Researchers have developed a non-invasive sensor that can detect cholera and other diseases. The sensor is placed above a toilet, and is paired with a machine learning algorithm that can classify excretion events. Researchers say it can be deployed to areas where cholera outbreaks are a persistent risk, as well as contaminated disaster zones.
Cholera, a bacterial disease that induces diarrhea, affects millions of people and results in about 150,000 deaths each year. An early detection system for a cholera outbreak would help professionals improve the allocation of resources and aid, but the monitoring of fecal samples is not an easy or non-invasive task.
Now, Maia Gatlin, a research engineer at the Georgia Institute of Technology, has pioneered the “Diarrhea Detector.” The detector itself is a microphone sensor that can be placed over a toilet. The team paired it with a machine learning algorithm that can identify cholera and other bowel diseases—without identifiable personal information.
Gatlin and her team tested the technique on audio data from online sources. Each audio sample of an excretion event was transformed into a spectrogram, which essentially captures the sound in an image. Different events produce different features in the audio and the spectrogram. For example, urination creates a consistent tone, while defecation may have a singular tone. In contrast, diarrhea is more random. Spectrogram images were then fed to a machine learning algorithm that learned to classify each event based on its features.
According to Gatlin’s findings, presented at the 183rd Meeting of the Acoustical Society of America, the AI correctly classified an excretion event as diarrheal or non-diarrheal with 98 percent accuracy if background noise was filtered out, and 96 percent accuracy if background noise was kept in.
"The hope is that this sensor, which is small in footprint and non-invasive in approach, could be deployed to areas where cholera outbreaks are a persistent risk," said Gatlin. "The sensor could also be used in disaster zones where water contamination leads to spread of waterborne pathogens, or even in nursing/hospice care facilities to automatically monitor bowel movements of patients. Perhaps someday, our algorithm can be used with existing in-home smart devices to monitor one's own bowel movements and health."
Next, Gatlin and her colleagues said they will gather real-world acoustic data so their machine learning model can adapt to work in a variety of bathroom environments.