Background: Despite excellent prediction performance, noninterpretability has undermined the value of applying deep-learning algorithms in clinical practice. To overcome this limitation, attention mechanism has been introduced to clinical research as an explanatory modeling method. However, potential limitations of using this attractive method have not been clarified to clinical researchers. Furthermore, there has been a lack of introductory information explaining attention mechanisms to clinical researchers.Objective: The aim of this study was to introduce the basic concepts and design approaches of attention mechanisms. In addition, we aimed to empirically assess the potential limitations of current attention mechanisms in terms of prediction and interpretability performance.Methods: First, the basic concepts and several key considerations regarding attention mechanisms were identified. Second, four approaches to attention mechanisms were suggested according to a two-dimensional framework based on the degrees of freedom and uncertainty awareness. Third, the prediction performance, probability reliability, concentration of variable importance, consistency of attention results, and generalizability of attention results to conventional statistics were assessed in the diabetic classification modeling setting. Fourth, the potential limitations of attention mechanisms were considered.Results: Prediction performance was very high for all models. Probability reliability was high in models with uncertainty awareness. Variable importance was concentrated in several variables when uncertainty awareness was not considered. The consistency of attention results was high when uncertainty awareness was considered. The generalizability of attention results to conventional statistics was poor regardless of the modeling approach.Conclusions: The attention mechanism is an attractive technique with potential to be very promising in the future. However, it may not yet be desirable to rely on this method to assess variable importance in clinical settings. Therefore, along with theoretical studies enhancing attention mechanisms, more empirical studies investigating potential limitations should be encouraged.

Background: Synthetic data generation (SDG) based on generative adversarial networks (GANs) is used in health care, but research on preserving data with logical relationships with synthetic tabular data (STD) remains challenging. Filtering methods for SDG can lead to the loss of important information. Objective: This study proposed a divide-and-conquer (DC) method to generate STD based on the GAN algorithm, while preserving data with logical relationships. Methods: The proposed method was evaluated on data from the Korea Association for Lung Cancer Registry (KALC-R) and 2 benchmark data sets (breast cancer and diabetes). The DC-based SDG strategy comprises 3 steps: (1) We used 2 different partitioning methods (the class-specific criterion distinguished between survival and death groups, while the Cramer V criterion identified the highest correlation between columns in the original data); (2) the entire data set was divided into a number of subsets, which were then used as input for the conditional tabular generative adversarial network and the copula generative adversarial network to generate synthetic data; and (3) the generated synthetic data were consolidated into a single entity. For validation, we compared DC-based SDG and conditional sampling (CS)–based SDG through the performances of machine learning models. In addition, we generated imbalanced and balanced synthetic data for each of the 3 data sets and compared their performance using 4 classifiers: decision tree (DT), random forest (RF), Extreme Gradient Boosting (XGBoost), and light gradient-boosting machine (LGBM) models. Results: The synthetic data of the 3 diseases (non–small cell lung cancer [NSCLC], breast cancer, and diabetes) generated by our proposed model outperformed the 4 classifiers (DT, RF, XGBoost, and LGBM). The CS- versus DC-based model performances were compared using the mean area under the curve (SD) values: 74.87 (SD 0.77) versus 63.87 (SD 2.02) for NSCLC, 73.31 (SD 1.11) versus 67.96 (SD 2.15) for breast cancer, and 61.57 (SD 0.09) versus 60.08 (SD 0.17) for diabetes (DT); 85.61 (SD 0.29) versus 79.01 (SD 1.20) for NSCLC, 78.05 (SD 1.59) versus 73.48 (SD 4.73) for breast cancer, and 59.98 (SD 0.24) versus 58.55 (SD 0.17) for diabetes (RF); 85.20 (SD 0.82) versus 76.42 (SD 0.93) for NSCLC, 77.86 (SD 2.27) versus 68.32 (SD 2.37) for breast cancer, and 60.18 (SD 0.20) versus 58.98 (SD 0.29) for diabetes (XGBoost); and 85.14 (SD 0.77) versus 77.62 (SD 1.85) for NSCLC, 78.16 (SD 1.52) versus 70.02 (SD 2.17) for breast cancer, and 61.75 (SD 0.13) versus 61.12 (SD 0.23) for diabetes (LGBM). In addition, we found that balanced synthetic data performed better. Conclusions: This study is the first attempt to generate and validate STD based on a DC approach and shows improved performance using STD. The necessity for balanced SDG was also demonstrated.

A screening model for undiagnosed diabetes mellitus (DM) is important for early medical care. Insufficient research has been carried out developing a screening model for undiagnosed DM using machine learning techniques. Thus, the primary objective of this study was to develop a screening model for patients with undiagnosed DM using a deep neural network. We conducted a cross-sectional study using data from the Korean National Health and Nutrition Examination Survey (KNHANES) 2013–2016. A total of 11,456 participants were selected, excluding those with diagnosed DM, an age < 20 years, or missing data. KNHANES 2013–2015 was used as a training dataset and analyzed to develop a deep learning model (DLM) for undiagnosed DM. The DLM was evaluated with 4444 participants who were surveyed in the 2016 KNHANES. The DLM was constructed using seven non-invasive variables (NIV): age, waist circumference, body mass index, gender, smoking status, hypertension, and family history of diabetes. The model showed an appropriate performance (area under curve (AUC): 80.11) compared with existing previous screening models. The DLM developed in this study for patients with undiagnosed diabetes could contribute to early medical care.

In healthcare, the most common type of data is tabular data, which holds high significance and potential in the field of medical AI. However, privacy concerns have hindered their widespread use. Despite the emergence of synthetic data as a viable solution, the generation of healthcare tabular data (HTD) is complex owing to the extensive interdependencies between the variables within each record that incorporate diverse clinical characteristics, including sensitive information. To overcome these issues, this study proposed a tabular transformer generative adversarial network (TT-GAN) to generate synthetic data that can effectively consider the relationships between variables potentially present in the HTD dataset. Transformers can consider the relationships between the columns in each record using a multi-attention mechanism. In addition, to address the potential risk of restoring sensitive data in patient information, a Transformer was employed in a generative adversarial network (GAN) architecture, to ensure an implicit-based algorithm. To consider the heterogeneous characteristics of the continuous variables in the HTD dataset, the discretization and converter methodology were applied. The experimental results confirmed the superior performance of the TT-GAN than the Conditional Tabular GAN (CTGAN) and copula GAN. Discretization and converters were proven to be effective using our proposed Transformer algorithm. However, the application of the same methodology to Transformer-based models without discretization and converters exhibited a significantly inferior performance. The CTGAN and copula GAN indicated minimal effectiveness with discretization and converter methodologies. Thus, the TT-GAN exhibited considerable potential in healthcare, demonstrating its ability to generate artificial data that closely resembled real healthcare datasets. The ability of the algorithm to handle different types of mixed variables efficiently, including polynomial, discrete, and continuous variables, demonstrated its versatility and practicality in health care research and data synthesis.

Multivessel disease (MVD) is an independent risk factor for poor prognosis in acute myocardial infarction patients. Although several global risk scoring systems (RSS) are in use in clinical practice, there is no dedicated RSS for MVD in ST-segment elevation myocardial infarction (STEMI). The primary objective of this study is to develop a novel RSS to estimate the prognosis of patients with MVD in STEMI. We used the Korean Acute Myocardial Infarction Registry (KAMIR) to identify 2,030 STEMI patients with MVD who underwent appropriate percutaneous coronary intervention (PCI). Their data were analyzed to develop a new RSS. The prognostic power of this RSS was validated with 2,556 STEMI patients with MVD in the Korean Working Group on Myocardial Infarction Registry (KORMI). Six prognostic factors related to all-cause death in STEMI patients with MVD were age, serum creatinine, Killip Class, lower body weight, decrease in left ventricular ejection fraction, and history of cerebrovascular disease. The RSS for all-cause death was constructed using these risk factors and their statistical weight. The RSS had appropriate performance (c-index: 0.72) in the KORMI validation cohort. We developed a novel RSS that estimates all-cause death in the year following discharge for patients with MVD in STEMI appropriately treated by PCI. This novel RSS was transformed into a simple linear risk score to yield a simplified estimate prognosis of MVD among STEMI patients.