.png&w=256&q=75)
🤝 Top Collabolators
.png&w=640&q=75)
.png&w=640&q=75)
.png&w=640&q=75)
🤓 Submissions
Multimodal Moderator
Current content moderators can process either text or images, but not both. Separate applications are needed to moderate multiple content types. Also, the moderators produce binary output that doesn’t explain why certain content is not appropriate. Our solution is the multimodal moderator. One application that can check if text or image is appropriate or not. It can “understand” the message and explain why certain content is not appropriate. The technology we used is Zapier with Discord as the front end. Zapier is a no-code solution to connect with an AI model, and does not require a constantly running program to connect with Discord. GPT-4 Vision hosted on Clarifai platform is the engine that can check if text or image is appropriate. It sends back a narrative to explain the reasoning.
GPDoc
Doctors spend a significant amount of time reading doctors' notes as part of their workflow. A study published in the Journal of General Internal Medicine in 2016 found that primary care physicians spent an average of 18.4 minutes per patient encounter reading and writing electronic health record notes. This includes reviewing previous notes, writing new notes, and reviewing test results. To mitigate this problem, we developed an AI assistant that can answer questions from doctors’ notes. Questions may be related to patient history, follow-ups, medication adjustments, treatment, diagnosis, procedures, and other categories.
LexaScan
According to a study by McKinsey, employees spend almost 2 hours per day searching and gathering information. Manual extraction processes cannot scale to handle large volumes of data. Automated data extraction solves the inefficiencies of manual information gathering. LexaScan is an AI solution that extracts data from text and images, delivering structured JSON format. We used GPT-4 Vision from OpenAI to extract data from text and images. We also used TruLens to ensure that the output is relevant and harmless. The total addressable market for data extraction software was estimated at $1.4 billion US dollars in 2022 and is expected to reach $3.8 billion dollars by 2028. Conservatively, we estimate the serviceable addressable market to be $140 million and the serviceable obtainable market to be $42 million.
