Application of Artificial Intelligence in Precision Medicine for Diabetic Macular Edema.

Diabetic macular edema (DME) is the primary cause of central vision impairment in patients with diabetes and the leading cause of preventable blindness in working-age people. With the advent of optical coherence tomography and antivascular endothelial growth factor (anti-VEGF) therapy, the diagnosis, evaluation, and treatment of DME were greatly revolutionized in the last decade. However, there is tremendous heterogeneity among DME patients, and 30%-50% of DME patients do not respond well to anti-VEGF agents. In addition, there is no evidence-based and universally accepted administration regimen. The identification of DME patients not responding to anti-VEGF agents and the determination of the optimal administration interval are the 2 major challenges of DME, which are difficult to achieve with the coarse granularity of conventional health care modality. Therefore, more and more retina specialists have pointed out the necessity of introducing precision medicine into the management of DME and have conducted related studies in recent years. One of the most frontier methods is the targeted extraction of individualized disease features from optical coherence tomography images based on artificial intelligence technology, which provides precise evaluation and risk classification of DME. This review aims to provide an overview of the progress of artificial intelligence-enabled precision medicine in automated screening, precise evaluation, prognosis prediction, and follow-up monitoring of DME. Further, the challenges ahead of real-world applications and the future development of precision medicine in DME will be discussed.

EE-Explorer: A Multimodal Artificial Intelligence System for Eye Emergency Triage and Primary Diagnosis.

PURPOSE: To develop a multimodal artificial intelligence (AI) system, EE-Explorer, to triage eye emergencies and assist in primary diagnosis using metadata and ocular images. DESIGN: A diagnostic, cross-sectional, validity and reliability study. METHODS: EE-Explorer consists of 2 models. The triage model was developed from metadata (events, symptoms, and medical history) and ocular surface images via smartphones from 2038 patients presenting to Zhongshan Ophthalmic Center (ZOC) to output 3 classifications: urgent, semiurgent, and nonurgent. The primary diagnostic model was developed from the paired metadata and slitlamp images of 2405 patients from ZOC. Both models were externally tested on 103 participants from 4 other hospitals. A pilot test was conducted in Guangzhou to evaluate the hierarchical referral service pattern assisted by EE-Explorer for unspecialized health care facilities. RESULTS: A high overall accuracy, as indicated by an area under the receiver operating characteristic curve (AUC) of 0.982 (95% CI, 0.966-0.998), was obtained using the triage model, which outperformed the triage nurses (P < .001). In the primary diagnostic model, the diagnostic classification accuracy (CA) and Hamming loss (HL) in the internal testing were 0.808 (95% CI 0.776-0.840) and 0.016 (95% CI 0.006-0.026), respectively. In the external testing, model performance was robust for both triage (average AUC, 0.988, 95% CI 0.967-1.000) and primary diagnosis (CA, 0.718, 95% CI 0.644-0.792; and HL, 0.023, 95% CI 0.000-0.048). In the pilot test in the hierarchical referral settings, EE-explorer demonstrated consistently robust performance and broad participant acceptance. CONCLUSION: The EE-Explorer system showed robust performance in both triage and primary diagnosis for ophthalmic emergency patients. EE-Explorer can provide patients with acute ophthalmic symptoms access to remote self-triage and assist in primary diagnosis in unspecialized health care facilities to achieve rapid and effective treatment strategies.

A digital mask to safeguard patient privacy.

The storage of facial images in medical records poses privacy risks due to the sensitive nature of the personal biometric information that can be extracted from such images. To minimize these risks, we developed a new technology, called the digital mask (DM), which is based on three-dimensional reconstruction and deep-learning algorithms to irreversibly erase identifiable features, while retaining disease-relevant features needed for diagnosis. In a prospective clinical study to evaluate the technology for diagnosis of ocular conditions, we found very high diagnostic consistency between the use of original and reconstructed facial videos (κ ≥ 0.845 for strabismus, ptosis and nystagmus, and κ = 0.801 for thyroid-associated orbitopathy) and comparable diagnostic accuracy (P ≥ 0.131 for all ocular conditions tested) was observed. Identity removal validation using multiple-choice questions showed that compared to image cropping, the DM could much more effectively remove identity attributes from facial images. We further confirmed the ability of the DM to evade recognition systems using artificial intelligence-powered re-identification algorithms. Moreover, use of the DM increased the willingness of patients with ocular conditions to provide their facial images as health information during medical treatment. These results indicate the potential of the DM algorithm to protect the privacy of patients' facial images in an era of rapid adoption of digital health technologies.

EyeHealer: A large-scale anterior eye segment dataset with eye structure and lesion annotations.

Anterior segment eye diseases account for a significant proportion of presentations to eye clinics worldwide, including diseases associated with corneal pathologies, anterior chamber abnormalities (e.g. blood or inflammation), and lens diseases. The construction of an automatic tool for segmentation of anterior segment eye lesions would greatly improve the efficiency of clinical care. With research on artificial intelligence progressing in recent years, deep learning models have shown their superiority in image classification and segmentation. The training and evaluation of deep learning models should be based on a large amount of data annotated with expertise; however, such data are relatively scarce in the domain of medicine. Herein, the authors developed a new medical image annotation system, called EyeHealer. It is a large-scale anterior eye segment dataset with both eye structures and lesions annotated at the pixel level. Comprehensive experiments were conducted to verify its performance in disease classification and eye lesion segmentation. The results showed that semantic segmentation models outperformed medical segmentation models. This paper describes the establishment of the system for automated classification and segmentation tasks. The dataset will be made publicly available to encourage future research in this area.

Deep-learning models for the detection and incidence prediction of chronic kidney disease and type 2 diabetes from retinal fundus images.

Regular screening for the early detection of common chronic diseases might benefit from the use of deep-learning approaches, particularly in resource-poor or remote settings. Here we show that deep-learning models can be used to identify chronic kidney disease and type 2 diabetes solely from fundus images or in combination with clinical metadata (age, sex, height, weight, body-mass index and blood pressure) with areas under the receiver operating characteristic curve of 0.85-0.93. The models were trained and validated with a total of 115,344 retinal fundus photographs from 57,672 patients and can also be used to predict estimated glomerulal filtration rates and blood-glucose levels, with mean absolute errors of 11.1-13.4 ml min-1 per 1.73 m2 and 0.65-1.1 mmol l-1, and to stratify patients according to disease-progression risk. We evaluated the generalizability of the models for the identification of chronic kidney disease and type 2 diabetes with population-based external validation cohorts and via a prospective study with fundus images captured with smartphones, and assessed the feasibility of predicting disease progression in a longitudinal cohort.