EYAL PERRY

This story starts with an authentic medical case, where an individual who suffers from Crohn's disease had to undergo surgery, post of which a temporary Ileostomy Stoma, a bypass of a section of the bowel, was required. The expected period of hospitalization was 5-7 days.

​

Immediately after surgery, the patient started to suffer from High Output Stoma (HOS). For most patients, the bowel eventually adjusts to its shorter length. In this case, unusual amount of output was consistently measured, which led to an hospitalization period of 6 weeks. During this time, the cause of HOS was unknown and standard treatment procedures failed to stabilize the output.

 

This predicament led the patient to start a rigorous documentation of his daily routine, specifically his nutrition - every food eaten and beverage drank were weighed and recorded. This process allowed the patient to intuitively reduce his excessive output, by altering his daily diet, which directly led to his release from the hospital.

​

In this project, my partner, Dan Lahav and I performed posteriori statistical analysis on the self recorded data generated during the aforementioned period. We cross referenced the food items to known food databases, thus breaking down the daily intake into macro and micro nutrients. We ratified the scientific consensus regarding the influence of fiber on the bowel, and also discovered weaker, but interesting correlations with other micro-nutrients.

​

This project was prepared as part of Statistics course in Tel Aviv University, and we were kindly advised by Tal Galili and Prof. Saharon Rosset

 

As a follow up to this project, Dan and I started to work on an app prototype allowing individuals in similar situations to record their daily life routine, and learn about interesting correlations within their data. At the time, our work didn't mature into a production-ready solution, but a few years later I did continue the same path as part of iCarbonX, and you can now download MEUM ME to daily log your data and get personal data-driven insights.