There are about 110 million diabetic patients in China, and diabetic retinopathy is one of the most serious complications, with an incidence rate of 24.7%-37.5%. With the advent of an aging society, the incidence of age-related macular degeneration has continued to rise. However, at this stage, the shortage of professional ophthalmologists in China, the lack of medical resources in primary hospitals, and the inefficient diagnosis methods have always restricted the diagnosis of ophthalmology.
The remote intelligent reading platform produced by the combination of artificial intelligence (AI) and ophthalmology will accurately screen the patient's fundus image through intelligent reading, which will effectively assist the doctor's diagnosis and improve the diagnostic efficiency. Especially in areas where doctor resources are relatively scarce, through the efficient telemedicine system, grassroots hospitals are expected to gradually approach the same diagnostic level as large hospitals.
Ophthalmic graded diagnosis becomes possible
Nowadays, the form of eye health is getting more and more serious, and the myopia of teenagers is high. According to data released by the National Health and Family Planning Commission, more than 50% of people do not receive routine eye examinations, and more than 90% of them are treated after the onset of illness. Early intervention can be done for elderly eye diseases and chronic diseases.
As a subsidiary of Aier Ophthalmology, Hunan Extreme Vision Technology Co., Ltd., as the main body of its Internet development strategy, has been serving eye health inspection, medical optometry, eye health products, private eye health doctors, remote medical services, referral services, It has created an online and offline eye health "Internet +" service model.
However, the Chinese ophthalmic medical service industry has the status of “national dispersion and regional concentrationâ€. The development of telemedicine, intelligent equipment, big data and other technologies has also enabled grading diagnosis and treatment in ophthalmology. Intel United Aier Ophthalmology and Polar Vision, Jin Hong Technology launched an artificial intelligence ophthalmic disease identification solution to create a remote intelligent reading platform to further enhance the screening of diabetic retinopathy (DR) and age-related macular degeneration (AMD) Accuracy and efficiency.
Neighboring remote intelligent reading platform
“Looking at the remote intelligent reading platform on the one hand to achieve graded diagnosis and treatment, on the one hand to achieve eye health management.†Xu Ming, general manager of Hunan Polar Vision Internet Technology Co., Ltd. said that doctors at the grassroots level are very lacking, and the reading platform can be centralized. Resources address the fragmented demand for decentralization.
Data collection from community health centers, optometry centers, county-level hospitals, optical shops, etc., through the remote intelligent reading platform for intelligent reading, the abnormal film will be sent to the reader for reading, and finally all The fundus photo is used to accumulate data and continue learning, and the diagnosis results are fed back to the patient and referred.
"The neighboring remote intelligent reading platform will play a role at the grassroots level because the grassroots ophthalmologists are scarce and the diagnostic ability is insufficient." Sun Tao, deputy general manager and CTO of Hunan Extreme Vision Technology Co., Ltd., said that by putting equipment into community health services At the center, the fundus photos can be initially screened by machine self-learning, and the photographs in question are sent to the reading doctor remotely for further judgment.
At present, nearly 20 community service centers and more than 10 Aier Ophthalmology self-built community eye health service centers have begun to use the neighboring remote intelligent reading platform, which is expected to reach 50 by the end of the year. The main pilot cities are concentrated in Changsha and Beijing. Chengdu, Tianjin, Shenzhen.
AI reading takes only 10 seconds
The neighboring remote intelligent reading platform is divided into two major components: software and hardware. The software is artificial intelligence + reading system, and the hardware is the reading center + doctor team. The neighboring remote intelligent reading platform uses a convolutional neural network (CNN), which belongs to a branch of machine learning and implements pattern recognition.
The neighboring remote intelligent reading platform chose DR and AMD as the first screening of artificial intelligence, because the two diseases have the highest prevalence in the ophthalmology industry and feature extraction is relatively convenient. Sun Tao pointed out that the platform has studied the fundus of 100,000 labeled lesions for one and a half years, and the recognition accuracy of DR and AMD has reached 93%. The identification of other diseases is also being developed.
Aier Ophthalmology and Extreme Vision, Intel and Jinhong Technology play different roles respectively. Aier Ophthalmology provides a fundus photo for learning through hundreds of hospitals across the country, and develops and operates a reading platform based on the Internet. Intel has developed the chip for the film, and Jinhong Technology provides the production of fundus camera equipment and artificial intelligence models.
From left to right: Zheng Zhuming, Chairman and Co-founder of Jinhong Technology (Taiwan) Co., Ltd., Zhang Xiaobo, Head of Asia Pacific, Intel Data Center Medical and Life Sciences Group, Xu Ming, General Manager of Hunan Extreme Vision Technology Co., Ltd., Hunan As the deputy general manager of the Internet Technology Co., Ltd., CTO Sun Tao
The neighboring remote intelligent reading platform construction and image storage are all on the third-party cloud platform, and the artificial intelligence calculation is placed in the local data center, and the three are connected through the dedicated network.
Zhang Xiaobo, head of Asia Pacific at Intel Data Center Medical and Life Sciences Group, said that the remote intelligent reading platform is a cloud-to-end practice. Based on the existing AI of Aier Ophthalmology, Intel combines the characteristics of ophthalmic image analysis and processing to provide customized hardware solutions from the perspective of processor and related hardware optimization, including artificial intelligence algorithm framework, Caffe and overall optimization. The calculation platform is improved by 3-5 times and more accurate, and the image judgment time is shortened from 1 minute to less than 10 seconds.
Due to the different qualifications of the reading doctors, some doctors cannot determine the symptoms, and the photos will be submitted to the readers at the next level for help. The platform also developed a track record to allow the primary reviewer to see the final reading.
Regardless of any fundus camera, the remote intelligent reading platform can analyze photos, so the platform is universally popular. The most important promotion problem is the deployment of fundus cameras. Zheng Zhuming, chairman and co-founder of Jinhong Technology (Taiwan) Co., Ltd., said that the application of handheld fundus inspection equipment in the future will greatly accelerate the use of the platform.
Fundus camera
Importance: 1. Bottle should be used within 60 days of first opening. 2. Keep away from light and moisture. 3. Do not leave in the bathroom to avoid moisture. 4. Tightly replace cap after removing strips from the bottle. 5. Do not remove desiccants or touch test areas. 6. Use within expiration date. Reading the results: Negative: reduce carbohydrates intake or increase excercise. Trace/ small/ moderate: your health is satisfactory and you are burning fat. Large: increase carbohydrates or reduce exercise.
Urine Test Strips Cvs,Home Urine Test Strips,Negative Uti Test Strip,Urine Test Strips For Ovulation
Changchun LYZ Technology Co., Ltd , https://www.lyzstrips.com