this is my first ML project, that I decided to work on. Here, i am using the dataset provided by @a3amat02 on kaggle.
The dataset includes patient data, and here is an overview:
| Column | Non-Null Count | Dtype | |
|---|---|---|---|
| 0 | Age | 4900 non-null | float64 |
| 1 | Gender | 4900 non-null | object |
| 2 | Smoking Status | 4900 non-null | object |
| 3 | Lung Capacity | 4900 non-null | float64 |
| 4 | Disease Type | 4900 non-null | object |
| 5 | Treatment Type | 4900 non-null | object |
| 6 | Hospital Visits | 4900 non-null | float64 |
| 7 | Recovered | 4900 non-null | object |
although the dataset is not a very large one, and the resulting trained model won't be the most accurate, there is, it could serve a "fun" purpose for those, who are willing to try and see some predictions about ones risks and chances on the topic lung diseases.
So it well neither be an exact, nor a trustworthy estimator for one's condition as complex as health, but it could spread some awareness, that is based on pure statistics, for curiosity reasons.
- Find out gender spread lung diseases
- Find out which diseases are the most dangerous and otherwise
- Find the best treatment for each disease
- Train a ML model on the given data to predict further chances based on new data
- Build a simple hosted application and host the model with the statistics and input field for user data to check on their chances for the selected stochastic model
The stochastic model will include:
1. What disease does the patient tend to have #TODO
2. Chances to survive the disease (done)
3. What treatment is better for the patient regarding his data and disease #TODO
4. How smoking influences his chances #TODO
5. Optimal hospital visits for his disease #TODO
This model will not be my main focus, becaus the dataset is not the best one for high accuracy, but i still want to add some features like:
1. What disease does the patient tend to have #TODO
2. What treatment is better for the patient regarding his data and disease #TODO
3. How smoking influences his chances #TODO
4. Optimal hospital visits for his disease #TODO
- If you have some suggestions, or find errors, feel free to add some pull requests, and i will review them ASAP :)
- The project is completely open-source and u can use it however you like