ICTNEWS : "WE NEED URGENT IMPLEMENTATION OF SANDBOX MECHANISM IN MEDICAL FIELD TO PROMOTE "MAKE IN VIETNAM" PRODUCT"

There are still many barriers to further expanding the application of DrAid

At the end of August 2021, the AI ​​product "DrAid for Radiology" of VinBrain Company, Vingroup was awarded the ACM SIGAI Industry Award 2021 for the best AI product at the AI ​​Technology Festival within the International General Conference on AI (IJICAI 2021).

According to experts, the ACM SIGAI  award is one of the leading prestigious awards in the field of AI in the world. Each year, the award is given to only one AI product that has been applied in practice and best meets 4 criteria: Uniqueness and novelty of the application; The breakthrough and technical excellence of the application; The importance of AI technology in application; The real value and lasting impact that the app brings to society. The award was given to Microsoft Corporation the previous year.

Mr. Truong Quoc Hung, CEO of VinBrain said that realizing that the DrAid Doctor Assistant product fully meets the 4 criteria of the award, especially the long-term value that the application brings to society, VinBrain decided to apply for the award, with the desire to spread awareness about an AI product 'Make in Vietnam'.

"Besides proving VinBrain's product-making capabilities and pioneering position in AI application in healthcare, the award is also an affirmation of Vietnamese wisdom, putting VinBrain on par with other technology giants and bringing DrAid to the world", said Mr Hung. When attending the award, the company's leadership team put their trust in their products because of the impressive numbers that DrAid for diagnostic imaging has achieved within just 2 years of development such as: The ability to diagnose and screen 21 abnormal and pathological signs on a straight chest X-ray image within 5 seconds with an accuracy of more than 89%.

In addition, the product has been practically deployed in many hospitals with more than 600 physician users, about 450,000 medical images have been uploaded to the application. Before that, the product has also received a number of other recognitions and awards such as the AI ​​model for the best TB screening as recognized by FIT (German non-profit organization working in the field of prevention and control of tuberculosis) or the Second Prize of the Best AI Solution for Health of the International SIIA CODiE Awards - an Oscar in the world computer industry.

The first AI product in Vietnam to support both diagnosis and prognosis of COVID-19 

According to Mr. Hung, the COVID-19 pandemic has raged and caused the whole world to struggle for nearly 2 years, but the pandemic is also an unprecedented "bump" for remote healthcare service. Many barriers have been broken down, new requirements due to the impact of COVID-19 such as contact restrictions, movement restrictions, etc. have created impetus to address infrastructure and regulatory needed to pave the way for telehealth. 

Then, after more than a year of research and development, recently, VinBrain has officially completed DrAid for COVID-19 with the ability to comprehensively solve the COVID-19 problem from diagnostic support to image-based prognosis. The model is trained on a large dataset collected from open data sources around the world and Vietnam. To date, DrAid has the world's largest data source on COVID-19, including 21,421 chest X-ray images of patients positive for SARS-CoV-2 and 118,018 standard chest X-ray images.

Thereby, VinBrain's AI model proactively warns of COVID-19 in all health care situations of people, giving instant results, easy to deploy, and easy to replicate. For example, people who go for regular health checks are assigned to take X-rays, as soon as the photos are uploaded to the system, the AI ​​will immediately warn if it detects signs of suspected Covid-19 infection, thereby helping Hospitals or medical facilities have a plan for timely isolation, limiting infection.

When deployed in F1 concentrated isolation areas, AI helps to stream patients: suspected F1s are immediately localized and single-sample PCR tests are performed, unsuspected cases are aggregated PCR samples and isolated. normal. This both saves costs and helps limit the risk of cross-contamination in the F1 isolation areas. The model is developed based on X-ray images - which is a popular imaging technique for decades, has reasonable prices, and is widely implemented at all medical levels. 

With this solution for diagnosis and treatment of COVID-19 patients, VinBrain hopes to provide the health system with an effective tool to proactively respond to the pandemic, helping CCOVID-19 patients to be identified. early detection and timely treatment and rescue. In addition, realizing the potential of Telehealth in the long-term, VinBrain is focusing on developing the AI ​​SmartCare platform to serve the needs of comprehensive personal health care for all people with different utilities and values ​​through WHO.

In addition, AI SmartCare also helps users to store electronic health records in both times (deep history of medical visits and health information collected via wearable devices) and spatially. (medical record types vary from medical reports, laboratory results, medical imaging types, prescriptions). AI technology will be used by VinBrain to analyze medical data obtained from electronic health records, plus genetic factors and user habits to make prevention or treatment recommendations tailored for users.

Specifically, users can register for health care consultation and medical examination anytime, anywhere on the web or mobile devices, buy drugs online, manage personal health records as well as track the health care route. At the same time, the platform also provides AI assistants to assist Doctors in diagnosing diseases, increasing the efficiency and accuracy of diagnosis.