Link for Final Video: https://drive.google.com/open?id=1qaL98E0W4yiVGSAid69VWtqq_fg0WsCx
Link to Final Presentation here
After speaking with several medical professionals at CHOP, it became abundantly clear that identifying patients is a time-consuming process that often requires multiple individuals. Additionally, certain key pieces of information such as name, condition, and vitals would be extremely helpful for medical professionals to have. We built a pair of Augmented Reality glasses to help doctors and other medical professionals interact with patients. Our glasses use face recognition to identify patients. It then connects to a secure to pull and display patient information. The doctor uses the information to improve the patient-doctor experience and help reduce misinformation in the visit.
We started this semester with an interesting idea and a motivated team looking to create a product that would be useful in the medical sector. We connected with Dr. James Won who helped us flesh out our idea, a pair of augmented reality glasses that would help improve the doctor-patient relationship in hospital settings. Once we had this idea, we started researching to find out more about it. Through Dr. Won, we got the chance to speak with medical professionals to help us determine our exact market and the issues they face and how we were going to solve them with our product.
One of the issues we heard about from medical professionals was this need for two-person verification before a nurse administers any drug or procedure. We thought we could use a camera on our glasses to identify patients to improve efficiency for nurses. We created an architecture and road map for our project and then we got started building. At first, we tried to fit all of our components in our glasses apparatus, but through feedback from Dr. Won and TAs, we switched to having a box on the hip of the user to store the larger components.
We worked on a few different aspects to get our product working for fall demo day. We had to work on facial recognition, building and printing our 3d structure, coding the screen to show information from the facial recognition. We used python to code our facial recognition program and trained the code on a library of images so it could recognize us. We connected the display screen through the raspberry pi and synced it with our facial recognition program. With our custom built glasses apparatus, we had a successful working demo of our basic product.
Next semester, we want to work on improving the functionality by adding additional sensors to the apparatus while keeping the product sleek and not uncomfortable to wear. We will also create a custom PCB and iterate through different form factors as we improve upon our fall demo day product.
Fall Video: https://youtu.be/1Vnjfxuaq-4
Link to our 4 Minute Presentation from Fall Semester: 4 Min Presentation
Our 30 second Video from earlier in the Fall semester can be found here: https://youtu.be/LY-m3RRWRQc