Sorry, you need to enable JavaScript to visit this website.

Detecting a melange of diseases in milliseconds

Read time: 2 mins
Detecting a melange of diseases in milliseconds | IIT Delhi

Tuberculosis, a deadly bacterial disease, killed half a million people in India in 2016. An estimated 16,700 people died of another potentially fatal disease called malaria, spread by mosquitoes, in 2018. Cervical cancer, a common type of cancer in women, leads to about 74,000 deaths annually in the country. Intestinal parasites, including those that cause diarrhoea are another leading cause of death. What’s common among all these diseases? Most of the deaths could be prevented if diagnosed early enough. However, the lack of timely, qualified diagnostics comes in the way.

Now, researchers at the Indian Institute of Technology Delhi (IIT Delhi) have developed a cost-effective, low power consuming hardware device to detect these disease using Artificial Intelligence and Deep Learning. These are computer-based machine learning techniques that “learn” by looking at existing data. Many innovative applications in healthcare use these techniques.

“While many software AI models exist for healthcare and diagnostic-related applications, the 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”, says Prof Manan Suri, from  IIT Delhi, who led the research.

The researchers have built a hardware system that runs on Raspberry Pi, a tiny computer, which is attached to an infrared-sensitive camera. Unlike heavy laboratory equipment, it is portable and requires minimum power. The hardware system is programmed with AI-based algorithms that can detect microorganisms in the biological samples. The device can diagnose malaria, tuberculosis, intestinal parasites and cervical cancer with high accuracy and can be used in resource-strapped environments like a primary healthcare centre.

In the absence of quality diagnostics, these diseases are mostly treated based on their symptoms, which can be error-prone, leading to higher mortality, drug resistance, and the economic burden of buying unnecessary drugs. Hence, a quicker, accurate diagnostic solution augurs well in such scenarios and could save thousands of lives.

A proof-of-concept of the device has been presented at international healthcare conferences in the US and Italy. It had also received the prestigious Gandhian Young Technology Innovation Award (GYTI) in 2018. The researchers hope that their platform will be deployed soon in rural and resource-constrained areas and improve access to diagnostic healthcare.

This article is based on a press release from IIT Delhi.