Personalized Diabetes, hypertension, hyperLipidimia (DHL) Counselling Tool using AI (PERDICT.AI)
Project Reference :
National University of Singapore (NUS), Singapore General Hospital (SGH)
Principal Investigator :
Prof Wynne Hsu
Technology Readiness :
5 (Technology validated in relevant environment)
Technology Categories :
AI- Deep Learning
At present, there is no standard patient counselling tool for diabetes in Singapore. Physicians typically counsel patients verbally without any visual aids or tools. The reason for this is information is available in piecemeal form and is not easily accessible. Furthermore, the information in these materials are mostly from western population studies that may not be relevant to the Asian population.
PERDICT.AI is an evidence-based personalised integrated solution to support clinical management of DHL patients and improve disease outcomes. It employs data analytics and artificial intelligence in considering different sources of information from clinical practice guidelines, electronic health records (EHR), including demographics, vital sign, laboratory test results, prescribed medications and their corresponding dosages, and external drug interaction repositories.
PERDICT.AI has four modules:
- Patient similarity algorithm for comparison with peers to help patients visualise the relative severity of their disease.
- Risk prognostication based on patient similarity analytics to illustrate and explain complication risks to the patient to increase medication adherence and motivate diet and lifestyle changes.
- Medication recommender system that suggests a list of medications, given patient-specific historical information.
- Personalized care planning taking into consideration clinical practice guidelines, target weight and biomarker values.
PERDICT.AI can be used by physicians to facilitate diabetes-related education, counselling and shared decision-making with patients, resulting in improved patient enablement and medication adherence.
PERDICT.AI’s patient similarity algorithm is the first to consider medication dosage information together with diverse types of variables in the electronic health records.
PERDICT.AI’s medication recommender algorithm outperforms state-of-the-art methods while achieving the best trade-off between accuracy and drug-to-drug interaction.
PERDICT.AI has the potential to be developed into an application that could help primary healthcare teams stop or slow disease progression and complication development in patients with diabetes, hypertension and hyperlipidemia through education, counselling and shared decision-making.
We welcome interest from the industry for collaboration/ co-development / customisation of the technology into a new product or service. If you have any enquiries or are keen to collaborate, please contact us.