utils/organize_data.py

import os
import shutil
import pandas as pd

# Paths
base_dir = r"c:\Users\ASUS\lung_ai_project"
extracted_dir = os.path.join(base_dir, "data", "extracted_cough", "Respiratory_Sound_Dataset-main")
audio_txt_dir = os.path.join(extracted_dir, "audio_and_txt_files")
diagnosis_file = os.path.join(extracted_dir, "patient_diagnosis.csv")

output_healthy = os.path.join(base_dir, "data", "cough", "healthy")
output_sick = os.path.join(base_dir, "data", "cough", "sick")

# Create directories if not exist
os.makedirs(output_healthy, exist_ok=True)
os.makedirs(output_sick, exist_ok=True)

# Read diagnosis
df = pd.read_csv(diagnosis_file)
print("Columns:", df.columns)
print(df.head())

# Map Patient_ID to Diagnosis
diagnosis_map = dict(zip(df['Patient_ID'], df['DIAGNOSIS'])) # CSV header implies 'Patient_ID' and 'DIAGNOSIS'

# Process files
files = os.listdir(audio_txt_dir)
wav_files = [f for f in files if f.endswith('.wav')]
print(f"Found {len(wav_files)} wav files")

count_healthy = 0
count_sick = 0

for wav_file in wav_files:
    # Filename format: 101_1b1_Al_sc_Meditron.wav
    try:
        patient_id = int(wav_file.split('_')[0])
    except ValueError:
        print(f"Skipping {wav_file}")
        continue
    
    diagnosis = diagnosis_map.get(patient_id, "Unknown")

    
    if diagnosis == 'Healthy':
        shutil.copy2(os.path.join(audio_txt_dir, wav_file), os.path.join(output_healthy, wav_file))
        count_healthy += 1
    elif diagnosis != "Unknown":
        shutil.copy2(os.path.join(audio_txt_dir, wav_file), os.path.join(output_sick, wav_file))
        count_sick += 1
    else:
        print(f"Unknown diagnosis for patient {patient_id}")

print(f"Result: {count_healthy} healthy files, {count_sick} sick files.")