(Hong Kong, Shanghai, 26 August 2020) Ping An Insurance (Group) Company of China, Ltd. (hereafter “Ping An” or the “Group”, HKEx:2318; SSE:601318) is pleased to announce that Ping An Smart Healthcare, the Group’s healthcare arm, has developed an automated system with artificial intelligence (AI) that can accurately identify diseases of the retina and make urgent referrals for specialist consultation. The effectiveness of the system is shown in a study published in the August issue of Translational Vision Science & Technology (TVST)[1], a respected peer-reviewed international ophthalmology journal.

Diseases such as diabetic retinopathy, age-related macular degeneration, and pathologic myopia, are major causes of visual impairment and blindness, particularly among elderly populations. Although most diseases can be cured through early diagnosis and treatment, many retinal pathologic cases are not identified in time due to the lack of ophthalmology specialists in some communities and rural areas.

Ping An’s intelligent system uses feature pyramid network-based deep learning, a type of machine learning, to automatically identify retinal disease and make urgent referrals to specialists. To test the system, the study, published in TVST, obtained 28,664 optical coherence tomography (OCT) images from 2,254 patients at the Eye and Ear, Nose and Throat Hospital of Fudan University (EENT Hospital) and the Shanghai Tenth People’s Hospital (TENTH Hospital). The system detected 15 categories of retinal pathologies from OCT images as common signs of various retinal diseases. In a second step, the pathologies detected in the OCT images and measurements that indicated changes in retinal thickness were combined and categorized using a random forest decision making tool. Cases were classified as urgent referral (within one week); routine (within three months) and observation (normal). 

The result showed that the system produced significant results (see table).

Since the beginning of 2020, the model has been used to screen more than 13,000 patients in China via AskBob Doctor, Ping An’s AI-based diagnosis and treatment assistant tool. Nearly 4,000 patients were diagnosed with eye diseases of varying degrees, and 1,300 required immediate treatment to prevent vision loss. 

Dr. Xie Guotong, Ping An Group Chief Healthcare Scientist, said, “Early diagnosis and treatment help to prevent vision loss. However, a shortage of eye care resources, especially in rural areas, reduces the possibility of a timely response. We developed this novel intelligence system in the hope we could help improve the capability of primary medical institutions and offer a solution to communities that lack ophthalmic resources. The accuracy and efficiency of the AI model was achieved with the support of the expert team led by EENT and TENTH hospitals and their high-quality ophthalmic data. The study shows that this intelligent system has a great value and practicality in communities that are shortage of ophthalmologists.”

Introducing new technologies to healthcare to serve online and offline patients, providers and payers, is an integral part of Ping An’s “finance + ecosystem” strategy. As of July 2020, Ping An Smart Healthcare supported a network of nearly 17,000 medical institutions across 90 cities in China. AskBob Doctor has been used 33 million times by 430,000 doctors for medical decision support, patient follow-up and patient education since its launch in June 2019.