mood.ai

Inspiration

Depression is the leading cause of disability in the U.S. for ages 15 to 44. 264 million people worldwide live with depression. The total economic burden of depression is estimated to be $210.5 billion in the U.S. For every dollar of direct costs, an estimated $6.60 is spent on comorbidities, workplace costs, and suicide-related costs. Despite these alarming numbers, not even 50% of people with mental health conditions, including depression, receive treatment. Screening instruments are frequently inaccurate at identifying individuals who have a diagnosis of depression. (Despite the high prevalence of depression in primary care (10% to 12%), the incidence of screening is extremely low, at 2% to 4% of visits.) With the current COVID-19 pandemic, its psychological impacts, and the shift towards telehealth, detecting patients with depression is ever more critical in today’s world. Annual screening of depression with current methods is costly, with more than 190,000$ needed to gain 1 year of life with quality. (190,000$/Quality-adjusted life-year) Teleheath use has increased >80% and there are no suitable tools to detect depression during these visits,

What it does

A screening tool powered by machine learning to automatically detect patients with possible depression during telehealth visits.

How we built it

Challenges we ran into

Accomplishments that we're proud of

What we learned

What's next for mood.ai

References

https://docs.google.com/document/d/1IdvyqoI61CcHBKu_LpGksxaF6VYaljgoc3U8F6OGg_c/edit?usp=sharing

Built With

• amazon-web-services
• pyaudio
• python
• pytorch
• spacy