NEW DELHI: The researchers at IIT-Delhi have designed an “Artificial Intelligence (AI)-based low-power electronic hardware system” to detect four different diseases including cervical cancer in milliseconds, according to a statement released on Tuesday.
According to World Health Organisation, cervical cancer is the fourth most frequent cancer in women with an estimated 570,000 new cases in 2018 representing 6.6 per cent of all female cancers.
The institute said that this work is focused on building an intelligent Neuromorphic system which can be used for healthcare access in resource-constrained areas with limited access to human specialists.
The researchers have demonstrated a Proof-of-Concept (PoC) low-power rapid AI hardware implementation-based microscopy diagnostic support system for four different diseases -- malaria, tuberculosis, cervical cancer and intestinal parasite infection.
Malaria is a life-threatening mosquito-borne blood disease and nearly half of the world’s population is at risk of malaria. Tuberculosis (TB) is one of the top ten causes of death worldwide. The rapid screening of TB is possible, but the service accessibility is still poor in rural areas and requires specialized equipment that is not readily available.
Intestinal parasites infect the gastrointestinal tract of humans and have a consistent external and internal morphology throughout the different stages of development that is egg, larva and adult stages.
The research team led by Professor Manan Suri, Department of Electrical Engineering, has four student researchers working on this project, Khushal Sethi, Narayani Bhatia, Vivek and Shridu Verma, who were awarded two Summer Undergraduate Research Awards, in 2017 and 2018.
“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 cutting edge-AI systems accessible to all in low resource environment,” the professor said.
“The approach demonstrated by the researchers is portable, low-power and can classify with high accuracy in detection of the diseases. The long-term impact and goal of this work will be to enable potential future deployment of the platform in rural and resource-constrained areas and improve the access to diagnostic health-care,” he added.