Inspiration
After finding out how many people are affected by procedural errors, we wanted to create a solution using tech to increase patient safety and reduce medical errors.
What it does
SafeOR uses modern technologies such as computer vision, NLP, and machine learning to bring visibility to patient safety and the need for tech-enabled solutions to reduce medical errors.
How we built it
We built the front end using react.js and tailwind CSS. For the backend, we used Firebase, Python, and Flask.
Challenges we ran into
With a combination of being new hackers and working with difficult tools for the first time, as aforementioned, we ran into many problems. Initially, we found computer vision and speech-to-text recognition to be easier than we had imagined. However, as we continued to work with the libraries opencv and google cloud speech to text, we found many handfuls of hardships. This included many frustrations with real-time data analysis as well as figuring out how to handle multiprocessing and even exploring threading for a bit. As for the front end, we were also new to creating a web app and had to learn Next.js. Later on down the road, we struggled to figure out how to connect our backends of NLP and computer vision together, as well as create a connection between the front end and the back end.
Accomplishments that we're proud of
We are proud to have created such a technically difficult problem and having created and trained our own data set for the machine learning portion and successfully implemented everything together from tying the NLP together with firebase and connecting that to the computer vision portion and finally pulling it all together with the front end. In addition, being new to many of the tools and technologies we used in this project, we are happy to say that we created a successful and well-working product.
What we learned
We learned many tools such as google cloud speech to text, opencv, nextjs, react, and the most important of them all, teamwork! Our team had 2 members participating for the first time, and we aimed this project to learn and practice what we learned.
What's next for SafeOR
SafeOR can grow a lot and add additional features based on medical research. As students, we tried to research on our own, but we believe talking to the end user, medical professionals would give us an insight into their processes and what they would like to see from our end.
Check it out!
https://github.com/DanierLiu/SafeOR
https://drive.google.com/drive/folders/1HpAMb1kSGurD9awt251XNis3XhqGvJmn?usp=share_link
Log in or sign up for Devpost to join the conversation.