Inspiration
Medically Speaking was created as a solution to a problem experienced by one of its founders. They had a frustrating experience in the hospital and healthcare system, where they struggled with a lack of clear and consistent information about their diagnosis. The founders recognized that many patients face this same issue, as doctors may not have the time to provide detailed explanations and online research can be overwhelming and unreliable. Medically Speaking aims to improve the patient experience by providing accurate and accessible information, saving research time for patients, and reducing miscommunication errors in routine patient care.
What it does
Medically Speaking is designed to help patients and medical professionals obtain accurate information regarding their personal conditions. Utilizing Natural Language Processing (NLP), previous conversations, and a medical history storage system, our bot identifies the intent behind a message, scans for any possible context (such as past medical history or previously mentioned conditions), and chooses the most appropriate way to respond.
How we built it
By planning out our objectives as a team and then separating into smaller objectives that more pertained to our strengths, we were able to simultaneously work on several pieces of the project at the same time, and put it together at the end, rather than focusing on one piece at a time.
Challenges we ran into
As our bot was created in Python, we were initially set to create and connect a Flask API to ReactJS so that we could not only display the chatbot in a more user-friendly and appealing manner, but we also would have created an API that could be integrated into (basically) any other web or even mobile application. However, our chatbot was created with the intent of retaining memories (such as past conversations). Due to time constraints and a general lack of familiarity with APIs and the Flask framework, we chose to preserve this functionality by developing a webpage using HTML and CSS instead.
Additionally, less than an hour before submission, the devpost we had put a ton of heart into decided to delete itself when we pressed save and continue. This really caused our final product (the devpost part of it) to be more rushed than our initial one we had worked on. (Although hopefully it doesn’t show!)
Accomplishments that we're proud of
Our biggest accomplishment was fully prioritizing the functionality of our chatbot without compromising on the user experience. This allowed for the chatbot to be more capable of answering a wider variety of questions and providing more information.
Another accomplishment is working together as a brand new team. This ended up being a huge benefit to us, as the group's ability to work together actually ended up accelerating our project, rather than hindering it.
Lastly, our final product. Despite having little to no experience with any of the systems, programs, and concepts we utilized, we managed to produce a final program that we believe is absolutely outstanding.
What we learned
We were able to understand the Flask framework. We also learned a lot about ReactJS and APIs in general despite not being able to implement it into our final project.
We absolutely should have written our text for the devpost in a google document.
Next time, we will absolutely focus on learning the systems and frameworks (like Flask) beforehand!!
What's next for Medically Speaking
We would love to be able to integrate our program into a fully functional website. Our bot has the capability to use personal and family medical history, prescriptions, and other related context information, to better personalize the information it provides to the user. Being able to showcase this feature at its full force would be amazing!
Log in or sign up for Devpost to join the conversation.