Computer-Aided Detection and Diagnosis Practice Questions

Question 1

Digital radiography differs from conventional in

Accepted Answer

Radiation receptors are different

Answer

X-rays are not required for imaging

Answer

Images cannot be printed

Answer

Uses radiation other than X-rays

Question 2

PACS in medical imaging stands for:

Accepted Answer

Picture archiving and communication system

Answer

Planned archiving computerized system

Answer

Planned archiving common system

Answer

Picture archiving or computerized system

Question 3

Best imaging modality for acute pulmonary embolism

Accepted Answer

CT pulmonary angiogram

Answer

V/Q scan

Answer

Chest X-ray

Answer

MRI

Question 4

In a screening test for DM out of 1000 population, 90 were positive. When the gold standard test was applied to the entire population, 100 were found to have the disease. Assuming all 90 screening positives were confirmed as true positives by the gold standard, calculate the sensitivity.

Accepted Answer

True positives divided by total actual positives (90%)

Answer

All positives identified by the test assumed as true positives (100%)

Answer

Underestimated true positives divided by total actual positives (80%)

Answer

Total positives identified by the test divided by total actual positives (90%)

Question 5

HIV RNA by PCR can detect as low as

Accepted Answer

50 copies of viral RNA/ml of blood

Answer

60 copies of viral RNA/ml of blood

Answer

40 copies of viral RNA/ml of blood

Answer

30 copies of viral RNA/ml of blood

Question 6

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 7

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 8

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 9

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 10

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

Computer-Aided Detection and Diagnosis — MCQs

Computer-Aided Detection and Diagnosis — MCQs

Want unlimited practice?