| 1 |
What is the primary goal of the article according to its introduction?
|
3. To explore advancements, applications, and challenges of generative AI in medical imaging |
|
The article’s introduction and abstract clearly state that its main goal is to review generative AI models for medical imaging.
Other choices are incorrect because it does not focus on hospital management, economic impacts, AI policies, or creating new models.
|
Generative AI models learn data patterns to create new images or data, which can be applied to medical imaging for education, surgery, and training purposes.
|
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 2 |
How do generative AI models differ from traditional discriminative models in healthcare applications?
|
2. Generative models produce new data rather than only classify or interpret |
|
The paper mention, that generative models learn data patterns to create new synthetic images or signals, while discriminative models only analyze or classify existing data. Other options are incorrect because generative models do not interpret, do not always require manual labeling, and can handle multimodal inputs.
|
Generative AI models: learn the patterns of medical data to create synthetic images or signals
discriminative models : only analyze, classify, or interpret existing data.
Generative artificial intelligence is a class of deep learning models capable of creating content that diverges from traditional discriminative models focused on interpretation or decision making.
|
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 3 |
What is meant by the term “model as a dataset”?
|
3. Sharing trained model weights instead of raw data |
|
Model as a dataset means distributing the trained model itself so others can use its knowledge without accessing sensitive raw patient data.
|
The paper mention that, The advancement of generative artificial intelligence introduces a new concept in data sharing, which we refer to as a model as a dataset. In this concept, generative models learn and store patterns and characteristics of the original data in their internal parameters (weights).
|
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 4 |
Which statement correctly distinguishes physics-informed and statistical models?
|
3. Physics-informed models incorporate biological or physical principles |
|
The question ask that which choice distinguishes physics-informed and statistical models, the only correct answer is 3, because other are false.
|
physics-informed : primarily rule-based approaches that incorporate domain-specific knowledge and physics principles through mathematical equations and explicit constraints to generate realistic and physically plausible data.
statistical models : learn from data patterns and distributions
|
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 5 |
According to the article, what does the “image generation trilemma” describe?
|
2. Trade-offs among image diversity, quality, and speed |
|
The article defines the “image generation trilemma” as the difficulty of achieving high diversity, high quality, and fast generation at the same time.
|
Generative AI models must balance sample quality (realism), diversity (coverage of variations), and speed (computational efficiency).
|
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 6 |
What is the Human Turing Test used for in medical image synthesis?
|
2. To assess realism of synthetic medical images by experts |
|
The Human Turing Test evaluates whether experts, like radiologists, can distinguish real images from AI-generated ones.
|
If experts cannot tell synthetic images from real ones, the generative model is producing highly realistic medical images, which can be use in education and others medical purposes.
|
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 7 |
Which of the following is NOT mentioned as a potential benefit of synthetic data in healthcare?
|
4. Eliminating all medical biases permanently |
|
The article mentions only that synthetic data can increase diversity, protect privacy, support education, and enable collaborations.
The article never mention that they gonna remove all existing bias.
|
Eliminating all medical biases permanently, would cause sampling bias and error. They should not remove the real data.
|
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 8 |
What is one major ethical concern associated with generative AI in medical imaging?
|
2. Data copying and patient reidentification |
|
AI might memorize real patient data and reproduce it in synthetic images, risking privacy breaches.
|
Even if images are synthetic, sensitive patient information could be revealed.
A lot of people said that using even using synthetic image of patient, it is the same as the real image being reveal to other people.
|
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 9 |
What regulatory precedent did the article cite for synthetic data technologies?
|
2. FDA clearance of synthetic MRI as image-processing software |
|
FDA approval shows that synthetic medical images can meet regulatory standards.
|
Article mention Regulatory precdent provides a framework for safely integrating AI-generated images
|
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 10 |
What is the main purpose of the article?
|
2. To compare and evaluate ASCVD risk prediction models in East Asia |
|
The article focuses on how Western-developed ASCVD models perform in East Asian populations and the need for region-specific adjustments, because some of the results are over or under estimated.
|
Risk prediction models must reflect population-specific factors like baseline incidence, lifestyle, bodymass and other things.
|
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 11 |
Which of the following models was originally developed for a Western population?
|
1. Framingham Risk Score |
|
The Framingham Risk Score was developed using data from the Framingham Heart Study in the United States
|
Risk prediction models are population-specific, Framingham Risk Score are designed to fit only for US citizen only.
Us citizen may have more body mass or fat, which could cause more risk, if compare to asian.
|
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 12 |
Why might Western-based risk prediction models overestimate ASCVD risk in East Asian populations?
|
2. East Asians have lower baseline incidence of ASCVD |
|
Western models like the Framingham Risk Score assume a higher baseline risk, so applying them to East Asian populations overestimates individual risk.
|
Risk prediction accuracy depends on baseline incidence, bodymass and genetics mismatched models can lead to overtreatment or misclassification.
|
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 13 |
What is the key advantage of the China-PAR model compared to Western-based models?
|
4. It was calibrated using national data representing diverse regions in China |
|
The China-PAR model uses large-scale, nationally representative Chinese data, making its predictions more accurate for the local population.
|
Calibration with local epidemiological data ensures the model accounts for population-specific risk factors, lifestyles, and disease incidence,
|
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 14 |
Which of the following variables is not typically included in ASCVD risk prediction models discussed in the article?
|
4. Genetic ancestry markers |
|
Standard ASCVD models like Framingham, China-PAR include age, blood pressure, cholesterol, diabetes, and smoking, but never use genetic ancestry markers.
|
Risk models focus on modifiable and easily measurable clinical factors to predict ASCVD in patient, tracking Genetic ancestry markers would take generations of patients and a lot of time.
|
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 15 |
What is a major difference between the Suita Score and the Framingham Risk Score?
|
2. Suita Score was designed for a Japanese population using local epidemiological data |
|
The Suita Score is calibrated for the Japanese population, whereas the Framingham Risk Score was developed in a Western population.
|
Population-specific calibration ensures more accurate risk prediction by using differences in baseline incidence, lifestyle, and epidemiology.
|
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 16 |
According to the article, what is a potential benefit of developing East Asia–specific risk models?
|
3. They improve accuracy and reduce overestimation of risk |
|
East Asia–specific models like China-PAR follow Western-based models baseline causing, overestimation of risk that occurs when using Western-based models.
|
Risk models must reflect population-specific factors to have the most accurate to the population.
|
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 17 |
Which factor was highlighted as influencing ASCVD risk differences among East Asian countries?
|
2. Cultural and dietary variations, such as salt intake and lifestyle |
|
The article mention that differences in diet, lifestyle, and cultural habits influence ASCVD risk
|
Environmental and lifestyle factors influence cardiovascular health, so region-specific models must consider these to accurately predict risk.
|
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 18 |
What future direction does the article suggest for improving ASCVD risk prediction?
|
2. Using multimodal AI-based prediction integrated with regional data |
|
The article mention, to combining clinical, lifestyle, genetic, and imaging data with AI models that was created for regional populations to improve accuracy.
|
creating multiple data types with regional calibration makes predictions more accurate, reflecting true ASCVD risk for specific populations.
|
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 19 |
Which statement best explains the key difference in how VAEs, GANs, and DDPMs generate medical images according to the figure?
|
3. DDPMs iteratively remove noise through reverse diffusion rather than using encoder–decoder or discriminator structures. |
|
According to the article, VAEs use an encoder–decoder architecture, GANs use a generator and discriminator with adversarial feedback, and DDPMs generate images by gradually denoising random noise.
|
Different generative models have distinct mechanisms for different purpose such as quality, speed and divesity.
|
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 20 |
Which of the following best explains the trend shown in Figure comparing age-standardized and crude CVD mortality rates among East Asian countries?
|
3. Despite differences in age structures, Japan maintains low mortality rates in both measures, suggesting effective prevention and healthcare systems. |
|
Japan has lowest Age standardlized CVD mortality rate compare to other countries.
Japan has the second lowest mortality rate for Stroke and IHD.
|
Japan have both low age-standardized and crude CVD mortality rates in east asian, indicating a effective prevention and healthcare systems.
|
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|