New AI Software can Predict Patterns of Cholera Risk in Humans


A 3D still photo of the cholera bacteria. / Photo by: Getty Images


In 2016, about 132,121 cholera cases and 2420 deaths worldwide have been reported by the World Health Organization. The disease is a major health problem in certain regions that lack access to drinking water and sanitation. To help reduce cholera cases, a team of scientists developed an artificial intelligence that detects patterns of the gut microbiome to calculate cholera risk.

AI-Powered Algorithm Calculates Risk of Cholera

Scientists from Duke University, Massachusetts General Hospital, and the International Center for Diarrheal Disease Research in Bangladesh created a software equipped with an algorithm that analyzes patterns of bacteria in the human gut. The objective of the software is to determine the risk of an individual to cholera infection.

“Our study found that this ‘predictive microbiota’ is as good at predicting who gets ill with cholera as the clinical risk factors that we’ve known about for decades,” stated Dr. Regina C. LaRocque, a senior author of the study from the Massachusetts General.

To create the software, the scientists collected rectal swab samples from 76 residents of Dhaka, Bangladesh. The residents lived in the same household wherein a patient has been hospitalized due to cholera, which increases their risk of contracting the disease. About 33 percent of the residents developed symptoms of cholera, while 66 percent remained uninfected during the follow-up period.

Next, they used a sequencing technology to extract microbe DNA on the swab samples and uploaded the data into a computer. The data were analyzed by the computer and the results were compared to 4,000 different bacterial taxa. Then, the AI looked for distinct patterns in the gut microbiome in those who got sick. The AI linked a set of 100 microbes to the disease based on the 4,000 bacterial taxa and the gut microbiome of cholera patients.

The scientists concluded that the AI can generate a model that could predict cholera better, compared to the system used by infectious disease experts.

“While some people are warning about artificial intelligence leading to killer robots, we are showing the positive impact of AI in its potential to overcome disease,” explained Dr. Lawrence A. David, a senior author of the study at Duke.