| 1 |
How does the concept of “model as a dataset” reshape traditional data-sharing practices in medical imaging?
|
It enables sharing of learned model weights instead of sensitive raw images. |
|
By the definition of the model, the key information is shared instead of the entire thing. |
According to the passage, model as a data set enables the sharing of keys data the model learned, promoting a more convenient way of training. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 2 |
Which analytical conclusion can be drawn about the trade-offs between physics-informed and statistical models?
|
Physics-informed models always produce higher diversity. |
|
|
|
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 3 |
Why is “mode collapse” considered a critical problem in GAN-based medical image synthesis?
|
It reduces image realism and variety by producing repetitive outputs. |
|
The problem in data's domain can results in inaccuracy and quality of the image. |
According to the paper, with repeatedly trainning on GANs model can degrades the quality of the output leading to the fail of capturing data variations and mode coverage known as mode collapse. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 4 |
Why are healthcare-specific metrics preferred over general-purpose metrics such as FID or SSIM?
|
They better capture clinical accuracy and diagnostic relevance. |
|
because FID and SSIM is a prediction process based on the pre-trained image. |
|
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 5 |
What does the article identify as the key tension between privacy preservation and image fidelity?
|
Higher realism may risk reproducing identifiable patient data. |
|
Higher quality image may accidentally contains patient's identifications or unique features. |
One of the challenges mentioned is the anonymisation of the generated data which have potential of reidentification especially with higher realism, due to the fact that medical images can possibly contains patient's unique features without consent. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 6 |
Why is the FDA’s approval of synthetic MRI technology significant for future AI-generated data?
|
It establishes a framework for validating synthetic data equivalence in clinical use. |
|
With the approval it can leads to more further development in applications. |
From the conclusion of the paper, FDA approval can create more advancnments and research by using it diagnostic performance to proof read or post market surviellance commitments monitor for future clinical trials. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 7 |
Which strategy would best mitigate demographic bias in generative models according to the article?
|
Applying diversity-aware training and fairness constraints |
|
It is the only choice mentioned in the article. |
The paper suggested diversity aware sampling during training and including debiasing techniques for more non-bias results. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 8 |
How do DDPMs exemplify versatility in healthcare image synthesis?
|
They can perform multiple tasks such as denoising, inpainting, and anomaly detection without retraining. |
|
It is the only choice correctly defines the ability of DDPMS model. |
According to the passage this model is known for it's high quality image and its ability to access wide ranges of domain creating a diversity of capbilities. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 9 |
What analytical insight does the article provide about integrating AI-generated medical images into education and research?
|
It enhances training by providing diverse, realistic datasets without ethical breaches. |
|
It is the only choice mentioning the main idea of the article. |
The article includes a further application for the generated image by focussing on it strong points, such as the diversity and advancnment while also concerning about ethical issues. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 10 |
Why is regional calibration essential when applying risk prediction models across countries?
|
To adjust for population-specific incidence and lifestyle differences |
|
People living across the world have different ways of life which can affect their risk for the disease. |
The main problem mentioned creating the wrong estimation of predictions calculated is the factors not specific enough for the cohort resulting a inaccurate of information. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 11 |
What analytical conclusion can be drawn when comparing the China-PAR and Framingham models?
|
China-PAR uses local epidemiological data, leading to improved predictive validity. |
|
China-PAR model is mention to use local cohort and dataset too analyzing in order to develop a more specific and accurate way for chinese population. |
It is mentioned in the article that Framingham model uses large cohort which are more specific to western population, while the China-PAR model is develop from Chinese epidemiological data for Chinese citizen. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 12 |
Based on CVD mortality data, what analytical inference can be made about Japan’s position compared to neighboring countries?
|
Japan’s data cannot be compared internationally. |
|
|
|
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 13 |
What analytical limitation arises when using Western-derived coefficients in East Asian models?
|
It reduces model interpretability. |
|
Without the same lifestyle it can not use the same way of evaluation risk. |
With cultural and lifestyle differences it can results in inaccuracy of the predictions. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 14 |
What policy implication can be derived from country-specific risk models?
|
|
|
|
|
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 15 |
If a model excludes socioeconomic variables, what analytical consequence might occur?
|
Ignored non-biological determinants of disease |
|
Socioeconomic factor is one of the most crucial factor in analyzing to prediction score. |
According to the passage, without socioeconomic variables can make the data inaccurate due to the significant difference the factor capable. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 16 |
How might AI improve next-generation ASCVD risk prediction in East Asia?
|
By integrating multimodal data, including imaging and lifestyle informa |
|
It is the only choice mentioned in the article. |
The paper mentioned the problem of information sharing with the projections of using ai to lessen to gap of diversity cohorts to create more accurate risk culculations. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 17 |
What conclusion can be drawn from comparing Mongolia’s and South Korea’s CVD mortality rates?
|
|
|
|
|
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 18 |
What is the most logical future direction for improving ASCVD models across East Asia?
|
Establishing multinational data-sharing platforms to harmonize regional models |
|
By sharing the data o a platform can result in more knowledge being used for future developments. |
According to the article, the main problem is the lack of diversity of the information with multinational data sharing can result in a better way of risk predictions. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 19 |
According to the “image generation trilemma” shown in the figure, what analytical conclusion can be drawn about the relative strengths of VAEs, GANs, and DDPMs in medical image synthesis?
|
GANs provide a balance between image quality and diversity but may suffer from mode collapse. |
|
It is the only option correctly analyzing the diagram. |
According from the paperVAEs is known for its speed and wide domain in generating image, while GANs is famous for its high quality and speed despite it lack of diversity. On the other hand DDPMs is the best option due to it's diversity of domain ad the realism of the image which outweigh it slowness of the process. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 20 |
Based on Figure, what analytical conclusion can be drawn regarding the distribution of cardiovascular disease (CVD) subtypes across East Asian countries?
|
|
|
|
|
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|