Anaesthesia OCR Project

I worked on a year-long project where a camera takes pictures of an anaesthesia monitor every few seconds (an anaesthesia monitor displays medical data of a patient such as blood pressure and heart rate.) From those images, Optical Character Recognition (OCR) extracted the patient data displayed on the monitor to medical software for analysis. I developed a desktop user interface using Python with existing OCR components. I also developed a mobile app using Kivy (a Python framework), which communicated with the desktop program over WebSockets. I was one of eight students selected to present at the showcase presentations at university. Below is a video showing the system in action:

Throughout the project, I sought feedback from my supervisors, helping to ensure that my system is aligned with end user requirements. I used AI tools to accelerate the development of the project. I built an installer using Pyinstaller and Nullsoft Scriptable Install System. This allowed both my program and the medical software to be installed on any Windows PC easily without any time consuming and error-prone manual setup steps. I wrote documentation files on how to set up and use the system.