import pandas as pd import os import torch from torch.utils.data import Dataset class SurvivalDataset(Dataset): def __init__(self, csv_file, root_dir): # Using the correct filename you just provided self.metadata = pd.read_csv(csv_file) self.root_dir = root_dir def __len__(self): return len(self.metadata) def __getitem__(self, idx): # 1. Get the patient ID patient_id = self.metadata.iloc[idx]['ID'] # 2. IMPORTANT: Check your folder structure on the D: drive. # If your folders are named just "UCSF-PDGM-004" instead of "UCSF-PDGM-004_nifti", # remove the "_nifti" from the line below. img_folder = os.path.join(self.root_dir, patient_id) # 3. Load your T1c path t1c_path = os.path.join(img_folder, f"{patient_id}_T1c.nii.gz") # 4. Get your Survival Targets survival_time = self.metadata.iloc[idx]['OS'] event_status = self.metadata.iloc[idx]['1-dead 0-alive'] return { 't1c_path': t1c_path, 'time': torch.tensor(float(survival_time)), 'event': torch.tensor(float(event_status)) } # Quick Test dataset = SurvivalDataset("clinicaldata.csv", r"D:\jainam\UCSF-PDGM-v5") print(dataset[0])