โจทย์ ปรนัย
What is the primary goal of the article according to its introduction?

Generative AI is applied in the creation and analysis of medical images. This covers technological advancements, practical applications, and challenges encountered, providing readers with an overview.

From Saab K, Tu T, Weng W-H, et al. Capabilities of Gemini models in medicine

โจทย์ ปรนัย
How do generative AI models differ from traditional discriminative models in healthcare applications?

Generative AI is designed to generate new data similar to the training data, such as creating simulated X-ray images for training or synthesizing new drug molecular structures.

Observe from Khosravi B, Li F, Dapamede T, et al. Synthetically enhanced: unveiling synthetic data’s potential in medical imaging research.

โจทย์ ปรนัย
What is meant by the term “model as a dataset”?

Using trained models (such as generative models) as a data repository instead of directly using raw data ensures security and privacy. These models can learn statistical patterns from real data and generate new data with similar characteristics to the original data.

From Koetzier LR, Wu J, Mastrodicasa D, et al. Generating synthetic data for medical imaging.

โจทย์ ปรนัย
Which statement correctly distinguishes physics-informed and statistical models?

Physics-Informed Models are designed to incorporate domain knowledge, such as physics laws or biological principles, into the model's structure to improve prediction accuracy, unlike typical statistical models that focus solely on finding relationships from data.

Rouzrokh P, Khosravi B, Faghani S, Moassefi M, Vahdati S, Erickson BJ. Multitask brain tumor inpainting with diffusion models: a methodological report.

โจทย์ ปรนัย
According to the article, what does the “image generation trilemma” describe?

GANs (Generating Animated Models) often excel in speed and quality but lack variability (mode collapse), while Diffusion Models tend to prioritize quality and variability but take considerably longer to generate images (sampling speed).

Xiao Z, Kreis K, Vahdat A. Tackling the generative learning trilemma with denoising diffusion GANs.

โจทย์ ปรนัย
What is the Human Turing Test used for in medical image synthesis?

In medical image synthesis, the Human Turing Test is used to evaluate the quality of AI-generated images. Real and synthetic images are combined, and experts (such as radiologists) examine and differentiate between real and fake images,if the experts cannot accurately distinguish between the real and fake images, the algorithm has generated an image with a very high degree of realism, comparable to the actual data.

Khosravi B, Rouzrokh P, Mickley JP, et al. Few-shot biomedical image segmentation using diffusion models: beyond image generation.

โจทย์ ปรนัย
Which of the following is NOT mentioned as a potential benefit of synthetic data in healthcare?

Simulation data facilitates multi-center collaboration without the need for data exchange.

From Data Simulation Concepts.

โจทย์ ปรนัย
What is one major ethical concern associated with generative AI in medical imaging?

The problem with generative AI in medicine is data privacy. Even though the data used to train the AI ​​is anonymized, the AI ​​model may recognize unique characteristics of the original image. If the data is accidentally copied, it could be traced back to the actual patient who owned the image.

Ktena I, Wiles O, Albuquerque I, et al. Generative models improve fairness of medical classifiers under distribution shifts.

โจทย์ ปรนัย
What regulatory precedent did the article cite for synthetic data technologies?

The FDA has established a legal precedent by approving software that uses Synthetic MRI, or synthesized data for medical image processing, considering this technology as SaMD that enhances image quality.

Wolleb J, Bieder F, Sandkühler R, Cattin PC. Diffusion models for medical anomaly detection. In: Wang L, Dou Q, Fletcher PT, Speidel S, Li S, eds. Medical image computing and computer assisted intervention.

โจทย์ ปรนัย
What is the main purpose of the article?

The primary objective is to compare and evaluate cardiovascular disease risk prediction models (ASCVD), with a focus on East Asia. This is typically a study to determine whether standard models used in the West can be applied to Asian populations.

Zhao D. Epidemiological features of cardiovascular disease in Asia.

โจทย์ ปรนัย
Which of the following models was originally developed for a Western population?

The Framingham Risk Score is a model developed from the Framingham Heart Study, a population study project in Framingham.

U.S. Census Bureau. 2019 American Community Survey.

โจทย์ ปรนัย
Why might Western-based risk prediction models overestimate ASCVD risk in East Asian populations?

Risk models developed in the West are often based on demographic data showing higher rates of cardiovascular disease (ASCVD) than in East Asia.

Volgman AS, Palaniappan LS, Aggarwal NT, et al. Atherosclerotic cardiovascular disease in South Asians in the United States: Epidemiology, risk factors, and treatments: A Scientific Statement From the American Heart Association.

โจทย์ ปรนัย
What is the key advantage of the China-PAR model compared to Western-based models?

The main advantage of the China-PAR (Prediction for ASCVD Risk in China) model is that it utilizes large datasets specifically from the Chinese population, which is more comprehensive and allows for more accurate assessment of cardiovascular disease (ASCVD) risk than Western models.

Joint Committee for Guideline Revision. 2016 Chinese guidelines for the management of dyslipidemia in adults.

โจทย์ ปรนัย
Which of the following variables is not typically included in ASCVD risk prediction models discussed in the article?

โจทย์ ปรนัย
What is a major difference between the Suita Score and the Framingham Risk Score?

The Framingham Risk Score (FRS) is based on demographic data from the United States, which is predominantly white. This can sometimes lead to overestimation of risk when applied to Asian populations. The Suita Score, however, was created using data from epidemiological studies in Japan to provide greater accuracy specifically for the Japanese population, considering risk factors appropriate for the local context.

Nippon Data Research Group. Risk assessment chart for death from cardiovascular disease based on a 19-year follow-up study of a Japanese representative population NIPPON DATA80.

โจทย์ ปรนัย
According to the article, what is a potential benefit of developing East Asia–specific risk models?

East Asia-specific risk models help to make disease predictions more accurate because they are adapted to the local population, which often differs from standard models developed for Western populations. This reduces the likelihood of overestimating the risk.

Qi Y, Fan J, Liu J, et al. Cholesterol-overloaded HDL particles are independently associated with progression of carotid atherosclerosis in a cardiovascular disease-free population: a communitybased cohort study.

โจทย์ ปรนัย
Which factor was highlighted as influencing ASCVD risk differences among East Asian countries?

Even though they are all countries in the East Asian region, cultural differences, dietary habits , and lifestyles are major factors that contribute to varying risks of cardiovascular and ASCVD from country to country.

Qi Y, Fan J, Liu J, et al. Cholesterol-overloaded HDL particles are independently associated with progression of carotid atherosclerosis in a cardiovascular disease-free population: a communitybased cohort study.

โจทย์ ปรนัย
What future direction does the article suggest for improving ASCVD risk prediction?

Focusing on integrating complex and multimodal data with artificial intelligence (AI) will enable more accurate risk predictions.

Seyyed-Kalantari L, Zhang H, McDermott MBA, Chen IY, Ghassemi M. Underdiagnosis bias of artificial intelligence algorithms applied to chest radiographs in under-served patient populations.

โจทย์ ปรนัย
Which statement best explains the key difference in how VAEs, GANs, and DDPMs generate medical images according to the figure?

DDPMs (C) Use a reverse diffusion process to gradually remove noise from the image step by step until a perfect image is obtained, which is clearly different from the first two structures.

โจทย์ ปรนัย
Which of the following best explains the trend shown in Figure comparing age-standardized and crude CVD mortality rates among East Asian countries?

Graph A shows that Japan has the lowest death rate, which is a comparison after adjusting for differences in population age structure. Even though Japan has a large elderly population, as shown in Graph B, Japan still has a low death rate compared to other countries.

Hirooka N, Takedai T, D’Amico F. Assessing physical activity in daily life, exercise, and sedentary behavior among Japanese moving to westernized environment: a cross-sectional study of Japanese migrants at an urban primary care center in Pittsburgh.