Companies that aim to build a “poo image database” by recruiting images of poop appear


    When researchers study a particular subject, they collect large amounts of data, but sometimes they widely recruit things that people don’t expect, such as “penis photos.” Seed, a health-related startup, is implementing a project to create a database of feces images by collecting 100,000 “stool images” from people. 

In America, one in five people suffer from chronic bowel diseases such as irritable bowel syndrome, and Seed is such a person by building a fecal image database. we believe that we may be able to develop a system that will help us. Seed’s research team will provide tools to diagnose users’ health from images of stools so that they can manage their health.

“We don’t usually see stools as sources, but stools certainly directly output intestinal health,” Ara Katz, co-founder and CEO of Seed, told The Verge. To help build Seed’s fecal image database, access Seed’s “Stool Image Transmission Page” on your smartphone.

Tap the #GIVEaSHIT button at the bottom of the page.

Then, the submission form will open, so enter your email address and the time zone you normally take a stool into the form. Then, check the check box and tap “TAKE OR UPLOAD PHOTO”, upload as it is if you have already taken a photo, if you have not already taken a stool and then take a photo OK. The photos taken separate the personally identifiable email addresses and potential metadata and are used anonymously by Seed’s research team.

Seed’s research team has seven gastroenterologists who specialize in analyzing stool images sent by the public, and gastroenterologists diagnose and classify all stool images. Classification is carried out according to the criteria of Bristol stool scale, from solid ones such as “hard like nuts”, “sausage type but it is sticky”, to loose things such as “irregular shape and muddy” and “completely liquid”, It is classified into seven types in all. Images of stools tagged by doctors are thus used to train artificial intelligence models. It uses a training system similar to that of self-driving cars to identify trees and cats on the street, allowing artificial intelligence to check the condition of feces just like doctors. The ultimate goal of this fecal database creation project is to open source large numbers of stool images that are usually difficult to obtain and make them useful for academic purposes. David Hachuel, who is working on the database building project, said, “People with chronic bowel disease are struggling with the choice of what to eat and how much exercise to exercise. It’s important to build a database of stool images and develop a monitoring tool that allows patients to monitor their health at home.”


Add a Comment

Your email address will not be published. Required fields are marked *