjunctival diseases.

BACKGROUND: Artificial intelligence (AI) is a promising new technology that has the potential of diagnosing allergic conjunctival diseases (ACDs). However, its development is slowed by the absence of a tailored image database and explainable AI models. Thus, the purpose of this study was to develop an explainable AI model that can not only diagnose ACDs but also present the basis for the diagnosis. METHODS: A dataset of 4942 slit-lamp images from 10 ophthalmological institutions across Japan were used as the image database. A sequential pipeline of segmentation AI was constructed to identify 12 clinical findings in 1038 images of seasonal and perennial allergic conjunctivitis (AC), atopic keratoconjunctivitis (AKC), vernal keratoconjunctivitis (VKC), giant papillary conjunctivitis (GPC), and normal subjects. The performance of the pipeline was evaluated by determining its ability to obtain explainable results through the extraction of the findings. Its diagnostic accuracy was determined for 4 severity-based diagnosis classification of AC, AKC/VKC, GPC, and normal. RESULTS: Segmentation AI pipeline efficiently extracted crucial ACD indicators including conjunctival hyperemia, giant papillae, and shield ulcer, and offered interpretable insights. The AI pipeline diagnosis had a high diagnostic accuracy of 86.2%, and that of the board-certified ophthalmologists was 60.0%. The pipeline had a high classification performance, and the area under the curve (AUC) was 0.959 for AC, 0.905 for normal subjects, 0.847 for GPC, 0.829 for VKC, and 0.790 for AKC. CONCLUSIONS: An explainable AI model created by a comprehensive image database can be used for diagnosing ACDs with high degree of accuracy.

Diagnosability of Keratoconus Using Deep Learning With Placido Disk-Based Corneal Topography.

Purpose: Placido disk-based corneal topography is still most commonly used in daily practice. This study was aimed to evaluate the diagnosability of keratoconus using deep learning of a color-coded map with Placido disk-based corneal topography. Methods: We retrospectively examined 179 keratoconic eyes [Grade 1 (54 eyes), 2 (52 eyes), 3 (23 eyes), and 4 (50 eyes), according to the Amsler-Krumeich classification], and 170 age-matched healthy eyes, with good quality images of corneal topography measured with a Placido disk corneal topographer (TMS-4TM, Tomey). Using deep learning of a color-coded map, we evaluated the diagnostic accuracy, sensitivity, and specificity, for keratoconus screening and staging tests, in these eyes. Results: Deep learning of color-coded maps exhibited an accuracy of 0.966 (sensitivity 0.988, specificity 0.944) in discriminating keratoconus from normal eyes. It also exhibited an accuracy of 0.785 (0.911 for Grade 1, 0.868 for Grade 2, 0.920 for Grade 3, and 0.905 for Grade 4) in classifying the stage. The area under the curve value was 0.997, 0.955, 0.899, 0.888, and 0.943 as Grade 0 (normal) to 4 grading tests, respectively. Conclusions: Deep learning using color-coded maps with conventional corneal topography effectively distinguishes between keratoconus and normal eyes and classifies the grade of the disease, indicating that this will become an aid for enhancing the diagnosis and staging ability of keratoconus in a clinical setting.

Determination of probability of causative pathogen in infectious keratitis using deep learning algorithm of slit-lamp images.

Corneal opacities are important causes of blindness, and their major etiology is infectious keratitis. Slit-lamp examinations are commonly used to determine the causative pathogen; however, their diagnostic accuracy is low even for experienced ophthalmologists. To characterize the "face" of an infected cornea, we have adapted a deep learning architecture used for facial recognition and applied it to determine a probability score for a specific pathogen causing keratitis. To record the diverse features and mitigate the uncertainty, batches of probability scores of 4 serial images taken from many angles or fluorescence staining were learned for score and decision level fusion using a gradient boosting decision tree. A total of 4306 slit-lamp images including 312 images obtained by internet publications on keratitis by bacteria, fungi, acanthamoeba, and herpes simplex virus (HSV) were studied. The created algorithm had a high overall accuracy of diagnosis, e.g., the accuracy/area under the curve for acanthamoeba was 97.9%/0.995, bacteria was 90.7%/0.963, fungi was 95.0%/0.975, and HSV was 92.3%/0.946, by group K-fold validation, and it was robust to even the low resolution web images. We suggest that our hybrid deep learning-based algorithm be used as a simple and accurate method for computer-assisted diagnosis of infectious keratitis.

Prediction of keratoconus progression using deep learning of anterior segment optical coherence tomography maps.

BACKGROUND: To predict keratoconus progression using deep learning of the color-coded maps measured with a swept-source anterior segment optical coherence tomography (As-OCT) device. METHODS: We enrolled 218 keratoconic eyes with and without disease progression. Using deep learning of the 6 color-coded maps (anterior elevation, anterior curvature, posterior elevation, posterior curvature, total refractive power, and pachymetry map) obtained by the As-OCT (CASIA, Tomey), we assessed the accuracy, sensitivity, and specificity of prediction of keratoconus progression in such eyes. RESULTS: Deep learning of the 6 color-coded maps exhibited an accuracy of 0.794 in discriminating keratoconus with and without progression. For a single map analysis, posterior elevation map (0.798) showed the highest accuracy, followed by anterior curvature map (0.775), posterior corneal curvature map (0.757), anterior elevation map (0.752), total refractive power map (0.729), and pachymetry map (0.720), in distinguishing between progressive and non-progressive keratoconus. The use of the adjusted algorithm by age subgroups improved to an accuracy of 0.849. CONCLUSIONS: Deep learning of the As-OCT color-coded maps effectively discriminates progressive keratoconus from non-progressive keratoconus with an accuracy of approximately 85% using the adjusted age algorithm, indicating that it will become an aid for predicting the progression of the disease, which is clinically beneficial for decision-making of the surgical indication of corneal cross-linking (CXL).

Automated Detection of Macular Diseases by Optical Coherence Tomography and Artificial Intelligence Machine Learning of Optical Coherence Tomography Images.

PURPOSE: Although optical coherence tomography (OCT) is essential for ophthalmologists, reading of findings requires expertise. The purpose of this study is to test deep learning with image augmentation for automated detection of chorioretinal diseases. METHODS: A retina specialist diagnosed 1,200 OCT images. The diagnoses involved normal eyes (n=570) and those with wet age-related macular degeneration (AMD) (n=136), diabetic retinopathy (DR) (n=104), epiretinal membranes (ERMs) (n=90), and another 19 diseases. Among them, 1,100 images were used for deep learning training, augmented to 59,400 by horizontal flipping, rotation, and translation. The remaining 100 images were used to evaluate the trained convolutional neural network (CNN) model. RESULTS: Automated disease detection showed that the first candidate disease corresponded to the doctor's decision in 83 (83%) images and the second candidate disease in seven (7%) images. The precision and recall of the CNN model were 0.85 and 0.97 for normal eyes, 1.00 and 0.77 for wet AMD, 0.78 and 1.00 for DR, and 0.75 and 0.75 for ERMs, respectively. Some of rare diseases such as Vogt-Koyanagi-Harada disease were correctly detected by image augmentation in the CNN training. CONCLUSION: Automated detection of macular diseases from OCT images might be feasible using the CNN model. Image augmentation might be effective to compensate for a small image number for training.

Keratoconus detection using deep learning of colour-coded maps with anterior segment optical coherence tomography: a diagnostic accuracy study.

OBJECTIVE: To evaluate the diagnostic accuracy of keratoconus using deep learning of the colour-coded maps measured with the swept-source anterior segment optical coherence tomography (AS-OCT). DESIGN: A diagnostic accuracy study. SETTING: A single-centre study. PARTICIPANTS: A total of 304 keratoconic eyes (grade 1 (108 eyes), 2 (75 eyes), 3 (42 eyes) and 4 (79 eyes)) according to the Amsler-Krumeich classification, and 239 age-matched healthy eyes. MAIN OUTCOME MEASURES: The diagnostic accuracy of keratoconus using deep learning of six colour-coded maps (anterior elevation, anterior curvature, posterior elevation, posterior curvature, total refractive power and pachymetry map). RESULTS: Deep learning of the arithmetical mean output data of these six maps showed an accuracy of 0.991 in discriminating between normal and keratoconic eyes. For single map analysis, posterior elevation map (0.993) showed the highest accuracy, followed by posterior curvature map (0.991), anterior elevation map (0.983), corneal pachymetry map (0.982), total refractive power map (0.978) and anterior curvature map (0.976), in discriminating between normal and keratoconic eyes. This deep learning also showed an accuracy of 0.874 in classifying the stage of the disease. Posterior curvature map (0.869) showed the highest accuracy, followed by corneal pachymetry map (0.845), anterior curvature map (0.836), total refractive power map (0.836), posterior elevation map (0.829) and anterior elevation map (0.820), in classifying the stage. CONCLUSIONS: Deep learning using the colour-coded maps obtained by the AS-OCT effectively discriminates keratoconus from normal corneas, and furthermore classifies the grade of the disease. It is suggested that this will become an aid for improving the diagnostic accuracy of keratoconus in daily practice. CLINICAL TRIAL REGISTRATION NUMBER: 000034587.