Medical Condition Explainer

• 

  Image of the Project

Inspiration

The inspiration for this was that I wanted there would be a simpler way of getting to understand medical technical terms, instead of seeing this complex word that we didn't know. I wanted to create a tool that could take technical information and simplify it, making health knowledge more accessible and understandable to anyone.

What it does

How we built it

I used Visual Studio Code as my coding platform. For the languages that I had used, it was mostly Python and HTML. To start off the project, I asked AI for some examples and templates of projects that I could do. I ran into multiple errors and was able to fix them with the help of GPT and Claude.

Challenges we ran into

Deciding on what design and topic It took way too long to brainstorm a topic and settling down with a topic. There were multiple attempts on one topic, and if that didn't work properly, I would just extract from what I learned from the previous and use it on a different one. Errors with sentence pieces I couldn't install sentence piece in the terminal, which blocked my train.py from working properly.

Combing datasets I had used two datasets, of which I had no idea how to use exactly; I only knew how to use one dataset properly. I had used GPT and Claude to guide and help me through the process of dual parsing.

Local Website Not Updating Correctly due to sentence piece error When running app.py, it offers a local network link, that link's information is not being updated or supported by the model.

Accomplishments that we're proud of

The datasets were successful. When running the commands python load_dataset.py and python dualparse.py, I was able to see that in the output folder, the files were being updated with data in them. I'm proud of being able to create and train a model, but sadly, the training doesn't work properly because of the error with the installation of the sentence piece.

What we learned

I learned how to implement pretrained models, parse two datasets, update data, lots of command prompts for a Python virtual environment, and how Flask and transformers work with Python.

What's next for Medical Condition Explainer

I would need to successfully install SentencePiece and properly be able to train the model effectively. In the future, the medical condition explainer will be used as a stepping stone for a better and more advanced project, where this time it would include a camera to analyze details and patterns based on even more datasets.

Built With

• claude
• gpt
• html
• huggingface
• json
• medlineplus
• python
• torch