Union education minister Dharmendra Pradhan on October 15 announced the setting up of three Artificial Intelligence-Centres for Excellence (AI-COE) focused on — healthcare, agriculture, and sustainable cities — which will be led by IIT Delhi – AIIMS Delhi, IIT-Ropar and IIT-Kanpur
From early detection of various types of cancer to diagnosing vector-borne diseases, the Indian Institute of Technology (IIT)-Delhi will work to develop AI-based technology solutions to improve healthcare in its Artificial Intelligence-Centre for Excellence (AI-COE) announced by the union Education ministry. The institute will be partnering with the All India Institute of Medical Sciences (AIIMS) Delhi to create such deployable technologies and platforms.
Union education minister Dharmendra Pradhan on October 15 announced the setting up of three AI-COEs focused on — healthcare, agriculture, and sustainable cities — which will be led by IIT Delhi – AIIMS Delhi, IIT-Ropar and IIT-Kanpur.
While IIT-Delhi has been collaborating with AIIMS Delhi to develop technology for various medical applications for the past several years, this will be the first time when the two top-rated institutes will work on developing AI solutions for various ailments identified in the national health programmes.
Over the next one month, the two institutes will decide where the centre will be located.
RS 330 CRORE MOE FUNDING
According to Prof Chetan Arora, chief project manager, AI-COE on Healthcare, IIT-Delhi, the COE has been provided a funding of Rs 330 crore by the MoE, which will eventually turn into a self-sustaining revenue model.
“We will be developing AI algorithms for various healthcare problems such as cancer detection, chronic disease management, infectious diseases etc. closely working with clinicians. It means developing a platform and AI models that can screen for various diseases when a person goes for a regular ultrasound, say for example, due to abdominal pain. This is for early detection and management where during regular screenings, diseases such as cancer can be detected at an early stage, allowing appropriate remedial measures in time,” said Prof Arora, a faculty member in the Computer Science and Engineering Department, which is jointly associated with the School of AI, at IIT Delhi.
The project has a timeline of four years by the end of which teams from the two institutes will have to develop the platform and various AI models and deploy them at clinics and hospitals across the country.
“The idea is to develop one platform that can screen for multiple diseases including various kinds of cancer, vector-borne diseases as well as analyse the health parameters of mother and child, among others, when placed in healthcare facilities. This is intended for early prevention where the AI models can actively assist the doctors improve their effectiveness and reduce their load, allowing them to focus on treatment,” he said.
UPSCALING AND UPSKILLING
The plan is to create an AI platform that does both — upscaling (to scale up services) and upskilling (improving the effectiveness). As these kinds of screenings are currently expensive in India, the single platform for the same will also bring in affordability.
Currently, a lot of generative AI applications are being developed world over, and India may have an edge in specialised AI algorithm development as it has a lot of data of its own, over which novel AI algorithms can be developed and fine-tuned.
The IIT is already working in its lab in these areas. The IIT Delhi and AIIMS Delhi team submitted a Proof of Concepts (POC) in four areas to the ministry to be selected for the COE in phase-2. These include breast cancer detection from mammograms; breast cancer risk prediction; detection of diabetic retinopathy; and analyses of chest X-ray.
“We have been working in our own lab for the past five years on early detection of breast cancer, and gall-bladder cancer. We have taken a lot of data from AIIMS and PGIMER Chandigarh,” he added.
IIT-Delhi and AIIMS Delhi are in consortium with other technological institutions such as IIT-Ropar and IIT-Mandi, medical institutions such as PGIMER Chandigarh and Cachar hospital in Assam, as well as various industry partners. “The plans also include inviting proposals from these as well as other academic and medical institutes, including industry partners, on other emergent areas for research within the scope of the CoE,” said Prof Parag Singla, Head, School of AI, IIT-Delhi.
Over the four years, the Central Project Monitoring unit for all AI-COEs, appointed by the ministry, will be evaluating the work being done across the three verticals every three to six months, he added.
The AI-COE came up as part of the Centre’s vision to ‘Make AI in India and Make AI work for India’. The establishment of these centres was announced during the budget for 2023-24. In alignment with this, the union government has approved the creation of the three AI-COEs, with a total financial outlay of Rs 990 crore over the period of 2023-24 to 2027-28.
