Analyzing Secondary Cancer Risk: A Machine Learning Approach

Area of Science:

• Oncology
• Medical Physics
• Biostatistics

Background:

• Rising cancer rates necessitate effective diagnosis and treatment strategies.
• Cancer survivors face potential secondary cancer (SC) risks influenced by treatments and lifestyle.
• Understanding SC risk factors is crucial for long-term survivor health management.

Purpose of the Study:

• To establish a novel relationship between effective organ dose and SC risk using linear regression.
• To compare different prediction methods for SC risk in lung, colon, and breast cancer.
• To investigate the influence of effective dose on SC development in cancer survivors.

Main Methods:

• Utilized linear regression models to analyze the dose-SC risk relationship.
• Employed machine learning (ML) for forecasting SC likelihood based on effective organ doses.
• Compared prediction methods across lung, colon, and breast cancer datasets.

Main Results:

• A positive correlation was observed between effective organ dose and SC risk.
• Linear regression model coefficients reflect organ-specific radiation sensitivity.
• The model demonstrated effectiveness in predicting SC risk based on dose.

Conclusions:

• Linear regression models are significant for predicting SC risk from effective organ doses.
• Organ radiation sensitivity is a key factor in SC risk assessment.
• Findings aid in better understanding and managing long-term health for cancer survivors.