Help others by advancing science: we’re building the world’s first AI computer vision model for stool. 11% of the global population suffers from Irritable Bowel Syndrome (that’s over 800 million people) and the symptoms can be debilitating. For the average patient, 73 days per year are socially or professionally restricted. At Dieta, we’re building technology to help people improve their symptoms rapidly using data and AI. This computer vision model is a critical piece of that system.
Keep track of your poops: if you’re one of those people with IBS, or are just generally trying to improve your digestive health, the Dieta app is the easiest way to keep track of your bowel patterns. You can also track your diet in our app by taking photos of foods (we can already label these for you with AI) and our technology will identify patterns about what triggers or relieves your digestive symptoms.
Get insights about yourself: Once we collect enough images to train our AI models, we’ll run those models on all your poop pics. As a result, you’ll get access to these insights. We’ll let you know what your stool suggests about digestive health and what you can learn from others who have similar digestive systems to yours!
Why are we doing this?
Dieta’s mission is to end digestive suffering. Our way of contributing is by using data and AI to give people the best recommendations to quickly improve their digestive symptoms. To do that, we need to collect data on the inputs and outputs of digestive health for each individual. The inputs are food, drinks, medications, and supplements; the things we put in our bodies and the things we can fully control. The outputs are the results of what we put in our body: bowel movements and digestive symptoms. These are the things we can’t always fully control, but we strive to optimize them with our data-driven recommendations.
We believe the future of measuring digestive outputs is the stool computer vision model we’re building. Today, gastroenterologists around the world use the Bristol Stool Scale to measure discrete 1-7 values that characterize bowel movements. This outdated method is problematic because it is subjective and provides low resolution data that isn’t very useful (we’ll discuss in our next blog post). Dieta’s computer vision model will produce high-resolution data that is objective and can be comparable across thousands (and one day millions) of patients.
With the high resolution data we already collect in the Dieta app today, we can build a “bowel movement profile” for each person. These measure 6 characteristics at different times of the day:
Now, instead of just measuring a single 1-7 value, we have hundreds of features on each individual’s digestive health, allowing us to gain deeper insight and provide more personalized recommendations.
How To Do It
Take a clear image of the stool in the toilet
Log the 2 most important characteristics using the sliders, these are consistency and completeness
For how to navigate the app, check out the following instructional gif:
Wrapping up
Dieta’s CEO, Asaf, has already photographed 430 of his own poops to contribute to the database (CTO Ben is at 150 works-of-art, his bowels aren’t as crazy as Asaf’s). We hope you’ll have as much fun as we’ve had taking poop pics!
Questions? Comments? We love hearing from you, email us at hello@bigdieta.com
