This project is a submission for the AI Intel Global Creators Competition 2023. The Disease Predictor AI Website is designed to assist users in predicting the probability of having a particular disease based on the symptoms they experience. The prediction is made using a deep neural network model trained on a dataset of 132 symptoms and their associated major diseases.
Important: This project is for experimental purposes only and should not be considered as a substitute for professional medical advice. It is crucial to seek medical assistance from qualified healthcare professionals if you have any health concerns or require a diagnosis.
- User-friendly interface: The website offers an intuitive interface for users to input their symptoms.
- Symptom selection: Users can type in or select the tags of symptoms to create their symptom profile.
- Disease prediction: The deep neural network model processes the symptom profile and predicts the probabilities of having various diseases.
- Probability output: The website displays the probabilities of each disease, ranked by the highest probability.
- Disease information: Users can choose a disease of their choice to learn more about its symptoms.
- Open the Disease Predictor AI Website in your preferred web browser. Link is https://dhanvine321-intel-ai-competition-disease-predictor-a-app-ol16z1.streamlit.app/
- On the homepage, you will find a list of symptoms.
- Type in or select the tags of symptoms that you are currently experiencing.
- After entering/selecting your symptoms, press the "Predict" button.
- The website will process your symptom profile using the deep neural network model.
- Once the prediction is complete, the website will display the probabilities of various diseases, with the most probable disease listed first.
- To learn more about a specific disease, click on the disease name or select it from a dropdown menu in the next section
- The website can also provide information about the selected disease, including its common symptoms.
The dataset used for training the model was obtained from Kaggle. It contains 132 symptoms and 42 diseases. You can find the dataset here: https://www.kaggle.com/datasets/kaushil268/disease-prediction-using-machine-learning
It is essential to understand the limitations of the Disease Predictor AI Website:
- Accuracy: The predictions made by the model are based solely on the symptoms provided and may not be entirely accurate or comprehensive.
- Data limitations: The model's predictions depend on the quality and quantity of the training data. It may not capture rare or uncommon diseases that are not well-represented in the dataset.
- No diagnosis: This website is not intended to provide a diagnosis. It only provides probabilities based on symptoms and should not be considered a substitute for professional medical advice.
- False positives and negatives: The model may generate false positives (predicting a disease when it is not present) or false negatives (failing to predict a disease that is present).
As an experimental project, the Disease Predictor AI Website can be further improved in the following ways:
- Enhanced training data: The model's accuracy can be improved by incorporating a larger and more diverse dataset, encompassing a broader range of symptoms and diseases.
- Expert consultation: Collaborating with medical professionals to validate and refine the model can enhance its accuracy and reliability.
- User feedback: Gathering feedback from users to identify and address any issues or limitations encountered during their interactions with the website.
- Risk assessment: Incorporating risk factors and medical history to provide more personalized predictions and recommendations.
The Disease Predictor AI Website serves as an experimental project which I developed for the AI Intel Global Creators Competition 2023. It provides users with a glimpse into the potential applications of deep neural networks in predicting the probability of diseases based on symptoms. Remember, this website should not be taken as a substitute for professional medical advice. Always consult with healthcare professionals for accurate diagnosis and treatment.****