Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Kyoung Min Moon has a total of 126 co-authors. Here are the top co-authors by number of shared publications:

  • Yang Deok Lee

    Department of Internal Medicine, Eulji University School of Medicine, Daejeon, Korea

    4 shared publications
  • Hyun-Soo Choi

    Department of Computer Science and Engineering, Seoul National University of Science and Technology, Seoul 01811, Republic of Korea

    3 shared publications
  • Kyueng-Whan Min

    Department of Pathology, Uijeongbu Eulji Medical Center, Eulji University School of Medicine, 712, Dongil-ro, Uijeongbu-si 11759, Republic of Korea

    3 shared publications
  • Min Soo Han

    Department of Internal Medicine, Eulji University School of Medicine, Daejeon, Korea

    3 shared publications
  • Yongseon Cho

    Department of Internal Medicine, Eulji University School of Medicine, Daejeon, Korea

    3 shared publications
  • Byoung Kwan Son

    Department of Internal Medicine, Uijeongbu Eulji Medical Center Eulji University School of Medicine Uijeongbu Republic of Korea

    2 shared publications

Latest Publications

  • Comparative Safety of Potassium-Competitive Acid Blockers on the Risk of Clostridioides difficile Infection: A Multicenter Cohort Study

    by Gyeongmin Lim, Ju-Young Shin, Da Eun Hyeon, Jie-Hyun Kim, Yuna Kim, Hyun-Soo Kim, Woo Hyeon Yoo, Byeong Jun Song, Chang Won Jeong, Hang A. Park, et al.

    INTRODUCTION: Proton pump inhibitors (PPIs) have been associated with an increased risk of Clostridioides difficile infection (CDI) in previous observational studies. However, real-world evidence on the CDI risk of potassium-competitive acid blockers (P-CABs), a newer class of acid suppressive agents, remains limited. METHODS: We conducted a retrospective cohort study using electronic health record data from 15 hospitals standardized to the Observational Medical Outcomes Partnership Common Data Model and integrated through Federated E-Health Big Data for Evidence Renovation Network. Patients newly prescribed P-CABs, proton pump inhibitors (PPIs), or histamine 2 receptor antagonists (H2RAs) were included and the primary outcome was the occurrence of CDI. After 1:4 propensity score matching, pairwise comparisons were conducted using Cox proportional hazards models. Hazard ratios (HRs) and 95% confidence intervals (CIs) were then pooled across databases by meta-analysis. RESULTS: In the 1:4 matched cohorts, we identified 61,596 P-CAB users (mean follow-up, 72.6 ± 29.9 days) and 132,683 PPI users (74.9 ± 28.6 days), 42,794 P-CAB users (69.5 ± 32.4 days) and 89,415 H2RA users (69.9 ± 33.1 days), and 85,686 PPI users (71.1 ± 31.2 days) and 159,178 H2RA users (68.0 ± 33.8 days) across the 3 pairwise comparisons. The pooled HRs were 0.87 (95% CI, 0.69–1.10) for P-CABs vs PPIs, 0.85 (0.64–1.12) for P-CABs vs H2RAs, and 1.13 (0.98–1.29) for PPIs vs H2RAs. Results were generally consistent across subgroup and sensitivity analyses. DISCUSSION: In this cohort study, P-CAB use was not associated with a higher risk of CDI compared with PPIs or H2RAs.


  • TUBA: AI-Assisted Nasogastric Tube Placement Assessment System

    by GwiSeong Moon, Kyoung Min Moon, Inseo Park, Kanghee Lee, Doohee Lee, Woo Jin Kim, Yoon Kim, Hyun-Soo Choi

  • Open Access

    Association Between Urban Green Space and Acute Exacerbations of COPD in Korea: A Nationwide Study Using the NHIS-NSC Cohort

    by Hae In Jung, Ju Won Lee, Hyochan Kim, Hoyoung Cha, Jongjin Baik, Kyoung Min Moon, Changhyun Jun, Sun-Young Jung, Kang-Mo Gu

    Background: Chronic obstructive pulmonary disease (COPD) is a major health concern in Korea, with a higher burden of acute exacerbations (AE-COPD) compared to Western populations. Environmental exposures such as smoking and air pollution are known contributors, but the impact of urban green space remains underexplored. Methods: We conducted a cohort study using the Korean National Health Insurance Service–National Sample Cohort (2006– 2019), including 5,171 patients aged ≥ 40 years with at least two COPD-related prescriptions within one year. Urban green space exposure was defined as the proportion of designated park area to total district area (2017 KOSIS data) and categorized into quartiles. Cox proportional hazards models estimated associations with AE-COPD and all-cause mortality, adjusting for demographic and clinical factors. Subgroup analyses were conducted by age, sex, income, comorbidities, BMI, smoking, and physical activity. Results: Among 5,171 COPD patients (mean age, 67.5 years; 60.7% male), 1,431 AE-COPD events occurred over 40,486 person-years. AE-COPD incidence declined from 35.4 to 31.3 per 1,000 person-years across green space quartiles. Compared to the lowest quartile, the highest quartile showed a lower AE-COPD risk (adjusted hazard ratio [aHR], 0.75; 95% CI, 0.58– 0.96; p for trend = 0.016). Stronger trends were observed in younger adults, men, high-income individuals, and those with comorbidities, though interaction tests were not significant. In a health screening subgroup (n = 3,318), patterns were consistent. No significant association was found with all-cause mortality. Conclusion: Greater urban green space coverage may be associated with reduced AE-COPD risk. However, results should be interpreted with cautiously given model limitations and exploratory nature of subgroup findings.


  • Open Access

    Deep learning-based cough classification using application-recorded sounds: a transfer learning approach with VGGish

    by Sanghoon Han, Yu-Rim Lee, Ji-Ho Lee, JinHee Jeon, Choongki Min, Kyungnam Kim, Donghoon Kim, Myung Pyo Kim, Young Mi Park, Uiri An, et al.

    Coughing sounds contain various bio-metric information with regards to respiratory diseases that can help in the assessment of respiratory diseases. While clinicians find coughs insightful, non-experts struggle to identify abnormalities in cough sounds. Furthermore, respiratory diseases has characterized by widespread health complications and elevated mortality rates, the development of early diagnostic systems is imperative for ensuring timely intervention and improving outcomes for both clinicians and patients. Accordingly, we propose a deep learning–based model for early diagnosis. To enhance the reliability of the training data, we utilized annotations provided by multiple medical specialists. Additionally, we examined how clinical expertise and diagnostic input influence the model’s generalization performance. This study introduces a deep learning framework utilizing VGGish as a transfer learning model, enhanced with additional detection and classification networks. The detection model identifies cough events within recorded audio, and then the classification model determines whether a detected cough is normal or abnormal. Both models were trained on raw cough sound data collected via smartphones and labeled by medical experts through a rigorous inspection process. Experimental evaluations demonstrated that the cough detection model achieved an average accuracy of 0.9883, while the cough classification model attained accuracies of 0.8417, 0.8629, and 0.8662 among dataset1, 2, and 3. To enhance interpretability, we applied Grad-CAM to visualize the features that influenced the model’s decision-making. Model performance was further evaluated using the area under the receiver operating characteristic curve (AUROC) and the area under the precision-recall curve (AUPRC). Our proposed cough classification model has the potential to assist individuals with limited access to healthcare as well as medical professionals with limited experience in diagnosing cough-related conditions. By leveraging deep learning and smartphone-recorded cough sounds, this approach aims to enhance early detection and management of respiratory diseases.


  • Open Access

    Deep learning based dual stage model for accurate nasogastric tube positioning in chest radiographs

    by Inseo Park, Gwiseong Moon, Ji Young Hong, Jeongwon Heo, Hongseok Ko, Doohee Lee, Yoon Kim, Woo Jin Kim, Hyun-Soo Choi, Kyoung Min Moon

    Accurate placement of nasogastric tubes (NGTs) is crucial for ensuring patient safety and effective treatment. Traditional methods relying on manual inspection are susceptible to human error, highlighting the need for innovative solutions. This study introduces a deep-learning model that enhances the detection and analysis of NGT positioning in chest radiographs. By integrating advanced segmentation and classification techniques, the model leverages the nnU-Net framework for segmenting critical regions and the ResNet50 architecture, pre-trained with MedCLIP, for classifying NGT placement. Trained on 1799 chest radiographs, the model demonstrates remarkable performance, achieving a Dice Similarity Coefficient of 65.35% for segmentation and an Area Under the Curve of 99.72% for classification. These results underscore its ability to accurately distinguish between correct and incorrect placements, outperforming traditional approaches. This method not only enhances diagnostic precision but also has the potential to streamline clinical workflows and improve patient care. A functional prototype of the model is accessible at https://ngtube.ziovision.ai.


  • Open Access

    Prediction of immunochemotherapy response for diffuse large B‐cell lymphoma using artificial intelligence digital pathology

    by Jeong Hoon Lee, Ga‐Young Song, Jonghyun Lee, Sae‐Ryung Kang, Kyoung Min Moon, Yoo‐Duk Choi, Jeanne Shen, Myung‐Giun Noh, Deok‐Hwan Yang

    Diffuse large B‐cell lymphoma (DLBCL) is a heterogeneous and prevalent subtype of aggressive non‐Hodgkin lymphoma that poses diagnostic and prognostic challenges, particularly in predicting drug responsiveness. In this study, we used digital pathology and deep learning to predict responses to immunochemotherapy in patients with DLBCL. We retrospectively collected 251 slide images from 216 DLBCL patients treated with rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone (R‐CHOP), with their immunochemotherapy response labels. The digital pathology images were processed using contrastive learning for feature extraction. A multi‐modal prediction model was developed by integrating clinical data and pathology image features. Knowledge distillation was employed to mitigate overfitting on gigapixel histopathology images to create a model that predicts responses based solely on pathology images. Based on the importance derived from the attention mechanism of the model, we extracted histological features that were considered key textures associated with drug responsiveness. The multi‐modal prediction model achieved an impressive area under the ROC curve of 0.856, demonstrating significant associations with clinical variables such as Ann Arbor stage, International Prognostic Index, and bulky disease. Survival analyses indicated their effectiveness in predicting relapse‐free survival. External validation using TCGA datasets supported the model's ability to predict survival differences. Additionally, pathology‐based predictions show promise as independent prognostic indicators. Histopathological analysis identified centroblastic and immunoblastic features to be associated with treatment response, aligning with previous morphological classifications and highlighting the objectivity and reproducibility of artificial intelligence‐based diagnosis. This study introduces a novel approach that combines digital pathology and clinical data to predict the response to immunochemotherapy in patients with DLBCL. This model shows great promise as a diagnostic and prognostic tool for clinical management of DLBCL. Further research and genomic data integration hold the potential to enhance its impact on clinical practice, ultimately improving patient outcomes.


  • Open Access

    Ensemble Deep Learning Model to Predict Lymphovascular Invasion in Gastric Cancer

    by Jonghyun Lee, Seunghyun Cha, Jiwon Kim, Jung Joo Kim, Namkug Kim, Seong Gyu Jae Gal, Ju Han Kim, Jeong Hoon Lee, Yoo-Duk Choi, Sae-Ryung Kang, et al.

    Lymphovascular invasion (LVI) is one of the most important prognostic factors in gastric cancer as it indicates a higher likelihood of lymph node metastasis and poorer overall outcome for the patient. Despite its importance, the detection of LVI(+) in histopathology specimens of gastric cancer can be a challenging task for pathologists as invasion can be subtle and difficult to discern. Herein, we propose a deep learning-based LVI(+) detection method using H&E-stained whole-slide images. The ConViT model showed the best performance in terms of both AUROC and AURPC among the classification models (AUROC: 0.9796; AUPRC: 0.9648). The AUROC and AUPRC of YOLOX computed based on the augmented patch-level confidence score were slightly lower (AUROC: −0.0094; AUPRC: −0.0225) than those of the ConViT classification model. With weighted averaging of the patch-level confidence scores, the ensemble model exhibited the best AUROC, AUPRC, and F1 scores of 0.9880, 0.9769, and 0.9280, respectively. The proposed model is expected to contribute to precision medicine by potentially saving examination-related time and labor and reducing disagreements among pathologists.


  • Open Access

    Deep learning model to predict Epstein–Barr virus associated gastric cancer in histology

    by Yeojin Jeong, Cristina Eunbee Cho, Ji-Eon Kim, Jonghyun Lee, Namkug Kim, Woon Yong Jung, Joohon Sung, Ju Han Kim, Yoo Jin Lee, Jiyoon Jung, et al.

    The detection of Epstein–Barr virus (EBV) in gastric cancer patients is crucial for clinical decision making, as it is related with specific treatment responses and prognoses. Despite its importance, the limited medical resources preclude universal EBV testing. Herein, we propose a deep learning-based EBV prediction method from H&E-stained whole-slide images (WSI). Our model was developed using 319 H&E stained WSI (26 EBV positive; TCGA dataset) from the Cancer Genome Atlas, and 108 WSI (8 EBV positive; ISH dataset) from an independent institution. Our deep learning model, EBVNet consists of two sequential components: a tumor classifier and an EBV classifier. We visualized the learned representation by the classifiers using UMAP. We externally validated the model using 60 additional WSI (7 being EBV positive; HGH dataset). We compared the model’s performance with those of four pathologists. EBVNet achieved an AUPRC of 0.65, whereas the four pathologists yielded a mean AUPRC of 0.41. Moreover, EBVNet achieved an negative predictive value, sensitivity, specificity, precision, and F1-score of 0.98, 0.86, 0.92, 0.60, and 0.71, respectively. Our proposed model is expected to contribute to prescreen patients for confirmatory testing, potentially to save test-related cost and labor.


  • Open Access

    How Many Private Data Are Needed for Deep Learning in Lung Nodule Detection on CT Scans? A Retrospective Multicenter Study

    by Jeong Woo Son, Ji Young Hong, Yoon Kim, Woo Jin Kim, Dae-Yong Shin, Hyun-Soo Choi, So Hyeon Bak, Kyoung Min Moon

    Early detection of lung nodules is essential for preventing lung cancer. However, the number of radiologists who can diagnose lung nodules is limited, and considerable effort and time are required. To address this problem, researchers are investigating the automation of deep-learning-based lung nodule detection. However, deep learning requires large amounts of data, which can be difficult to collect. Therefore, data collection should be optimized to facilitate experiments at the beginning of lung nodule detection studies. We collected chest computed tomography scans from 515 patients with lung nodules from three hospitals and high-quality lung nodule annotations reviewed by radiologists. We conducted several experiments using the collected datasets and publicly available data from LUNA16. The object detection model, YOLOX was used in the lung nodule detection experiment. Similar or better performance was obtained when training the model with the collected data rather than LUNA16 with large amounts of data. We also show that weight transfer learning from pre-trained open data is very useful when it is difficult to collect large amounts of data. Good performance can otherwise be expected when reaching more than 100 patients. This study offers valuable insights for guiding data collection in lung nodules studies in the future.


  • Open Access

    Machine Learning Models for Predicting Mortality in 7472 Very Low Birth Weight Infants Using Data from a Nationwide Neonatal Network

    by Hyun Jeong Do, Kyoung Min Moon, Hyun-Seung Jin

    Statistical and analytical methods using artificial intelligence approaches such as machine learning (ML) are increasingly being applied to the field of pediatrics, particularly to neonatology. This study compared the representative ML analysis and the logistic regression (LR), which is a traditional statistical analysis method, using them to predict mortality of very low birth weight infants (VLBWI). We included 7472 VLBWI data from a nationwide Korean neonatal network. Eleven predictor variables (neonatal factors: male sex, gestational age, 5 min Apgar scores, body temperature, and resuscitation at birth; maternal factors: diabetes mellitus, hypertension, chorioamnionitis, premature rupture of membranes, antenatal steroid, and cesarean delivery) were selected based on clinical impact and statistical analysis. We compared the predicted mortality between ML methods—such as artificial neural network (ANN), random forest (RF), and support vector machine (SVM)—and LR with a randomly selected training set (80%) and a test set (20%). The model performances of area under the receiver operating curve (95% confidence interval) equaled LR 0.841 (0.811–0.872), ANN 0.845 (0.815–0.875), and RF 0.826 (0.795–0.858). The exception was SVM 0.631 (0.578–0.683). No statistically significant differences were observed between the performance of LR, ANN, and RF (i.e., p > 0.05). However, the SVM model was lower (p < 0.01). We suggest that VLBWI mortality prediction using ML methods would yield the same prediction rate as the traditional statistical LR method and may be suitable for predicting mortality. However, low prediction rates are observed in certain ML methods; hence, further research is needed on these limitations and selecting an appropriate method.