AI based system to diagnose diseases

New Delhi: Researchers at the Indian Institute of Technology (IIT) here have developed an artificial intelligence (AI)-based low-power electronic hardware system that can help in detecting malaria, tuberculosis, intestinal parasite and cervical cancer in a few milliseconds.

The research focuses on building a neuromorphic system which can be used for healthcare access in resource-constrained areas with limited access to human specialists. "While several software AI models exist for healthcare and diagnostic related applications, need of the hour is to efficiently map these models on portable dedicated low-power, low-cost hardware to enable edge-AI systems accessible to all in low resource environment," said Professor Manan Suri, Department of Electrical Engineering, IIT Delhi.

Suri, who is leading the team of researchers, said microscopy is particularly well adapted to low-resource, high disease burden areas, being both simple and versatile; even for diagnostic tasks. While newer technologies are available for diagnosis, the cost of specialised equipment may render it impractical in such places, he said.

"In contrast to alternatives such as rapid diagnostic tests, however, microscopy-based diagnosis does depend on the availability of skilled technicians, of which there is a critical shortage. As a result, diagnoses are often made on the basis of clinical signs and symptoms alone, which is error-prone and leads to higher mortality, drug resistance, and the economic burden of buying unnecessary drugs," the professor said. "There is therefore need for alternatives, which can help in providing the access to quality diagnosis that is currently routinely unavailable," he added.