Artificial Intelligence in Radiology Practice Questions

Q: A hospital implements an AI algorithm for detecting intracranial hemorrhage on CT scans. The algorithm was trained on data from a different population with different CT scanner protocols. The algorithm shows decreased performance. Which concept explains this phenomenon?

Dataset shift and lack of generalizability. ***Dataset shift and lack of generalizability*** - **Dataset shift** occurs when the distribution of data used during training differs significantly from the data encountered in clinical practice, such as different **scanner protocols**. - This leads to a lack of **generalizability**, where the AI performs poorly in new environments because it cannot adapt to variations in **population demographics** or imaging hardware. *Overfitting of the training data* - **Overfitting** happens when a model learns the noise and specific details of the training set too well, failing to predict outcomes on any new data. - While it affects generalizability, the specific issue of switching **scanner protocols** and **populations** is more accurately described as a shift in data domains. *Insufficient neural network layers* - Insufficient layers or **lack of depth** typically results in **underfitting**, where the model is too simple to capture the underlying patterns in the training data. - This is a structural limitation of the model architecture rather than an issue related to the **external validation** or the source of the data. *Poor image preprocessing* - **Preprocessing** involves cleaning or standardizing images before feeding them into the model; errors here would affect consistency across all datasets. - While standardized preprocessing helps mitigate differences, the root cause of decreased performance across different **institutional protocols** is the mismatch in the data distribution itself.

Question 1

A research team develops an AI algorithm using 100,000 CT scans from multiple institutions. The algorithm shows excellent performance (AUC 0.96) but requires extensive computational resources. To deploy it in resource-limited settings, they propose model compression techniques. Evaluate the potential trade-offs and propose the most balanced approach.

Accepted Answer

Use knowledge distillation to train a smaller model that mimics the larger model while accepting minimal performance decrease

Answer

Model compression always maintains performance while reducing size

Answer

Avoid compression as any performance loss is unacceptable in medical AI

Answer

Random pruning of neural network connections is sufficient

Question 2

A radiology department is evaluating two AI algorithms for fracture detection. Algorithm A has AUC-ROC of 0.95, while Algorithm B has AUC-ROC of 0.92 but provides explainable results showing which image regions influenced its decision. Considering clinical implementation and medicolegal aspects, which statement best evaluates the choice?

Accepted Answer

Algorithm B may be preferred despite lower AUC due to interpretability and accountability

Answer

Algorithm A should always be chosen due to superior performance metrics

Answer

AUC-ROC is the only relevant metric for clinical decision making

Answer

The difference in AUC is clinically insignificant so both are equivalent

Question 3

A hospital implements an AI algorithm for detecting intracranial hemorrhage on CT scans. The algorithm was trained on data from a different population with different CT scanner protocols. The algorithm shows decreased performance. Which concept explains this phenomenon?

Accepted Answer

Dataset shift and lack of generalizability

Answer

Overfitting of the training data

Answer

Insufficient neural network layers

Answer

Poor image preprocessing

Question 4

A 55-year-old male presents with chronic cough. A chest X-ray is analyzed by an AI algorithm that reports a 4mm lung nodule in the right upper lobe with 85% confidence. The human radiologist reviews the image but cannot identify the nodule. What is the most appropriate next step?

Accepted Answer

Obtain a second opinion from another radiologist and correlate with clinical findings

Answer

Accept the AI finding and proceed to CT scan immediately

Answer

Reject the AI finding as a false positive without further action

Answer

Report only what the radiologist can see and discard AI output

Question 5

How does a Generative Adversarial Network (GAN) work in the context of medical image synthesis?

Accepted Answer

Through a generator creating images and a discriminator distinguishing real from fake

Answer

By comparing two identical neural networks

Answer

By using regression algorithms to predict image quality

Answer

Through supervised learning with labeled datasets only

Artificial Intelligence in Radiology — MCQs

Artificial Intelligence in Radiology — MCQs

On this page

Practice by Chapter

Want unlimited practice?