Lab-on-an-App: AI Empowered Point-of-Care Diagnostics for Ageing Population

Lead Research Organisation: University College London
Department Name: Electronic and Electrical Engineering

Abstract

Mobile health technology - encompassing mobile sensor, computation, communication, and user-interface capability - has the potential to support healthcare of an increasingly ageing population. It offers the opportunity to perform clinical diagnoses on the patient side, contributing to a more sustainable demand for healthcare systems in developed nations but also more widespread use of healthcare in developing ones. The overarching challenge relates to the development of high-accuracy, low-cost, portable mobile health applications capable of diagnosing a range of conditions prevalent in the older population.

One such condition relates to anaemia that nowadays afflicts circa 20% of the older population. This condition is both under-recognised and under-treated; leads to additional morbidities ranging from organ damage (heart, lung), immune system disorders, or fatigue in turn contributing to falls (hence bone fractures); and contributes to the financial burden of healthcare systems. Adults with some types of anaemias were considered as a vulnerable patient group requiring shielding due to their high risk of severe SARS-CoV-2 infection during the COVID-19 pandemic.

The diagnosis of anaemia requires laboratory-based measurements of a venous blood sample but this is not always readily accessible to a large fraction of the older population, preventing timely interventions. Therefore, with an increasing number of diagnoses reported each year, there is also a demand for easily accessible portable diagnosic tools.

This project will develop a Lab-on-App to non-invasively diagnose anaemia and its causes (e.g. genetics, diet, or injury) that can be easily used by older people, carers, or healthcare professionals. It involves the development of:

(1) Sensor technology capable of extracting information / images from the body or body fluids including (i) electrochemical sensors to measure concentration of urea on urine or (ii) multi-spectral sensors to measure skin / body fluids appearance

(2) Machine learning technology that delivers diagnoses of anaemia and its causes given the data collected by the aforementioned sensors.

(3) An android / ios application offering users an interface to collect data, analyse data, and deliver the diagnostics.

Publications

10 25 50

Studentship Projects

Project Reference Relationship Related To Start End Student Name
EP/R513143/1 30/09/2018 29/09/2023
2580492 Studentship EP/R513143/1 30/09/2021 29/09/2025 Zhuo Zhi
EP/T517793/1 30/09/2020 29/09/2025
2580492 Studentship EP/T517793/1 30/09/2021 29/09/2025 Zhuo Zhi