Evaluation of an offline, artificial intelligence system for referable glaucoma screening using a smartphone-based fundus camera: a prospective study.

BACKGROUND/OBJECTIVES: An affordable and scalable screening model is critical for undetected glaucoma. The study evaluated the performance of an offline, smartphone-based AI system for the detection of referable glaucoma against two benchmarks: specialist diagnosis following full glaucoma workup and consensus image grading. SUBJECTS/METHODS: This prospective study (tertiary glaucoma centre, India) included 243 subjects with varying severity of glaucoma and control group without glaucoma. Disc-centred images were captured using a validated smartphone-based fundus camera analysed by the AI system and graded by specialists. Diagnostic ability of the AI in detecting referable Glaucoma (Confirmed glaucoma) and no referable Glaucoma (Suspects and No glaucoma) when compared to a final diagnosis (comprehensive glaucoma workup) and majority grading (image grading) by Glaucoma specialists (pre-defined criteria) were evaluated. RESULTS: The AI system demonstrated a sensitivity and specificity of 93.7% (95% CI: 87.6-96.9%) and 85.6% (95% CI:78.6-90.6%), respectively, in the detection of referable glaucoma when compared against final diagnosis following full glaucoma workup. True negative rate in definite non-glaucoma cases was 94.7% (95% CI: 87.2-97.9%). Amongst the false negatives were 4 early and 3 moderate glaucoma. When the same set of images provided to the AI was also provided to the specialists for image grading, specialists detected 60% (67/111) of true glaucoma cases versus a detection rate of 94% (104/111) by the AI. CONCLUSION: The AI tool showed robust performance when compared against a stringent benchmark. It had modest over-referral of normal subjects despite being challenged with fundus images alone. The next step involves a population-level assessment.

Development and validation of an artificial intelligence based screening tool for detection of retinopathy of prematurity in a South Indian population.

Purpose: The primary objective of this study was to develop and validate an AI algorithm as a screening tool for the detection of retinopathy of prematurity (ROP). Participants: Images were collected from infants enrolled in the KIDROP tele-ROP screening program. Methods: We developed a deep learning (DL) algorithm with 227,326 wide-field images from multiple camera systems obtained from the KIDROP tele-ROP screening program in India over an 11-year period. 37,477 temporal retina images were utilized with the dataset split into train (n = 25,982, 69.33%), validation (n = 4,006, 10.69%), and an independent test set (n = 7,489, 19.98%). The algorithm consists of a binary classifier that distinguishes between the presence of ROP (Stages 1-3) and the absence of ROP. The image labels were retrieved from the daily registers of the tele-ROP program. They consist of per-eye diagnoses provided by trained ROP graders based on all images captured during the screening session. Infants requiring treatment and a proportion of those not requiring urgent referral had an additional confirmatory diagnosis from an ROP specialist. Results: Of the 7,489 temporal images analyzed in the test set, 2,249 (30.0%) images showed the presence of ROP. The sensitivity and specificity to detect ROP was 91.46% (95% CI: 90.23%-92.59%) and 91.22% (95% CI: 90.42%-91.97%), respectively, while the positive predictive value (PPV) was 81.72% (95% CI: 80.37%-83.00%), negative predictive value (NPV) was 96.14% (95% CI: 95.60%-96.61%) and the AUROC was 0.970. Conclusion: The novel ROP screening algorithm demonstrated high sensitivity and specificity in detecting the presence of ROP. A prospective clinical validation in a real-world tele-ROP platform is under consideration. It has the potential to lower the number of screening sessions required to be conducted by a specialist for a high-risk preterm infant thus significantly improving workflow efficiency.

Diagnostic Accuracy of Automated Diabetic Retinopathy Image Assessment Softwares: IDx-DR and Medios Artificial Intelligence.

INTRODUCTION: Numerous studies have demonstrated the use of artificial intelligence (AI) for early detection of referable diabetic retinopathy (RDR). A direct comparison of these multiple automated diabetic retinopathy (DR) image assessment softwares (ARIAs) is, however, challenging. We retrospectively compared the performance of two modern ARIAs, IDx-DR and Medios AI. METHODS: In this retrospective-comparative study, retinal images with sufficient image quality were run on both ARIAs. They were captured in 811 consecutive patients with diabetes visiting diabetic clinics in Poland. For each patient, four non-mydriatic images, 45° field of view, i.e., two sets of one optic disc and one macula-centered image using Topcon NW400 were captured. Images were manually graded for severity of DR as no DR, any DR (mild non-proliferative diabetic retinopathy [NPDR] or more severe disease), RDR (moderate NPDR or more severe disease and/or clinically significant diabetic macular edema [CSDME]), or sight-threatening DR (severe NPDR or more severe disease and/or CSDME) by certified graders. The ARIA output was compared to manual consensus image grading (reference standard). RESULTS: On 807 patients, based on consensus grading, there was no evidence of DR in 543 patients (67%). Any DR was seen in 264 (33%) patients, of which 174 (22%) were RDR and 41 (5%) were sight-threatening DR. The sensitivity of detecting RDR against reference standard grading was 95% (95% CI: 91, 98%) and the specificity was 80% (95% CI: 77, 83%) for Medios AI. They were 99% (95% CI: 96, 100%) and 68% (95% CI: 64, 72%) for IDx-DR, respectively. CONCLUSION: Both the ARIAs achieved satisfactory accuracy, with few false negatives. Although false-positive results generate additional costs and workload, missed cases raise the most concern whenever automated screening is debated.

Agreement of a Novel Artificial Intelligence Software With Optical Coherence Tomography and Manual Grading of the Optic Disc in Glaucoma.

PRCIS: The offline artificial intelligence (AI) on a smartphone-based fundus camera shows good agreement and correlation with the vertical cup-to-disc ratio (vCDR) from the spectral-domain optical coherence tomography (SD-OCT) and manual grading by experts. PURPOSE: The purpose of this study is to assess the agreement of vCDR measured by a new AI software from optic disc images obtained using a validated smartphone-based imaging device, with SD-OCT vCDR measurements, and manual grading by experts on a stereoscopic fundus camera. METHODS: In a prospective, cross-sectional study, participants above 18 years (Glaucoma and normal) underwent a dilated fundus evaluation, followed by optic disc imaging including a 42-degree monoscopic disc-centered image (Remidio NM-FOP-10), a 30-degree stereoscopic disc-centered image (Kowa nonmyd WX-3D desktop fundus camera), and disc analysis (Cirrus SD-OCT). Remidio FOP images were analyzed for vCDR using the new AI software, and Kowa stereoscopic images were manually graded by 3 fellowship-trained glaucoma specialists. RESULTS: We included 473 eyes of 244 participants. The vCDR values from the new AI software showed strong agreement with SD-OCT measurements [95% limits of agreement (LoA)=-0.13 to 0.16]. The agreement with SD-OCT was marginally better in eyes with higher vCDR (95% LoA=-0.15 to 0.12 for vCDR>0.8). Interclass correlation coefficient was 0.90 (95% CI, 0.88-0.91). The vCDR values from AI software showed a good correlation with the manual segmentation by experts (interclass correlation coefficient=0.89, 95% CI, 0.87-0.91) on stereoscopic images (95% LoA=-0.18 to 0.11) with agreement better for eyes with vCDR>0.8 (LoA=-0.12 to 0.08). CONCLUSIONS: The new AI software vCDR measurements had an excellent agreement and correlation with the SD-OCT and manual grading. The ability of the Medios AI to work offline, without requiring cloud-based inferencing, is an added advantage.

Towards a Device Agnostic AI for Diabetic Retinopathy Screening: An External Validation Study.

Purpose: To evaluate the performance of a validated Artificial Intelligence (AI) algorithm developed for a smartphone-based camera on images captured using a standard desktop fundus camera to screen for diabetic retinopathy (DR). Participants: Subjects with established diabetes mellitus. Methods: Images captured on a desktop fundus camera (Topcon TRC-50DX, Japan) for a previous study with 135 consecutive patients (233 eyes) with established diabetes mellitus, with or without DR were analysed by the AI algorithm. The performance of the AI algorithm to detect any DR, referable DR (RDR Ie, worse than mild non proliferative diabetic retinopathy (NPDR) and/or diabetic macular edema (DME)) and sight-threatening DR (STDR Ie, severe NPDR or worse and/or DME) were assessed based on comparisons against both image-based consensus grades by two fellowship trained vitreo-retina specialists and clinical examination. Results: The sensitivity was 98.3% (95% CI 96%, 100%) and the specificity 83.7% (95% CI 73%, 94%) for RDR against image grading. The specificity for RDR decreased to 65.2% (95% CI 53.7%, 76.6%) and the sensitivity marginally increased to 100% (95% CI 100%, 100%) when compared against clinical examination. The sensitivity for detection of any DR when compared against image-based consensus grading and clinical exam were both 97.6% (95% CI 95%, 100%). The specificity for any DR detection was 90.9% (95% CI 82.3%, 99.4%) as compared against image grading and 88.9% (95% CI 79.7%, 98.1%) on clinical exam. The sensitivity for STDR was 99.0% (95% CI 96%, 100%) against image grading and 100% (95% CI 100%, 100%) as compared against clinical exam. Conclusion: The AI algorithm could screen for RDR and any DR with robust performance on images captured on a desktop fundus camera when compared to image grading, despite being previously optimized for a smartphone-based camera.

Design and Performance Characterization of a Novel, Smartphone-Based, Portable Digital Slit Lamp for Anterior Segment Screening Using Telemedicine.

Purpose: Telemedicine-enabled, portable digital slit lamps can help to decentralize screening to close-to-patient contexts. We report a novel design for a portable, digital slit lamp using a smartphone. It works on an advanced optical design and has the capability of instantaneous, objective photodocumentation to capture anterior segment images and is telemedicine-enabled. Methods: The device is constructed keeping its usability and the importance of design ergonomics for nonspecialized field personnel in mind. The optical design is described, and the resolution and magnification are compared with traditional desktop-based slit lamps. A Health Insurance Portability and Accountability Act (HIPAA)-compliant, patient management software is integrated to synchronize the captured images with a secure cloud server along with a sharpness algorithm to extract the best focused frames of the cornea, iris, and lens, from videos. We demonstrate its photodocumentation ability and teleophthalmology feasibility by capturing images in a pilot study from nine subjects. Results: Images were obtained in various illumination, magnification, and filter settings. Synchronous and asynchronous teleophthalmology consults were conducted. The performance of the device was shown to be limited by the smartphone sensor resolution and not the optical design, because the Air Force target resolution was found to be the same on smartphone-mounted traditional slit lamps despite a lower magnification. Conclusions: The novel, portable, digital slit lamp with advanced optical design using smartphones has the ability to screen for anterior segment pathologies using telemedicine. Translational Relevance: A portable, telemedicine-friendly, ergonomically designed, slit lamp used by nonspecialist personnel allows for both synchronous and asynchronous modes of consultation at remote locations, facilitating mass screening programs.