Upload 5 files

app.py

@@ -0,0 +1,107 @@
+import gradio as gr
+import numpy as np
+import pickle
+
+# Load the trained models
+with open("stroke_clf.pkl", "rb") as f:
+    clf_model = pickle.load(f)
+
+with open("stroke_reg.pkl", "rb") as f:
+    reg_model = pickle.load(f)
+
+def predict_stroke(chest_pain, shortness_breath, irregular_heartbeat, fatigue_weakness, 
+                   dizziness, swelling, pain_neck_jaw, excessive_sweating, 
+                   persistent_cough, nausea_vomiting, high_bp, chest_discomfort, 
+                   cold_hands_feet, snoring, anxiety, age):
+    """
+    Prepares the input features and returns stroke risk prediction using both the classification
+    and regression models.
+    """
+    # Convert boolean inputs to 1 (True) or 0 (False), keeping age as numeric.
+    features = [
+        1 if chest_pain else 0,
+        1 if shortness_breath else 0,
+        1 if irregular_heartbeat else 0,
+        1 if fatigue_weakness else 0,
+        1 if dizziness else 0,
+        1 if swelling else 0,
+        1 if pain_neck_jaw else 0,
+        1 if excessive_sweating else 0,
+        1 if persistent_cough else 0,
+        1 if nausea_vomiting else 0,
+        1 if high_bp else 0,
+        1 if chest_discomfort else 0,
+        1 if cold_hands_feet else 0,
+        1 if snoring else 0,
+        1 if anxiety else 0,
+        age
+    ]
+    
+    sample_input = np.array([features])
+    
+    # Get predictions
+    classification_result = clf_model.predict(sample_input)[0]
+    risk_class = "At Risk" if classification_result == 1 else "Not At Risk"
+    
+    regression_result = reg_model.predict(sample_input)[0]
+    risk_percentage = round(regression_result, 2)
+    
+    return (f"**Classification Result:** {risk_class}\n"
+            f"**Regression Result:** {risk_percentage}% stroke risk")
+
+# Build Gradio Interface using Blocks for a clean, organized UI
+with gr.Blocks(css=".output { white-space: pre-wrap; }") as demo:
+    gr.Markdown(
+    """
+    # Stroke Risk Prediction
+    This application uses two machine learning models to predict stroke risk:
+    
+    - **Classification Model:** Determines if an individual is at risk (Yes/No).
+    - **Regression Model:** Estimates the stroke risk percentage.
+    
+    **Provide the following information based on your symptoms:**
+    """
+    )
+    
+    with gr.Row():
+        with gr.Column():
+            chest_pain = gr.Checkbox(label="Chest Pain")
+            shortness_breath = gr.Checkbox(label="Shortness of Breath")
+            irregular_heartbeat = gr.Checkbox(label="Irregular Heartbeat")
+            fatigue_weakness = gr.Checkbox(label="Fatigue & Weakness")
+            dizziness = gr.Checkbox(label="Dizziness")
+            swelling = gr.Checkbox(label="Swelling (Edema)")
+            pain_neck_jaw = gr.Checkbox(label="Pain in Neck/Jaw/Shoulder/Back")
+            excessive_sweating = gr.Checkbox(label="Excessive Sweating")
+        with gr.Column():
+            persistent_cough = gr.Checkbox(label="Persistent Cough")
+            nausea_vomiting = gr.Checkbox(label="Nausea/Vomiting")
+            high_bp = gr.Checkbox(label="High Blood Pressure")
+            chest_discomfort = gr.Checkbox(label="Chest Discomfort (During Activity)")
+            cold_hands_feet = gr.Checkbox(label="Cold Hands/Feet")
+            snoring = gr.Checkbox(label="Snoring/Sleep Apnea")
+            anxiety = gr.Checkbox(label="Anxiety/Feeling of Doom")
+            age = gr.Number(label="Age", value=50)
+    
+    predict_btn = gr.Button("Predict Stroke Risk", variant="primary")
+    output = gr.Markdown(label="Prediction Results", elem_classes="output")
+    
+    predict_btn.click(
+        predict_stroke, 
+        inputs=[
+            chest_pain, shortness_breath, irregular_heartbeat, fatigue_weakness,
+            dizziness, swelling, pain_neck_jaw, excessive_sweating,
+            persistent_cough, nausea_vomiting, high_bp, chest_discomfort,
+            cold_hands_feet, snoring, anxiety, age
+        ],
+        outputs=output
+    )
+    
+    gr.Markdown(
+    """
+    ---
+    **Disclaimer:** This tool is intended for informational purposes only and should not be used as a substitute for professional medical advice.
+    """
+    )
+
+demo.launch()

requirements.txt

@@ -0,0 +1,7 @@
+gradio
+matplotlib==3.10.0
+numpy==1.26.4
+pandas==2.2.2
+seaborn==0.13.2
+scikit-learn==1.6.1
+xgboost==2.1.4

stroke_clf.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1bbc260597cd7b227c994ae30c014610b45bf247e6691feecef096b8cccc4cd8
+size 117998965

stroke_reg.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:83390fa3edb13b0119e41f4078a260da3bf1c69de4b46867e805765c550d4e06
+size 471045835