Consano

• 

• 

• 

  Voice recognition in action

• 

  Wellness score graph

• 

Inspiration

Hospitals are busy spaces; stressful places. When healthcare systems get busy, it becomes difficult for practitioners and professionals (nurses, doctors etc) to spend meaningful time with patients and keep tabs on how patients are doing — how are they feeling? Are there new symptoms? Have existing symptoms gotten worse?

What it does

Consano (latin for "to heal") is a tool that makes it easy for patients to talk about how they feel. A simple, friendly web interface allows patients to start talking with one click. Passages are transcribed into written text before running through an NLP model powered by cohere.ai, which scores the patient's status with sentiment analysis while also extracting key symptom words and terms, like "fever," "headache," "stomach pain" and more.

The information can be viewed in a simple dashboard that graphs the calculated scores and newly detected symptoms.

How we built it

We used a React frontend alongside Node.js, Express and the co:here NLP engine API. Our database of choice was CockroachDB!

Challenges we ran into

Getting things set up; we've never done any sort of ML model training (coming up with sample data); getting our Express APIs to behave

Accomplishments that we're proud of

All of it!

What's next for Consano

Stronger models, separate dashboard for health professionals, a more accessible web interface for visually/hearing impaired patients.

Built With

• chakraui
• co:here
• cockroachdb
• css
• express.js
• html
• javascript
• node.js
• react