MedLink AI

• 

  The home page

• 

  The emergency support

• 

  New patient assessment

• 

  Symptoms patients can select from

• 

  Sustainable develoment impact column

• 

  Patient History page

• 

  The Analytics dashboard

• 

  Risk distribution and average score by risk bar charts

• 

  Most reported symptoms barchart

What inspired me to create MedLink AI was that many people, especially those in underserved communities in Ghana, are unsure whether their symptoms require immediate medical attention. Delayed decisions can lead to preventable health complications, so I wanted to build a tool that provides quick, accessible health guidance while supporting healthcare workers with digital record keeping. I built the project using Python, Streamlit, and SQLite. The application uses a rule-based intelligent triage engine that evaluates symptoms, age, and severity to classify patients into low, medium, or high risk categories. It also stores patient records, generates downloadable PDF medical reports, provides analytics through dashboards, and allows patient data to be exported as CSV for reporting. Throughout the project, I learned how to design a full-stack Python application, manage persistent data with SQLite, create multi-page Streamlit interfaces, and generate professional PDF reports. The biggest challenge was balancing accuracy with simplicity. Since this is a decision-support tool rather than a diagnostic system, I focused on creating clear, responsible risk assessments while keeping the interface easy to use. The experience strengthened both my technical and problem-solving skills. I want to make it even more useful by adding hospital referrals, support for Twi since many Ghanaians speak Twi, and smarter health assessments so more people can get the right care quickly.

Built With

• python
• sqlite
• streamlit