Google trains its AI to detect sickness symptoms through sound

With growing technology, artificial intelligence has opened gates for multiple possibilities. AI has been a boom in every segment of the industry. And now, to take it even further, Google is training its AI to detect sickness symptoms through voice. According to a Bloomberg report, the tech giants are using 300 million audio samples, which include the sounds of coughing, sneezing, and laboured breathing, to detect symptoms of diseases like tuberculosis that generally go undetected.

Last week, Google partnered with Salcit Technologies, a respiratory healthcare AI startup based in India, which will help them integrate this technology into smartphones. This partnership is working on bioacoustics (a combination of biology and acoustics that studies sounds produced by animals and humans) and a healthcare model that will help in detecting disease from sounds. Previously, Google has worked in a similar venture to detect disease from scent.

This attempt by Google can be a breakthrough in the medical industry, making diagnosis easier and more accessible. This AI model, known as HeAR (Health Acoustic Representations), is revolutionary in its approach. The technology can be accessed through a smartphone, which will easily provide diagnosis in remote areas where big machinery like X-rays is not available.

Countering Tuberculosis

According to the World Health Organisation’s data, tuberculosis claims 4,500 lives and infects 30,000 individuals daily. While the disease is treatable, it generally goes undiagnosed due to a lack of awareness and availability. Through this revolutionary technology, the diagnosis will be much easier.

Reportedly, Google has trained its AI, referring to almost 300 million audio clips gathered from various corners of the world. These cough, sneeze, and breathing sounds are sourced from YouTube and recordings from TB screenings in hospitals.

This smartphone blend technology will use a microphone and AI software to detect the disease. The mobile technology can be taken to the remotest area to carry out the diagnosis. With the ability to study the cough pattern, the AI model can detect TB in its early phases. However, its accuracy cannot be judged right now as it is still in the developing phase.

Partnership with Salcit Technologies

Through its partnership with Salcit Technologies, Google aims to level up the accuracy of tuberculosis diagnosis and lung health assessments. According to reports, the India-based startup will incorporate Google’s AI model with its machine learning technology called Swaasa—an AI system named after the Sanskrit word for breath. Their association will bring a revolution in the field of respiratory health monitoring and diagnosis.

Through the advancement of this HeAR model, we can expect it to expand and cover multiple other diseases related to respiratory illness and even cardiovascular diseases.

This revolutionary AI model will open possibilities of early diagnosis and early treatment even in the most remote corner of the world. This will enhance the accessibility of diagnosis using the mobile model, making healthcare available in both rural and urban setups.

You might also be interested in – MK Stalin visits Apple, Google, Microsoft offices in US, Inks investment deals