Share the post "Is it possible to diagnose Parkinson’s or COVID-19 based on the user’s voice?"
It is difficult to detect Parkinson’s disease or COVID-19 in the early stages of those illnesses. Therefore there is a solution to that problem. An app called Aum detects both of those diseases in the early stages.
Dinesh Kumar, a professor at the Royal Melbourne Institute of Technology, and his colleagues conducted an experiment to discover whether the subtleties in a person’s voice could be detected by machine-learning algorithms. Therefore they invited 36 people with Parkinson’s disease and 36 without it.
The participants had to say different phenomes, which required sounds from the throat (/a/), the mouth (/o/), and the nose (/m/). The researchers recorded that with the IOS system, and then they developed and applied the algorithms which could differentiate people with Parkinson’s disease and without it. In IEEE Access, the researchers reported that their algorithm could identify people with Parkinson’s disease with 100 per cent accuracy. Kumar also said that they could differentiate between people with Parkinson’s disease who take medication and who do not.
The researchers then applied a different machine-learning algorithm to the previous one. It turned out that the features extracted from the vowel /i/ during the first three days after admittance to the hospital were the most effective at differentiating between people with a COVID-19 infection of the lungs and healthy people. The algorithm was accurate at 94 per cent.
In my opinion, Aum is beneficial for us. With that app, we can detect in the early stages illnesses like Parkinson’s or COVID-19. Therefore we can prepare for the future by buying assistive devices for a person with Parkinson’s disease or adapting our house to that person. Moreover, those people can start taking medicines from the very beginning of their disease, which may help them to slow down the effects of that illness. When it comes to COVID-19, taking medication from the beginning of that illness can make them recover. However, if the app got a diagnosis wrong, it could end badly for that person. It could cause unnecessary expenses, stress or breakdowns leading to mental health problems. I think that we should not believe the app in 100 per cent, and the researchers should continue to develop it.
Thanks for your time. Feel free to comment below 🙂
References:
It is an interesting innovation, but I think it is also only in the early stages of development and it has a long way to go, nonetheless, it will affect the healthcare system significantly.
In my opinion, the idea of diagnosing someone with a disease without physically seeing them is really smart and will make it a lot easier for people all around the world who maybe can’t afford to do a Covid 19 or Parkinson test. But on the other hand it will decrease employment, and also a noticeable error will occur when people are using the app and recording (for example wind might affect the accuracy of delivering this sound). So this project can be really helpful but when used correctly in a closed area with no external sounds or factors that might affect the recording.
Your blog absolutely shocked and intrigued me; I had never heard of such method before. Early stages of illnesses are particularly important for people, so it would be fantastic if this app could detect whether you are ill or not. However, as you pointed out, we should be careful while using it because a poor diagnosis could lead to serious issues. I think that there is still much work to be done for this system to be 100% safe and accurate.
This is the first time I’ve heard about such an application. And I can’t say that this is something reliable for me. Maybe I would try it just for the sake of interest, or if someone from my family got sick, I would check how it works. In general, there is a telegram bot that works roughly on the same principle, it determines whether you have a mole on your body, and I actually used this application, and once when he wrote me a suspicious result, I went to the doctor, but he said that I was fine. It’s interesting that people are trying to combine technology and health, but I think these developments should also include doctors, experienced specialists in their field.