Navigating Our Tutorial Notebooks

The following notebooks showcase examples of how you can use jarvAIs for classification, regression, and survival tasks!

All of the notebooks below use clinical data from RADCURE.

The RADCURE dataset was collected clinically for radiation therapy treatment 
planning and retrospectively reconstructed for quantitative imaging research.

Accessing the Data

You can find the raw and processed clinical data here.

Data Processing

The following function was used to process the raw data:

Source code in src/jarvais/utils/functional.py

def process_RADCURE_clinical(df):
    """
    Processes RADCURE clinical data.

    Raw data found here: https://www.cancerimagingarchive.net/collection/radcure/
    """
    df_converted = pd.DataFrame({
        'Study ID': df['patient_id'],
        'survival_time': df['Length FU'],
        'death': df['Status'].apply(lambda x: 1 if x == 'Dead' else 0),
        'age at dx': df['Age'],
        'Sex': df['Sex'],
        'T Stage': df['T'],
        'N Stage': df['N'],
        'Stage': df['Stage'],
        'Dose': df['Dose'],
        'Chemotherapy': df['Chemo'].apply(lambda x: 1 if x != 'none' else 0),
        'HPV Combined': df['HPV'].apply(lambda x: 1 if isinstance(x, str) and 'positive' in x.lower() else None),
        'Smoking Status': df['Smoking Status'],
        'Disease Site': df['Ds Site'].str.lower()
    })

    return df_converted