Fibropred - Bitsxlamarato - PAI

Inspiration

The inspiration for this project came from the need to improve the diagnosis and treatment of fibrotic diseases. By leveraging machine learning and data analysis, we aim to provide better predictive models that can assist healthcare professionals in making more informed decisions.

What it does

We have created a a predictive model that analyzes patient data to predict the progression of fibrotic diseases. It uses various indicators and patient history to provide insights into the likelihood of disease progression, helping doctors to tailor treatment plans more effectively.

How we built it

We built Fibropred using a combination of data preprocessing, exploratory data analysis, and machine learning models. The project involved:

• Cleaning and preprocessing the raw data.
• Conducting univariate and bivariate analysis to understand the relationships between different variables.
• Training various machine learning models, including Decision Trees, Random Forests, and Support Vector Machines, to predict disease progression.
• Evaluating the models using cross-validation and refining them based on performance metrics.

Challenges we ran into

One of the main challenges was handling the missing and inconsistent data in the dataset, as well as selecting the useful variables. We had to carefully preprocess the data to ensure that the models could learn effectively.

Accomplishments that we're proud of

We are proud of successfully building a predictive model that can provide valuable insights into the progression of fibrotic diseases. Our model achieved high accuracy and demonstrated the potential to assist healthcare professionals in making better treatment decisions. Additionally, we were able to preprocess and analyze a complex dataset effectively, which was a significant achievement. On top of that, we created an easy to use UI for the practitioners to use.

What we learned

Throughout this project, we learned the importance of data preprocessing and the impact it has on the performance of machine learning models. We also gained experience in exploratory data analysis and the use of various machine learning algorithms. Furthermore, we learned how to handle real-world data challenges, such as missing values and data inconsistencies.

What's next for Fibropred - Bitsxlamarato - PAI

The next steps for Fibropred include:

• Further refining the predictive models by incorporating additional data and features.
• Collaborating with healthcare professionals to validate the model's predictions in clinical settings, and collecting more data to improve the predictions.

This project has the potential to make a significant impact on the treatment of fibrotic diseases, and we are excited to continue its development.

Note: We had to copy the code from one private repo to a public one, as the data was already in the .git file, and making the original repo public would expose these confidential information.

Built With

• csv
• python
• scikit-learn