Vingroup developed disease diagnosis software through X-ray images

Diagnosis of diseases through medical images (such as X-ray images, cut-offs, magnetic resonance …) is usually done manually by imaging diagnostics. Doctors can sometimes miss small details but are important because of normal eye restrictions, pressure on time, number of patients, as well as work intensity … thereby affecting accuracy .

In order to overcome this problem, the Medical Picture Analysis team of VinTech Company (Vingroup) has put artificial intelligence (AI) into image diagnosis. Initially, this technology has been successfully applied in the diagnosis of chest X-ray images, which are also popular diagnostics and have the greatest demand today (accounting for 70-80%).

Using AI to diagnose diseases through X-ray images (photo: NVIDIA)

According to Dr. Nguyen Quy Ha – Head of Health Picture Analysis (Big Data Institute), if based on the evaluation criteria of the health sector, both the sensitivity and specificity of the algorithm will reach over 90%, even over 95% for some diseases. If based on speed criteria, if it is necessary to read a radiograph, the doctor will need 5-10 minutes, then the reader will only take 2-5 seconds.

Although initial positive results have been achieved, algorithms are being built based on foreign open source data. If it is possible to collect medical image data of Vietnamese people, the accuracy and practical applicability in Vietnam will be really high. The current difficulty is that medical data collection cannot be done on day one or two. For each type of disease, the algorithm needs to take photos from about 100,000 to 200,000 patients, a small number.

To solve the problem of input data, the Big Data Institute signed a Memorandum of Understanding to cooperate with the Information Technology Department – Ministry of Health, unifying coordination in developing regulations on data exchange. medical examination and treatment of people among medical examination and treatment establishments and organizations and individuals serving the research and development of artificial intelligence application products in Health.

Big Data Institute cooperates with Ministry of Health to exchange examination data

"Data after being standardized and fully annotated will not carry any personal information and will be shared widely to the community. When the software is complete, hospitals can be used for free. The doctors will receive maximum support, minimize the time to read photos, in lower hospitals, where there are not many experts, the application of AI will significantly increase the opportunity for patients to have access to Modern diagnostic methods, help doctors have more useful information to decide therapy, " said Prof. Vu Ha Van – Director of Big Data Institute.

Upon completion of the product, a feasible option is to deploy a remote diagnostics system using cloud computing technology. The user will be granted an account and just sending the image will have the readout result immediately after. This way of doing is beneficial is low cost and hospitals do not need to invest in private infrastructure.

Prof. Vu Ha Van also shared, to put the software into use, there are many things to do. In addition to collecting data of Vietnamese people, there are legal issues and licensing procedures. Our aim is that the software must meet international standards and become useful to domestic hospitals. We will aim for the necessary diagnoses for a number of dangerous and common diseases in Vietnam such as cancers or cardiovascular, neurological and diabetic diseases.