“For the First Time Ever, Artificial Intelligence is Applied in Healthcare to Identify Medication Error”

Applying Artificial Intelligence (AI) to help detect off-prescription drugs, VAIPE – a system developed by the research team at VinUni-Illinois Smart Health Center and partners – has enjoyed a great reception and won the AI ​​Awards 2022. The solution makes use of advanced technologies in the fields of artificial intelligence and data science, ensuring low cost as well as easy and mass accessibility in order to improve public health.

Let’s find out VAIPE’s features:

1️⃣ DETECTING MEDICATION ERROR/ OFF-PRESCRIPTION DRUGS

According to Dr. Pham Huy Hieu – co-leader of the project, this is the first time new AI algorithms were used to allow accurate identification of drugs from images and detect off-prescription drugs. This function of automatic drug identification via pill images will help users understand medication information (including uses and dosage) as well as possible side effects. This feature is expected to help monitor medication habits to avoid errors.

In particular, the AI ​​algorithm developed by the team uses graph neural networks and contrastive learning to warn users of medication errors when detecting off-prescription drugs. “The accuracy of these algorithms is completely superior to existing traditional methods,” said Dr. Hieu.

2️⃣ HELPING DOCTORS ACCESS PATIENTS’ HEALTH INDICATORS IN REAL TIME AND MAKE NECESSARY RECOMMENDATIONS

The system automatically extracts and recognizes information from prescription images, thereby helping to manage medication intake. With this function, users can access personal health analysis in a timely manner. Doctors can also access patients’ health indicators in real time and make necessary recommendations.

3️⃣ DETECTING CARDIAC IRREGULARITIES THROUGH DIGITALIZATION OF ELECTROCARDIOGRAM SIGNALS

The system automatically classifies medical measuring tools and identifies health indicators. Through digitizing electrocardiogram (ECG) signals from images, the system helps detect cardiac irregularities. At the same time, the system has an integrated learning platform that helps protect users’ personal data.

4️⃣ EARLY WARNING AND DISEASE MONITORING

The research team has built an integrated AI-based mobile platform that allows centralized data collection that can build health records and support treatment of diseases. Indicators such as medication information, heart rate, blood pressure, etc., are collected in real time, forming a large database for the development of AI algorithms for early warning and disease monitoring.

VAIPE’s feature of detecting medication error/ off-prescription drugs

This rich database will be open to the scientific community and form the foundation for policy-making and development of public health services, thereby benefiting the Vietnamese people.

“Creating new, low-cost, easy-to-use and accessible technologies for the average user is the goal of this project,” said Dr. Hieu. He added that smart healthcare technology only makes sense when people from all walks of life can access it easily and benefit from it directly.

VinUni-Illinois Smart Health Center is jointly established by VinUniversity and the University of Illinois Urbana-Champaign (UIUC) – USA, focusing on research into biotechnology and data science. This project received 13.5 million USD from Vingroup to work towards the goal of researching for positive impacts on public health at low cost and easy accessibility.