Create app.py

app.py

@@ -0,0 +1,132 @@
+#input values, note give None if disease_type is not mentioned.
+
+disease_type = input("Enter the Disease Type: ")
+symptoms = input("Enter the Symptoms (comma separated): ")
+age = float(input("Enter the Age: "))
+gender = input("Enter the Gender (Male/Female/Other): ")
+lab_results = float(input("Enter the Lab Results value: "))
+
+###################################################################
+
+import pandas as pd
+
+# Upload cured patients data from a CSV file
+# Assuming the CSV file is stored in the same directory as the Jupyter Notebook
+file_path = 'cured_patients.csv'  # Path to the CSV file
+
+# Load the data
+cured_patients = pd.read_csv(file_path)
+
+# Display the first few rows of the dataset to ensure it's loaded correctly
+
+from sklearn.preprocessing import LabelEncoder
+from sklearn.feature_extraction.text import CountVectorizer
+from sklearn.preprocessing import StandardScaler
+
+# Encode categorical data
+label_encoder = LabelEncoder()
+cured_patients['Gender'] = label_encoder.fit_transform(cured_patients['Gender'])
+cured_patients['Disease Type'] = label_encoder.fit_transform(cured_patients['Disease Type'])
+
+# Vectorize symptoms
+vectorizer = CountVectorizer()
+symptoms_matrix = vectorizer.fit_transform(cured_patients['Symptoms'])
+
+# Scale numerical features
+scaler = StandardScaler()
+cured_patients[['Age', 'Lab Results']] = scaler.fit_transform(cured_patients[['Age', 'Lab Results']])
+
+
+from sklearn.metrics.pairwise import cosine_similarity
+import pandas as pd
+
+# Define a function to find similar patients
+def find_similar_patients(current_patient, cured_patients, n=3):
+    # Handle NaN values in the cured_patients dataset
+    cured_patients.fillna({
+        'Age': cured_patients['Age'].median(),
+        'Gender': 'Unknown',
+        'Lab Results': cured_patients['Lab Results'].median(),
+        'Disease Type': 'Unknown',
+    }, inplace=True)
+
+    # Filter based on Disease Type if provided
+    if 'Disease Type' in current_patient and current_patient['Disease Type']:
+        filtered_cured_patients = cured_patients[cured_patients['Disease Type'] == current_patient['Disease Type']]
+        if filtered_cured_patients.empty:
+            print("No matching Disease Type found. Returning results from all patients.")
+            filtered_cured_patients = cured_patients
+    else:
+        print("Disease Type not provided. Using all patients for similarity calculation.")
+        filtered_cured_patients = cured_patients
+
+    # Combine features (Symptoms vector + numerical features)
+    features = pd.concat([
+        pd.DataFrame(symptoms_matrix.toarray()),
+        filtered_cured_patients[['Age', 'Gender', 'Lab Results']]
+    ], axis=1)
+
+    # Check and fill any remaining NaN values in features
+    features = features.fillna(0)
+
+    # Convert current patient into the same feature format
+    current_features = vectorizer.transform([current_patient['Symptoms']]).toarray()
+
+    # Handle unseen labels for 'Gender'
+    try:
+        current_gender = label_encoder.transform([current_patient['Gender']])[0]
+    except ValueError:
+        print(f"Warning: Unseen label '{current_patient['Gender']}' for Gender. Assigning default value 0.")
+        current_gender = 0
+
+    # The StandardScaler was fitted only on 'Age' and 'Lab Results'
+    current_numerical = scaler.transform([[current_patient['Age'], current_patient['Lab Results']]])
+
+    # Add 'Gender' to the transformed data
+    current_numerical_df = pd.DataFrame(current_numerical, columns=['Age', 'Lab Results'])
+    current_numerical_df['Gender'] = current_gender
+
+    # Combine current patient features
+    current_combined = pd.concat([pd.DataFrame(current_features), current_numerical_df], axis=1)
+
+    # Check and fill any NaN values in current_combined
+    current_combined = current_combined.fillna(0)
+
+    # Compute similarity scores
+    similarity_scores = cosine_similarity(current_combined, features)
+    filtered_cured_patients['Similarity'] = similarity_scores[0]
+
+    # Calculate match percentage
+    filtered_cured_patients['Match Percentage'] = (filtered_cured_patients['Similarity'] * 100).round(2)
+
+    # Retrieve top n similar patients
+    top_matches = filtered_cured_patients.sort_values(by='Similarity', ascending=False).head(8)
+    return top_matches[['Patient ID', 'Remedial Measures', 'Similarity', 'Match Percentage']]
+
+# Example current patient details
+current_patient = {
+    'Disease Type': disease_type,
+    'Symptoms': symptoms,
+    'Age': age,
+    'Gender': gender,
+    'Lab Results': lab_results,
+}
+
+# Find similar patients
+similar_patients = find_similar_patients(current_patient, cured_patients)
+print(similar_patients)
+
+
+
+
+
+
+##################################################
+# sample input:
+
+# Enter the Disease Type: COVID-19
+# Enter the Symptoms (comma separated): Dry cough, Fatigue, Loss of taste
+# Enter the Age: 30
+# Enter the Gender (Male/Female/Other): Male
+# Enter the Lab Results value: 27
+