Recent research shows that IBM's deep learning technology has reached 86% accuracy in the detection and classification of diabetic retinopathy. IBM said the diagnostic tool can help doctors better understand disease progression and determine treatment options. It is understood that diabetic retinopathy, also known as sugar net disease, is one of the complications of diabetes, the incidence is high, and diabetes always follows. In the early stage of the disease, it does not cause strong discomfort to the sick, but gradually erodes the patient's vision like a boiled frog. Therefore, when the patient's response to visual acuity drops significantly, it may be too late, resulting in loss of vision can not be restored. According to statistics, there are currently approximately 415 million diabetic patients worldwide at risk of retinopathy. It is especially important to effectively screen and take timely treatment of patients at risk through technical innovation to prevent blindness. International giants, including Google, are exploring the application of artificial intelligence to the screening of sugar net diseases. Domestic companies such as Shanggong Medical Letter and Peptide Building Blocks are also engaged in research in this area. This time, IBM trained on 35,000 fundus images to identify fundus lesions. It can classify diseases into five levels according to their severity: no diabetic retinopathy, mild, moderate, severe or proliferative diabetic retinopathy. The method combines deep learning, convolutional neural networks, and dictionary-based learning to incorporate disease-specific pathology. According to IBM, the traditional screening method is to monitor patients regularly to diagnose diabetic retinopathy through a clinician. This aspect takes a long time and is highly subjective. The computer vision diagnosis method can complete the diagnosis and classification of diabetic retinopathy within 20 seconds, helping clinicians screen out more patients and improve work efficiency. PTFE Coated Guide Wire,Guide Wire,hydrophilic guide wire,medical guide wire Anesthesia Medical Co., Ltd. , https://www.sinoanesthesia.com