icTus.io

Inspiration

Our icTus concept came together when my Italian professor of Internal Medicine explained what a life-threatening situation patients face because of Infective Endocarditis and Stroke.

What it does

icTus uses computer vision, machine learning, natural language processing, and wearable technology to help stroke patients and medical staff assess outcomes and risk of recurrence.

How we built it

We will be using Google Cloud and TensorFlow and applying machine learning and AI towards a more elegant application of Precision Rehabilitation for patients in ER and Neurology who are faced with complications due to IE & Stroke and managing anti-coagulants and related treatment.

Challenges we ran

We are trying to utilize real patient use cases and will be modeling the first set of algorithms using retrospective data.

Accomplishments that we're proud of

We successfully submitted icTus for Harvard Spaulding Innovation Challenge this January 2018 and are among the top viewed and voted projects.

What we learned

The importance of asking all the right questions in clinic and seeing the hierarchy of icTus SelfCare for patients/caregivers together with MD user requirements in more rapid differential diagnosis: http://box5348.temp.domains/~ictusio/stroke-ie-assessment/