Spaces:
Sleeping
Sleeping
import streamlit as st | |
import re | |
import pandas as pd | |
from sklearn.feature_extraction.text import TfidfVectorizer | |
from sklearn.naive_bayes import MultinomialNB | |
from sklearn.model_selection import train_test_split | |
# Load your symptom-disease data | |
data = pd.read_csv("Symptom2Disease.csv") | |
# Initialize the TF-IDF vectorizer | |
tfidf_vectorizer = TfidfVectorizer() | |
# Apply TF-IDF vectorization to the preprocessed text data | |
X = tfidf_vectorizer.fit_transform(data['text']) | |
# Split the dataset into a training set and a testing set | |
X_train, X_test, y_train, y_test = train_test_split(X, data['label'], test_size=0.2, random_state=42) | |
# Initialize the Multinomial Naive Bayes model | |
model = MultinomialNB() | |
# Train the model on the training data | |
model.fit(X_train, y_train) | |
# Set Streamlit app title with emojis | |
st.title("Health Symptom-to-Disease Predictor π₯π¨ββοΈ") | |
# Define a sidebar | |
st.sidebar.title("Tool Definition") | |
st.sidebar.markdown("__Tool Definition__") | |
st.sidebar.markdown("This tool helps you identify possible diseases based on the symptoms you provide.") | |
st.sidebar.markdown("This is my project for the KaggleX program.") | |
st.sidebar.markdown("Please note that this tool is for informational purposes only. Always consult a healthcare professional for accurate medical advice.") | |
# Initialize chat history | |
if "messages" not in st.session_state: | |
st.session_state.messages = [] | |
# Function to preprocess user input | |
def preprocess_input(user_input): | |
user_input = user_input.lower() # Convert to lowercase | |
user_input = re.sub(r"[^a-zA-Z\s]", "", user_input) # Remove special characters and numbers | |
user_input = " ".join(user_input.split()) # Remove extra spaces | |
return user_input | |
# Function to predict diseases based on user input | |
def predict_diseases(user_clean_text): | |
user_input_vector = tfidf_vectorizer.transform([user_cleaned_text]) # Vectorize the cleaned user input | |
predictions = model.predict(user_input_vector) # Make predictions using the trained model | |
return predictions | |
# Add user input section | |
user_input = st.text_area("Enter your symptoms (how you feel):", key="user_input") | |
# Add button to predict disease | |
if st.button("Predict Disease"): | |
# Display loading message | |
with st.spinner("Diagnosing patient..."): | |
# Check if user input is not empty | |
if user_input: | |
cleaned_input = preprocess_input(user_input) | |
predicted_diseases = predict_diseases(cleaned_input) | |
# Display predicted diseases | |
st.session_state.messages.append({"role": "user", "content": user_input}) | |
st.session_state.messages.append({"role": "assistant", "content": f"Based on your symptoms, you might have {', '.join(predicted_diseases)}."}) | |
st.write("Based on your symptoms, you might have:") | |
for disease in predicted_diseases: | |
st.write(f"- {disease}") | |
else: | |
st.warning("Please enter your symptoms before predicting.") | |
# Display a warning message | |
st.warning("Please note that this tool is for informational purposes only. Always consult a healthcare professional for accurate medical advice.") | |
# Create FAQs section | |
show_faqs = st.checkbox("Show FAQs") | |
if show_faqs: | |
st.markdown("## Frequently Asked Questions") | |
st.markdown("**Q: How does this tool work?**") | |
st.markdown("A: The tool uses a machine learning model to analyze the symptoms you enter and predicts possible diseases based on a pre-trained dataset.") | |
st.markdown("**Q: Is this a substitute for a doctor's advice?**") | |
st.markdown("A: No, this tool is for informational purposes only. It's essential to consult a healthcare professional for accurate medical advice.") | |
st.markdown("**Q: Can I trust the predictions?**") | |
st.markdown("A: While the tool provides predictions, it's not a guarantee of accuracy. It's always best to consult a healthcare expert for a reliable diagnosis.") | |
# Add attribution | |
st.markdown("Created with β€οΈ by Joas") | |