KKKangHyun Kim

KangHyun Kim

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


KangHyun Kim has a total of 11 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Real-world implementation and predictive performance of suicide prediction models: A systematic review and meta-analysis

    by KangHyun Kim, JuHee Kim, Myung-Gwan Kim, Hyun Wook Han

    Background: Machine learning (ML) models for suicide risk prediction show promise, but their real-world discriminative ability and the sources of performance variation remain unclear. Methods: We systematically searched databases for studies evaluating statistical or ML suicide prediction models in real-world clinical settings. Random-effects meta-analyses pooled the area under the receiver operating characteristic curve (AUC) overall and by outcome type. We conducted exploratory univariable meta-regression with cluster-robust inference for outcome type, evaluation timing, and model type. Sensitivity analyses assessed the influence of within-study dependence. Results: Nine studies (20 evaluations) were included. The overall pooled AUC was 0.849 (95% CI 0.827–0.869). Pooled AUCs were 0.865 (0.840–0.887) for suicide attempts, 0.835 (0.825–0.844) for suicidal ideation, and 0.842 (0.797–0.878) for suicide death. Heterogeneity was extreme (I 2 = 99.9%). Sensitivity analyses yielded similar pooled estimates. Exploratory univariable meta-regression showed no clear associations for outcome type, evaluation timing, or model type. Conclusions: Suicide prediction models show good real-world discrimination, but performance varies substantially across evaluations. We found no clear evidence that outcome type, evaluation timing, or model type explained this heterogeneity. These findings support rigorous local validation and workflow-sensitive implementation rather than assuming that more complex algorithms will perform better in practice.


  • Open Access

    KODA: A Blockchain-Based Healthcare Data Marketplace (Preprint)

    by KangHyun Kim, Sung Min Kim, YoungMin Park, EunSol Lee, SungJae Jung, Jeongyong Kang, DongUk An, Kyungil Min, Hyeong Won Yu, Hyun Wook Han

    BACKGROUND A large amount of healthcare data has been produced and accumulated in recent years: big data derived from hospital records, patient medical records, medical examinations, and other sources. These data require proper management and analysis to derive meaningful results to improve healthcare. For proper management, the data must be integrated, secured, and made interoperable between individuals, hospitals, and companies, while protecting patient privacy. OBJECTIVE The purpose of this research was to create a healthcare data marketplace using a blockchain-based business-to-consumer model that ensured the integration, security, and interoperability of healthcare data between individuals, hospitals, companies, and marketplaces. This marketplace should enable direct transactions of healthcare data between companies that need individual healthcare data and individuals who own the data. METHODS To investigate the essential elements of a healthcare data marketplace, we considered two aspects: technology and service. A healthcare data marketplace was created using MySQL 8.0, JavaScript library, and Node.js software. The marketplace system featured security of data, access for companies and individuals, and a data transaction monitor. Furthermore, anonymization and standardization of data were implemented. A web-based application was developed for implementation of the marketplace system, along with a mobile application system for use by individuals. Finally, we evaluated the performance of the data marketplace system in three scenarios: male and obese (BMI ≥30 kg/m2); male aged ≥40 years with abnormal liver function tests (aspartate aminotransferase ≥40 units/L and alanine aminotransferase ≥40 units/L); and female aged ≥60 years with high blood pressure (systolic pressure ≥130 mmHg and diastolic pressure ≥90 mmHg). RESULTS A web-based application was developed to implement the healthcare data marketplace; a mobile application was also developed for individuals who provide their healthcare data. The performance of the data marketplace was evaluated through the three scenarios by calculating the time taken for completion of each scenario. A data query was executed fully within about 1–2 s; approximately 9.5 healthcare data queries were processed per minute in each scenario. CONCLUSIONS We successfully created a prototype of a healthcare data marketplace that ensured the integrity and security of data through its honest broker design.