AI that can listen for CANCER in farts is developed by scientists… who had to listen to hours of toilet audio tapes
- Research is in early stages but has the potential to prevent thousands of deaths
- Diarrheal diseases such as cholera as well as cancer are targets of the device
- A sensor would record sounds and AI technology could potentially diagnose
Scientists have developed a device they hope can diagnose cancer by listening to your bathroom sounds.
It works using artificial intelligence to look for subtle changes in the sound when someone defecates, urinates, or flatulates.
Researchers built a database with hours of audio and video samples of excretions from healthy and unwell patients to establish a baseline to prime the machine-learning algorithm.
David Ancalle, the lead researcher from Georgia Tech University who helped make the device said: ‘We’re trying to find a non-invasive way where people can get a notification on whether or not they should go get checked out. Like “Hey, your urine is not flowing at the rate that it should. Your farts are not sounding the way they should. You should check it out.”’
It comes as experts explore novel ways to diagnose cancer without invasive biopsies. Japanese experts are using tiny worms to sniff out pancreatic tumors — which are notoriously hard to catch and deadly.
The prototype of the sensor package includes a microphone through which bathroom sounds could be recorded. Different sounds produced through urination, flatulence, solid defecation, and diarrhea are all influenced by the pathways in your body and can indicate something is wrong
Cooling down brain tumor cells could lead to better cancer survival rates, a promising study suggests.
Changes in the tracts like the urethra and the rectum can be caused by diseases such as colorectal cancer, and sound waves can reflect that.
The engineers assembled a mechanical device equipped with pumps, nozzles, and tubes that recreate the physics behind each bathroom sound, which they determined by sifting through publicly available audio and video samples of excretions.
After analyzing the sound waves and frequency spectrum for each, the team fed the information into a machine-learning algorithm. Their AI technology was able to learn from the mountain of collected sound data.
Armed with the analyses of sound waves from different bathroom noises, the researchers primed the mechanical device that, fittingly, is called Synthetic Human Acoustic Reproduction Testing. Yes, SHART.
They pumped SHART with water, recreating those sound waves.
Their algorithm identified the source of the sounds – urination, flatulence, solid defecation, and diarrhea – 98 per cent of the time.
The ultimate goal is to employ an inexpensive sensor to pick up the sounds and feed them into the AI technology.
Maia Gatlin, an aerospace engineer at the Georgia Tech Research Institute who worked on the device said: ‘A lot of thought went into each of the sounds. There was a subsystem for each sound on this little machine.’
The SHART machine works by mimicking the physics behind different bathroom sounds using its plethora of nozzles and tubes. Researchers pump it with water and sounds that mirror the real thing are produced
How will the device work?
A sensor will pick up the sound of bathroom excretions – pooping, peeing, etc.
Researchers behind the device have explored putting the sensors in toilets.
The Synthetic Human Acoustic Reproduction Testing machine is now preparing an AI algorithm to one day pick up on deadly diseases using sound waves.
The machine recreates The researchers pump the SHART machine with water and it is able to recreate the physics behind the sounds that are picked up by the sensor.
The distinct sounds result from changes in the tracts like the urethra and the rectum, which can be caused by diseases such as colorectal cancer
The resulting sound waves are fed into an AI device for analysis
The machine-learning algorithm, which has been primed with publicly availble samples of bathroom sounds, could potentially spot discrepancies in sounds pointing to health issues.
The device would have a huge impact globally in spotting outbreaks early and preventing them from spreading further. Cholera, for instance, is a diarrheal disease that the device would be able to detect, potentially averting thousands of deaths.
‘And as we classify those events, we can start to collect that data. It can say, “Hey we’re seeing an outbreak of lots of diarrhea.’ Then we can start to quickly diagnose what’s going on in an area,”’ Ms Gatlin said.
An estimated 1.3 to four million cases of cholera are diagnosed worldwide each year and the disease kills between 20,000 and 143,000 annually.
Diarrheal diseases like cholera are an especially acute problem in parts of the world where people are more likely to be malnourished and deprived of quality, accessible healthcare, such as in Sub-Saharan Africa and rural parts of South Asia. This class of diseases is the third leading cause of child mortality globally, falling just behind pneumonia and preterm birth complications.
Research findings from the Georgia Tech team, presented last week at the American Physical Society’s Annual Meeting of the Division of Fluid Dynamics, represent just the latest example of sewage materials being used for the good of public health.
Wastewater surveillance proved an important tool for monitoring Covid-19 surges days before cases crop up, as genetic material from the virus can be detected in fecal matter.
Mr Ancalle stressed that his team is embarking on a mission to find an affordable and widely accessible sensor device so that the technology can be scaled up to cover wider populations.
He said: ‘We’re not trying to come up with million-dollar equipment. We are trying to make this something that can be afforded by just everyone, particularly since the project is focused on urban areas with weak health systems. The affordability aspect is very important for us.’
Source: Read Full Article