AI is rapidly impacting how we identify diseases and health issues. Recently, researchers from the Sardar Vallabhbhai National Institute of Technology in Surat, India used AI technology to determine if someone had a cold or not by analyzing vocal patterns.
Led by electronic engineer Pankaj Warule, the research team collected 630 human voices, 111 of them belonged to individuals who had a cold. By testing the regular pattern of harmonics that diminishes in amplitude as the frequency increases, the researchers were able to find out the difference between a healthy voice and the voice of someone with a cold. The subjects were asked to count from one to 40 before describing what they did at the weekend and recite Aesop’s fable “The North Wind and the Sun”. With the help of machine-learning algorithms, it was possible for the team to detect cold successfully with 70% accuracy.
The team believes that their technology could radically change medical diagnosis by eliminating the need for expensive and time-consuming visits to the doctor. It can also play a role in managing workplace absenteeism. Americans don’t take many days off even if they are unwell, and over one quarter of private sector employees don’t have access to paid sick leave. An AI-based system could help business owners pinpoint people who are genuinely taking a leave and to potential fakers.
The results of this research are one of the latest examples of AI’s increasingly prominent role in the medical world. Last year, a research group successfully used AI to detect Parkinson’s disease by closely observing patients’ breathing patterns. AI is also helping researchers detect depression and multiple forms of cancer with the analyzed vocal patterns.
The research team, Pankaj Warule and his team, headed up the study that shows the potential of AI to help identify the health conditions without requiring time consuming and costly doctor’s visits. The study demonstrates the power and versatility of AI to analyze various aspects of an individual’s health and provide accurate results in real time. Warule, an electronic engineer, includes the use of machine learning algorithms to detect differences in amplification. As such, this study opens the door to numerous opportunities to use AI to detect pre-existing conditions before they become serious. In addition, the technology can give businesses the option of using AI to monitor their employees’ vocal health and remove any uncertainty about their well-being.