Explainable predictions of a machine learning model to forecast the postoperative length of stay for severe patients: machine learning model development and evaluation.

Background: Predicting the length of stay in advance will not only benefit the hospitals both clinically and financially but enable healthcare providers to better decision-making for improved quality of care. More importantly, understanding the length of stay of severe patients who require general anesthesia is key to enhancing health outcomes.

Objective: Here, we aim to discover how machine learning can support resource allocation management and decision-making resulting from the length of stay prediction.

Association between heat and hospital admissions in people with disabilities in South Korea: a nationwide, case-crossover study.

Background: Despite extensive findings on the hazardous impacts of environmental heat exposure, little is known about the effect on people with disabilities. This study aimed to estimate the association between environmental heat exposure and emergency department admissions for people with disabilities compared with people without disabilities.

Methods: In this nationwide, case-crossover study, we linked data on emergency department admissions (cases) for any cause in the warm season in South Korea from the Korean National Health Insurance Service (NHIS)-National Sample Cohort database (a nationally representative database of 1 million systematically sampled beneficiaries covering all ages) from Jan 1, 2002, to Dec 31, 2019, and short-term daily mean temperature exposure (measured via Google Earth Engine at a 9 km spatial grid, aggregated to district).

The impact of depression, stress, and self-esteem on quality of life among older adults in South Korea.

Background & Objective: Aging is a global trend, and Korea is also entering an aging society, which threatens the mental health of the elderly due to isolation, etc. In line with the growing domestic and international interest in elderly issues, this study aimed to identify the effects of depression, stress and self-esteem on the lives of the elderly in South Korea and to provide basic data for welfare measures.

Methods: Depression, stress, self-esteem, and quality of life were measured in 104 South Korean seniors (32 men, 72 women, average age 72.

Are polypharmacy side effects predicted by public data still valid in real-world data?

Background And Objective: Although interest in predicting drug-drug interactions is growing, many predictions are not verified by real-world data. This study aimed to confirm whether predicted polypharmacy side effects using public data also occur in data from actual patients.

Methods: We utilized a deep learning-based polypharmacy side effects prediction model to identify cefpodoxime-chlorpheniramine-lung edema combination with a high prediction score and a significant patient population.

Biological response of nonhuman primates to controlled levels of acute blood loss.

Introduction: The global shortage of human blood for medical use has prompted the development of alternative blood sources. Nonhuman primates (NHPs) are commonly used owing to their physiological similarities to humans. The objective of the current study was to establish a controlled-blood-loss model in NHPs to explore their clinical and biological responses.

A Comparison Between ; and ; Pig Kidneys Transplanted in Nonhuman Primates.

Because there is a shortage of donor kidneys, researchers are exploring the possibility of using genetically modified pig kidneys for transplantation. Approaches involving knockout of carbohydrate genes or knockin of protective proteins have been attempted to determine the best gene modifications. In this study, we utilized ; and ; pigs for transplantation in nonhuman primates (NHPs).

Machine learning models to predict the warfarin discharge dosage using clinical information of inpatients from South Korea.

As warfarin has a narrow therapeutic window and obvious response variability among individuals, it is difficult to rapidly determine personalized warfarin dosage. Adverse drug events(ADE) resulting from warfarin overdose can be critical, so that typically physicians adjust the warfarin dosage through the INR monitoring twice a week when starting warfarin. Our study aimed to develop machine learning (ML) models that predicts the discharge dosage of warfarin as the initial warfarin dosage using clinical data derived from electronic medical records within 2 days of hospitalization.

Initial investigation on the feasibility of porcine red blood cells from genetically modified pigs as an alternative to human red blood cells for transfusion.

The decline in blood donation rates and the ongoing shortage of blood products pose significant challenges to medical societies. One potential solution is to use porcine red blood cells (pRBCs) from genetically modified pigs as an alternative to human red blood cells (hRBCs). However, adverse immunological reactions remain a significant obstacle to their use.

LDP-GAN : Generative adversarial networks with local differential privacy for patient medical records synthesis.

Electronic medical records(EMR) have considerable potential to advance healthcare technologies, including medical AI. Nevertheless, due to the privacy issues associated with the sharing of patient's personal information, it is difficult to sufficiently utilize them. Generative models based on deep learning can solve this problem by creating synthetic data similar to real patient data.

Comparative transcriptome analysis between long- and short-term survival after pig-to-monkey cardiac xenotransplantation reveals differential heart failure development.

Cardiac xenotransplantation is the potential treatment for end-stage heart failure, but the allogenic organ supply needs to catch up to clinical demand. Therefore, genetically-modified porcine heart xenotransplantation could be a potential alternative. So far, pig-to-monkey heart xenografts have been studied using multi-transgenic pigs, indicating various survival periods.

Recent trends in perioperative blood transfusion during elective kidney transplantation.

Background: Accurately predicting the demand for blood transfusions is crucial for blood banks. Given the potential for emergency situations, it is imperative that blood banks maintain a sufficient inventory of blood products. In this study, we examined the use of perioperative transfusions in patients undergoing elective kidney transplants.

Impact of the COVID-19 Pandemic on the Usage of Blood for Transfusions: A 2-Year Experience from a Tertiary Center in Korea.

The coronavirus disease (COVID-19) outbreak affected the utilization and management of blood products in hospitals. Blood shortages occurred owing to social distancing policies and reduction in blood donors. However, only a few studies examined whether these changes affected blood usage and transfusion patterns.

Clinical Applicability of Automated Hematology Analyzer Research Parameters for Disease Diagnosis: a Reference Interval Study.

Background: We established reference intervals for research parameters of complete blood cell count and examined their usefulness for diagnosing certain diseases.

Methods: Reference intervals for 26 basic and 38 research parameters were established for 3,457 and 1,325 men and 2,742 and 830 women aged 20 - 59 and ≥ 60 years, respectively. Research parameter values for patients with iron deficiency anemia (IDA), appendicitis, sepsis, and myelodysplastic syndromes (MDS) were compared against gender- and age-matched reference values.

Changes of Pepsin Concentration in Saliva Sample According to Storage Period.

Objectives: To determine whether the concentration of pepsin in the saliva sample changes according to the storage period of the sample.

Methods: Forty eight patients with suspected laryngopharyngeal reflux were included in this study. Saliva samples were collected from each patient and each sample divided into six and stored for different period of time.

Heterogeneous graph construction and HinSAGE learning from electronic medical records.

Graph representation learning is a method for introducing how to effectively construct and learn patient embeddings using electronic medical records. Adapting the integration will support and advance the previous methods to predict the prognosis of patients in network models. This study aims to address the challenge of implementing a complex and highly heterogeneous dataset, including the following: (1) demonstrating how to build a multi-attributed and multi-relational graph model (2) and applying a downstream disease prediction task of a patient's prognosis using the HinSAGE algorithm.

Visceral obesity as a risk factor of incisional hernia after single-port laparoscopic gynecologic surgery.

Objective: We aimed to evaluate associations between abdominal fat distribution (AFD) parameters and incisional hernia (IH) in patients who underwent transumbilical single-port laparoscopic surgery (SPLS) for gynecological disease.

Methods: Medical records of 2116 patients who underwent SPLS for gynecological disease at Daejeon St. Mary's Hospital between March 2014 and February 2021 were reviewed.

Treatment strategies for Lennox-Gastaut syndrome: outcomes of multimodal treatment approaches.

Background: Multimodal treatment approaches are often considered for patients with Lennox-Gastaut syndrome (LGS). Creating an algorithm that can guide healthcare providers in selecting treatment options for patients with LGS remains a challenge. Herein, we assessed the long-term seizure-free and neurodevelopmental outcomes of stepwise multimodal treatment in patients with LGS.

RIDAB: Electronic medical record-integrated real world data platform for predicting and summarizing interactions in biomedical research from heterogeneous data resources.

Background And Objective: With the advent of bioinformatics, biological databases have been constructed to computerize data. Biological systems can be described as interactions and relationships between elements constituting the systems, and they are organized in various biomedical open databases. These open databases have been used in approaches to predict functional interactions such as protein-protein interactions (PPI), drug-drug interactions (DDI) and disease-disease relationships (DDR).

Relationship between metabolic syndrome and follicle-stimulating hormone in postmenopausal women.

Depletion of ovarian reserve during menopausal transition raises follicle-stimulating hormone (FSH) markedly and menopause is related to an increased risk for metabolic syndrome (MetS). This study examined the relationship between FSH and MetS in postmenopausal women.We evaluated the anthropometric values, lipid profiles, high-sensitivity C-reactive protein (hs-CRP) level, Homeostasis model assessment for insulin resistance (HOMA-IR), and serum adipokines levels in 219 postmenopausal women.

Effect of Factor H on Complement Alternative Pathway Activation in Human Serum Remains on Porcine Cells Lacking -Glycolylneuraminic Acid.

Background: Triple knockout (TKO) donor pigs lacking alpha-1,3-galactose (Gal), -glycolylneuraminic acid (Neu5Gc), and Sd(a) expressions were developed to improve the clinical success of xenotransplantation. Neu5Gc, a sialic acid expressed on cell surfaces, recruits factor H to protect cells from attack by the complement system. Lack of Neu5Gc expression may cause unwanted complement activation, abrogating the potential benefit of gene-modified donor pigs.