Using Predictive Analytics for Diagnosis of Patients with Irritable Bowel Syndrome

Irritable Bowel Syndrome, or IBS, is a common gastrointestinal disorder affecting the large intestine. It is estimated to affect ten to fifteen percent of the world’s population and 25 to 45 million people in the United States alone. IBS is classified as a chronic condition which is broadly defined by the Center for Disease Control and Prevention (CDC) as a condition which lasts one or more years, requires ongoing medical attention, or limits daily activities.² Symptoms of IBS include cramping, abdominal pain, bloating, constipation, diarrhea, or both.¹ There are no definitive causes for IBS nor is there a definitive way to diagnose patients with IBS. Doctors base their diagnosis on patient-doctor discussions and the elimination of other possible conditions. This long and often tedious process is a reason why this research group sought out to see if any specific conditions within the body were consistent for certain subtypes of IBS, specifically the microbiome of the gut.

The purpose of this study was to see if the use of decision trees could find correlations within the data provided by Cooper Medical School to accurately diagnose patients with IBS based on their gut microbiome. Many studies have shown that the microbiome of the gut plays a key role in patients with IBS and that diet can have a large effect on the gut condition. The data consists of 4 stool samples for 3 different IBS conditions, IBS-C, IBS-D, IBS-M and a control group of patients without IBS, this totaled to 16 samples. Of each sample, values for 48 different tests were received. This data was used to model mock data to increase the amount of data points contained within the decision tree model. The use of a decision tree model determined that six bacteria were found to be most important in determining a patient’s subtype of IBS, Megasphaera, Alistipes onderdonkii, Bacteroides, Ruminococcus, Akkermansia muciniphila, and Barnesiellaceae . This information was then used to develop an online application where the user could input their own values for these bacteria and get a prediction for the subtype of IBS the patient could have. With an accurate prediction for patients, doctors would be able to suggest a specific diet for that patient in order to improve their symptoms.

Sources:

[1] International Foundation for Gastrointestinal Disorders - About Us. Retrieved from https://www.aboutibs.org/what-is-ibs/facts-about-ibs-2.html.

[2] About Chronic Diseases. (2019, October 23). Retrieved from https://www.cdc.gov/chronicdisease/about/index.htm.