Inspiration
Our project builds on the foundation of a previous initiative developed by one of our team members: a Covid-19 chatbot designed to assist users during the pandemic and to see whether someone potentially has COVID-19. Recognizing the potential of that earlier work, we decided to reimagine and significantly expand its scope and functionality. Our new chatbot goes beyond providing information; it aims to simulate the role of a nurse by collecting and processing patient data, performing preliminary assessments, and seamlessly relaying the information to medical professionals. (Although we ask the patient to relay it directly to a doctor via their emails using our email prompt in order to create a general idea for a potential diagnoses that a real doctor can provide) By bridging the gap between patients and healthcare providers, this innovative tool has the potential to enhance efficiency, improve patient outcomes, and reduce the workload on healthcare systems while maximizing the efficiency of doctors. This evolution demonstrates our commitment to leveraging technology to address critical challenges in healthcare delivery, ensuring accessible and effective support for those in need.
What it does
ClearDr AI is designed to assist patients by streamlining the diagnostic process. It begins by gathering detailed information about the symptoms a patient is experiencing. Using this input, it generates relevant diagnostic questions to further refine and narrow down potential causes. By guiding patients through this interactive process, ClearDr AI helps them gain a clearer understanding of their own condition. Which can also help raise awareness towards self care. The system then equips patients with structured and concise information to effectively communicate their symptoms and concerns to medical professionals. This not only saves valuable time during consultations but also ensures healthcare providers receive accurate and comprehensive insights, ultimately contributing to faster and more effective diagnosis and treatment.
How we built it
To build ClearDr AI, we employed a robust and well-integrated technology stack designed to optimize both functionality and user experience. Here’s a breakdown of how we approached the development process.We used HTML as the foundation for structuring the user interface of the application. HTML provided the framework to design a responsive and intuitive layout for users to interact with, the friendly UI part for the criteria. CSS was utilized to enhance the visual aesthetics, ensuring a clean and professional design. This included styling forms, buttons, and other UI components to deliver a user-friendly and visually stunning experience. For interactivity and functionality, JavaScript played a pivotal role. JavaScript was responsible for handling user inputs and creating a seamless interaction between the user and the application. The backend of ClearDr AI was developed using Python, which powered the artificial intelligence functionalities. Python’s robust libraries and frameworks made it an ideal choice for implementing the AI-driven diagnostic engine. We integrated the Gemini API model as the backbone for the AI component. The Gemini model provided advanced natural language processing (NLP) capabilities, enabling the chatbot to generate intelligent diagnostic questions and analyze patient inputs with precision. Basically the prediction parts of it was what we used Gemini for. We were planning to use logistic regression to make our own predictions but we were short on time. By leveraging this model, we ensured accurate symptom analysis and meaningful suggestions.
Challenges we ran into
Our project journey has been marked by a significant restart at the 3 hour mark and numerous errors, each presenting valuable learning opportunities. While we are still in the early stages of mastering coding and lack extensive experience, we have approached these challenges with resilience and a growth mindset (which led us to improve our productivity by a lot). Every hurdle has been a chance to deepen our understanding, refine our skills, and strengthen our teamwork. Through persistence and collaboration, we have made remarkable progress, overcoming obstacles and steadily improving with every step forward. Integrating four distinct programming languages—HTML, CSS, JavaScript, and Python—into a cohesive project was one of the most complex challenges we faced during development. Each language serves a specific purpose within the application, and ensuring seamless communication between them requires meticulous planning and problem-solving.
Accomplishments that we're proud of
We are proud of the remarkable progress we’ve made despite the many challenges faced throughout your project. From overcoming the significant hurdle of integrating four different programming languages—HTML, CSS, JavaScript, and Python—into a cohesive system, to navigating through errors and setbacks that provided invaluable learning opportunities, we’ve embraced every obstacle as a chance to grow. Our persistence and teamwork have been the driving force behind developing ClearDr AI, a tool that bridges technology and healthcare, and you’ve significantly advanced your coding skills along the way. We are particularly proud of how we’ve used the Gemini API to integrate artificial intelligence into the project, enabling ClearDr AI to simulate the role of a nurse and streamline the diagnostic process for patients. Despite still being in the mid stages of our coding journey, the progress made is a testament to your resilience, growth, and ability to learn quickly. We’ve transformed frustrations into breakthroughs, continuously improving. The journey has been a testament to our commitment, and as a team, We are proud of how far we’ve come, the obstacles we’ve conquered, and the potential we’ve unlocked.
What we learned
Throughout the course of our hackathon, learning to use Flask played a pivotal role in enabling us to rapidly develop a functional web application. Initially, none of us had any experience with Flask, and we were unsure how to integrate it into our project. However, through diligent research and hands-on experimentation, we quickly grasped the framework’s fundamentals and were able to harness its power effectively. Flask, with its lightweight and flexible structure, provided the perfect environment for us to focus on building our application without being overwhelmed by unnecessary complexity. Its minimalist design made it incredibly easy to understand and implement, even for first-time users like us. As a web framework for Python, Flask allows us to seamlessly connect the frontend and backend of our application, streamlining the process of building dynamic web features. One of the most valuable lessons we learned was how Flask’s built-in development server and debugger significantly accelerated our testing and troubleshooting process. This feature allowed us to catch errors early, iterate quickly, and refine our application with efficiency. Moreover, Flask’s compatibility with a wide array of extensions gave us the flexibility to customize our project and scale it as needed. We learned how to integrate essential tools like databases and APIs, and how to fine-tune our application for performance and functionality. In addition to the technical skills we developed, Flask taught us the importance of simplicity and adaptability in software design. It reinforced the idea that a minimalistic approach, when used correctly, can lead to powerful and scalable solutions. Ultimately, Flask not only helped us build a functional web app within a tight timeframe but also empowered us to gain valuable insights into web development, improve our problem-solving skills, and learn the importance of choosing the right tools for the task at hand.
What's next for ClearDr AI
The future of ClearDr AI holds exciting possibilities, as there are several directions in which the project can evolve and grow to enhance its functionality, accessibility, and impact.
As the AI backend is powered by the Gemini API model, there’s potential to refine the diagnostic capabilities further. By incorporating more advanced AI models or expanding the dataset used for training, ClearDr AI could improve its accuracy in diagnosing a broader range of symptoms and conditions, leading to better patient outcomes. Additionally, integrating medical images or more detailed patient histories could make the AI more comprehensive and capable of offering deeper insights.
While the current user interface (UI) is functional, enhancing it for an even more intuitive, personalized experience will be crucial. Implementing features like voice recognition for symptom reporting or adaptive questionnaires based on previous responses could make the chatbot even more interactive and user-friendly. Adding multilingual support will expand ClearDr AI’s reach, enabling it to serve patients in different regions and make healthcare more accessible for non-English speakers.
One of the next steps could involve integrating ClearDr AI with existing electronic health record (EHR) systems, allowing seamless data transfer between the chatbot and healthcare providers. This would make the AI-driven information more directly actionable by medical professionals, enhancing the overall workflow of healthcare settings. Implementing features like appointment scheduling or medication tracking could further streamline the patient experience, making ClearDr AI an all-encompassing tool for healthcare assistance.
As the project grows, transitioning from a local website hosted through Flask to a scalable cloud-based solution could ensure ClearDr AI’s reliability and availability for a larger number of users. This would enable the system to handle increased traffic, provide real-time updates, and scale globally. Exploring deployment on cloud platforms like AWS or Azure would be beneficial.
Log in or sign up for Devpost to join the conversation.