# @seed on Instagram

- **Type:** Image
- **Original URL:** https://www.instagram.com/p/CLCjAtNpMqT
- **Gondola URL:** https://gondola.cc/posts/34247018-seed-instagram
- **Thumbnail:** https://img.gondola.cc/tr:w-,h-,fo-auto/postThumbnails/c5c86625c1.jpg
- **Posted:** 2021-02-08T17:35:13.000+00:00
- **Account Owner:** Seed (@seed) — https://gondola.cc/seed

## Caption

The latest from Seed: our new AI 🤖 

Today we announced our acquisition of auggihealth, a digital health company with a suite of artificial intelligence and machine learning technologies that track and analyze digestive health¹—this includes an AI algorithm for real-time Bristol stool typing and a mobile monitoring application for clinical research. 

Auggi’s deep learning algorithms utilize computer vision and deep convolutional neural networks to objectively quantify and characterize an individual’s stool (💩) with 94.07% accuracy (measured in the ability to predict the right Bristol class within one Bristol level) as compared to 75% accuracy¹ from self-assessment.² 🤯

And thanks to many of you and our 2019 Community Science initiative, #GiveAShit, the AI was trained by the largest bank of stool imagery assembled to date in collaboration with gastroenterologists at Massachusetts General Hospital. 

So what are we going to do with it? 

The clinical trial monitoring app is currently deployed in our ongoing Phase II randomized, triple-blind, and placebo-controlled clinical trial to investigate the role of the gut microbiome in patients with Irritable Bowel Syndrome at Beth Israel Deaconess Medical Center. 

Link in bio to read more. 🤓

And what else? 

We can’t fully tell you 🤭 (yet)...but here’s a hint: 💩+ 🤖+ 📸+📲

—
¹ Blake, M.R., Raker, J.M. and Whelan, K. (2016), Validity and reliability of the Bristol Stool Form Scale in healthy adults and patients with diarrhoea‐predominant irritable bowel syndrome. Aliment Pharmacol Ther, 44: 693-703. https://doi.org/10.1111/apt.13746
² The Bristol stool chart, first published in the Scandinavian Journal of Gastroenterology in 1997, classifies human stool into seven categories and is a globally-recognized diagnostic tool. However, traditional methods of tracking and characterizing stool are subjective, inconsistent, burdensome, and hampered by poor adherence.

## Stats

- **Views:** 0
- **Likes:** 1,708
- **Shares:** 0
- **Comments:** 97

## Tags

giveashit

---
Copyright (c) Gondola