Answer: Transfer learning

## Explanation **Transfer learning** is the most efficient approach in this scenario because: 1. **Reuse of pre-trained model**: The hospital already has a model trained to detect pneumonia from chest X-rays, which has learned relevant features from medical imaging data. 2. **Similar domain and data type**: Both tasks involve chest X-rays and medical image analysis, so the feature representations learned for pneumonia detection are likely relevant for lung cancer detection. 3. **Efficiency benefits**: - **Reduced training time**: Instead of training from scratch, you can fine-tune the existing model - **Less data required**: Transfer learning typically requires less labeled data for the new task - **Better performance**: The model starts with useful feature representations rather than random initialization 4. **Process**: - Keep the early layers (feature extractors) of the pneumonia model - Replace the final classification layer for lung cancer detection - Fine-tune the model on lung cancer X-ray data **Why other options are incorrect**: - **A) Unsupervised learning**: This involves finding patterns in unlabeled data, not adapting a pre-trained model - **B) Self-supervised learning**: This is a type of unsupervised learning where models create their own labels from data, not suitable for adapting existing models - **D) Reinforcement learning**: This involves learning through trial-and-error with rewards, not appropriate for adapting image classification models Transfer learning is particularly effective in medical imaging where labeled data is scarce and expensive to obtain, making it the optimal choice for adapting an existing model to a related task.

Author: Jin H