Korea University Ansan Hospital Research Articles

Deep learning image reconstruction of diffusion-weighted imaging in evaluation of prostate cancer focusing on its clinical implications.

Image-based assessment of prostate cancer (PCa) is increasingly emphasized in the diagnostic workflow for selecting biopsy targets and possibly predicting clinically significant prostate cancer (csPCa). Assessment is based on Prostate Imaging-Reporting and Data System (PI-RADS) which is largely dependent on T2-weighted image (T2WI) and diffusion weighted image (DWI). This study aims to determine whether deep learning reconstruction (DLR) can improve the image quality of DWI and affect the assessment of PI-RADS ≥4 in patients with PCa. In this retrospective study, 3.0T post-biopsy prostate magnetic resonance imaging (MRI) of 70 patients with PCa in Korea University Ansan Hospital from November 2021 to July 2022 was reconstructed with and without using DLR. Four DWI image sets were made: (I) conventional DWI (CDWI): DWI with acceleration factor 2 and conventional parallel imaging reconstruction, (II) DL1: DWI with acceleration factor 2 using DLR, (III) DL2: DWI with acceleration factor 3 using DLR, and (IV) DL3: DWI with acceleration factor 3 and half average b-value using DLR. Apparent diffusion coefficient (ADC) value, signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR) were measured by one reviewer, while two reviewers independently assessed overall image quality, noise, and lesion conspicuity using a four-point visual scoring system from each DWI image set. Two reviewers also performed PI-RADSv2.1 scoring on lesions suspected of malignancy. A total of 70 patients (mean age, 70.8±9.7 years) were analyzed. The image acquisition time was 4:46 min for CDWI and DL1, 3:40 min for DL2, and 2:00 min for DL3. DL1 and DL2 images resulted in better lesion conspicuity compared to CDWI images assessed by both readers (P<0.05). DLR resulted in a significant increase in SNR, from 38.4±14.7 in CDWI to 56.9±21.0 in DL1. CNR increased from 25.1±11.5 in CDWI to 43.1±17.8 in DL1 (P<0.001). PI-RADS v2.1 scoring for PCa lesions was more agreeable with the DL1 reconstruction method than with CDWI (κ value CDWI, DL1; 0.40, 0.61, respectively). A statistically significant number of lesions were upgraded from PI-RADS <4 in CDWI image to PI-RADS ≥4 in DL1 images for both readers (P<0.05). Most of the PI-RADS upgraded lesions were from higher than unfavorable intermediate-risk groups according to the 2023 National Comprehensive Cancer Network guidelines with statistically significant difference of marginal probability in DL1 and DL2 for both readers (P<0.05). DLR in DWI for PCa can provide options for improving image quality with a significant impact on PI-RADS evaluation or about a 23% reduction in acquisition time without compromising image quality.

Development of a predictive model for hypoxia due to sedatives in gastrointestinal endoscopy: a prospective clinical study in Korea.

Sedation has become a standard practice for patients undergoing gastrointestinal (GI) endoscopy. However, considering the serious cardiopulmonary adverse events associated with sedatives, it is important to identify patients at high risk. Machine learning can generate reasonable prediction for a wide range of medical conditions. This study aimed to evaluate the risk factors associated with sedation during GI endoscopy and develop a predictive model for hypoxia during endoscopy under sedation. This prospective observational study enrolled 446 patients who underwent sedative endoscopy at the Korea University Ansan Hospital. Clinical data were used as predictor variables to construct predictive models using the random forest method that is a machine learning algorithm. Seventy-two of the 446 patients (16.1%) experienced life-threatening hypoxia requiring immediate medical intervention. Patients who developed hypoxia had higher body weight, body mass index (BMI), neck circumference, and Mallampati scores. Propofol alone and higher initial and total dose of propofol were significantly associated with hypoxia during sedative endoscopy. Among these variables, high BMI, neck circumference, and Mallampati score were independent risk factors for hypoxia. The area under the receiver operating characteristic curve for the random forest-based predictive model for hypoxia during sedative endoscopy was 0.82 (95% confidence interval, 0.79-0.86) and displayed a moderate discriminatory power. High BMI, neck circumference, and Mallampati score were independently associated with hypoxia during sedative endoscopy. We constructed a model with acceptable performance for predicting hypoxia during sedative endoscopy.

Hearing Recovery Prediction for Patients with Chronic Otitis Media Who Underwent Canal-Wall-Down Mastoidectomy.

Background: Chronic otitis media affects approximately 2% of the global population, causing significant hearing loss and diminishing the quality of life. However, there is a lack of studies focusing on outcome prediction for otitis media patients undergoing canal-wall-down mastoidectomy. Methods: This study proposes a recovery prediction model for chronic otitis media patients undergoing canal-wall-down mastoidectomy, utilizing data from 298 patients treated at Korea University Ansan Hospital between March 2007 and August 2020. Various machine learning techniques, including logistic regression, decision tree, random forest, support vector machine (SVM), extreme gradient boosting (XGBoost), and light gradient boosting machine (light GBM), were employed. Results: The light GBM model achieved a predictive value (PPV) of 0.6945, the decision tree algorithm showed a sensitivity of 0.7574 and an F1 score of 0.6751, and the light GBM algorithm demonstrated the highest AUC-ROC values of 0.7749 for each model. XGBoost had the most efficient PR-AUC curve, with a value of 0.7196. Conclusions: This study presents the first predictive model for chronic otitis media patients undergoing canal-wall-down mastoidectomy. The findings underscore the potential of machine learning techniques in predicting hearing recovery outcomes in this population, offering valuable insights for personalized treatment strategies and improving patient care.

Modified intraocular lens power selection method according to biometric subgroups Eom IOL power calculator

This study evaluates the accuracy of a newly developed intraocular lens (IOL) power calculation method that applies four different IOL power calculation formulas according to 768 biometric subgroups based on keratometry, anterior chamber depth, and axial length. This retrospective cross-sectional study was conducted in at Korea University Ansan Hospital. A total of 1600 eyes from 1600 patients who underwent phacoemulsification and a ZCB00 IOL in-the-bag implantation were divided into two datasets: a reference dataset (1200 eyes) and a validation dataset (400 eyes). Using the reference dataset and the results of previous studies, the Eom IOL power calculator was developed using 768 biometric subgroups. The median absolute errors (MedAEs) and IOL Formula Performance Indexes (FPIs) of the Barrett Universal II, Haigis, Hoffer Q, Holladay 1, Ladas Super, SRK/T, and Eom formulas using the 400-eye validation dataset were compared. The MedAE of the Eom formula (0.22 D) was significantly smaller than that of the other four formulas, except for the Barrett Universal II and Ladas Super formulas (0.24 D and 0.23 D, respectively). The IOL FPI of the Eom formula was 0.553, which ranked first, followed by the Ladas Super (0.474), Barrett Universal II (0.470), Holladay 1 (0.444), Hoffer Q (0.396), Haigis (0.392), and SRK/T (0.361) formulas. In conclusion, the Eom IOL power calculator developed in this study demonstrated similar or slightly better accuracy than the Barrett Universal II and Ladas Super formulas and was superior to the four traditional IOL power calculation formulas.

Audiologic results of total ossicular replacement prosthesis bypassing the stapes suprastructure in canal wall down mastoidectomy

ObjectivesSometimes performing PORP adequately is challenging when the stapes is tilted or the suprastructure is partially damaged owing to inflammation or infection. In such cases, the implementation of a TORP bypassing the stapes can be a useful alternative. This study aimed to investigate whether bypassing the stapes suprastructure during total ossicular replacement prosthesis (TORP) affects postoperative complications or audiological outcomes. Material and methodsAmong 104 patients who underwent open cavity mastoidectomy and ossiculoplasty using a titanium prosthesis at Korea University Ansan Hospital between January 2012 and December 2019, we compared the preoperative and postoperative audiological results and surgical complications of 52, 21, and 31 patients who underwent partial ossicular replacement prosthesis (PORP), TORP bypassing the remaining stapes suprastructure, and TORP on the stapes footplate or oval window, respectively. ResultsThe air-bone gap before surgery was significantly different in the TORP on the stapes footplate group (34.2 ± 12.0 dB) than that in the PORP (22.9 ± 13.8 dB) and TORP bypassing the stapes groups (20.7 ± 11.5 dB, p < 0.001). After surgery, there were no significant differences among the groups (p = 0.818). The air-bone gap difference before surgery was associated with the presence of stapes before surgery (p < 0.001). There was no difference in the proportion of postoperative tympanic perforation, whether it was a revision surgery, malleus status, or the size of perforation of the tympanic membrane among the three groups. ConclusionWhen performing ossiculoplasty using TORP, bypassing the stapes did not affect surgical and audiologic outcomes.

490. Brain-derived neurotrophic factor (BDNF) is a prognostic biomarker and therapeutic target through AMP-activated protein kinase (AMPK) signal pathway for sepsis

Abstract Background Brain-derived neurotrophic factor (BDNF) is a pleiotropic cytokine that not only acts on the central nervous system, but is also implicated in metabolic regulation, development of lung diseases, and support of hematopoiesis. Recently, a study report that the plasma levels of BDNF were independently associated with mortality in critically ill patients. We aimed to evaluate the role of BDNF in sepsis patients and animal sepsis model. Methods Prospectively enrolled adult sepsis patients (n=168) who visited the emergency room (ER) of Korea University Ansan Hospital from January 2017 to December 2018., and healthy volunteers (n=48) were included in the present study. Serum BDNF level was determined and analyzed in septic patients and healthy volunteers. The effects of BDNF on pro-inflammatory cytokines, AMP-activated protein kinase (AMPK) and NF-κB signaling in septic condition of RAW264.7 cells and mouse peritoneal macrophage, and mouse sepsis model by cecal ligation and puncture (CLP). Results Serum BDNF levels (mean) were significantly decreased in sepsis patients compared to healthy controls (11.78ng/ml vs 18.91ng/ml, p&amp;lt; 0.001) and significantly decreased in non-survivors compared to survivors at discharge (3.16ng/ml vs 13.86ng/ml, p&amp;lt; 0.05). BDNF treatment suppressed LPS-induced pro-inflammatory cytokine (TNF-α, iL-6, and iL-1β) in RAW264.7 cells. Although LPS stimulation diminished AMPK phosphorylation, BDNF was able to decrease the effects of LPS/toll-like receptor 4 (TLR4) engagement on down stream signaling events, particularly LPS-induced IκBα degradation in RAW cells and peritoneal macrophage. In CLP model, intra-muscular injection of BDNF after CLP significantly decreased serum levels of TNF-α which was increased compared to sham, and significantly improved survival of mouse (p&amp;lt; 0.01). Conclusion Sepsis results in a decreased production of BDNF, which might be a potential prognostic biomarker and therapeutic target through AMPK signal pathway for sepsis. Disclosures All Authors: No reported disclosures.

Toward Personalized Hemoglobin A1c Estimation for Type 2 Diabetes

A considerable gap is noted between a measured value of hemoglobin A1c (HbA1c) and its estimation from blood glucose (BG) levels; this gap needs to be resolved to improve glycemic control in people with diabetes. We aimed to develop clustering-based personalized models that estimate the HbA1c level from continuous glucose monitoring (CGM) value using a real-world clinical dataset and a novel machine learning (ML) approach. We used data from 101 insulin-treated Koreans with type 2 diabetes at Korea University Ansan Hospital (Ansan, South Korea). The dataset consisted of CGM values, activity level, nutritional data, antidiabetic medication for ten days, and HbA1c levels. Using the dataset, we optimized a novel ML approach, <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${K}$ </tex-math></inline-formula> -means local nonlinear regressor (NLR), and developed two types of estimation models. The proposed models showed 0.877% and 0.857% mean absolute differences (MADs) in estimating HbA1c levels, whereas Nathan’s and Bergenstal’s general models showed 1.296% and 1.157%, respectively, and Grossman’s personalized models showed either 0.952% or 0.923%. The proposed model outperformed commonly used models, and the performance improvement in our model came from appropriate personalization—-estimating HbA1c higher for patients with poor glycemic control. We expect that the proposed CGM-based HbA1c estimator can serve as a more powerful motivator for glucose control. The proposed model, along with the CGM data, can provide real-time integrated information on glycemic control, which allows diabetic patients to adjust their lifestyles and physicians to modify management plans more promptly. Ultimately, this may result in better glycemic control than using conventional HbA1c measurement.

Safety and efficacy of knotless barbed suture in cesarean section using postpartum ultrasound: a retrospective cohort study

Background: While speed, safety, and efficacy are necessary in the operation of cesarean section (CS), a number of new devices have been introduced to support the technique. This retrospective study was conducted to assess the usability and safety of knotless barbed suture, compared to conventional method, for closure of uterine myometrium during cesarean section. Methods: Patients who underwent cesarean delivery at Korea University Ansan Hospital between August 2018 and December 2019 were reviewed. Surgical outcomes including uterine closure time, operating time, estimated blood loss, and sonographic results of residual myometrial thickness at the site of incision at 6 to 8 weeks after operation were analyzed. Results: Out of total 78 women, 44 patients had knotless barbed suture and 34 patients underwent conventional suture. Compared to conventional method, the uterine closure time was significantly reduced (p = 0.012). Operative time, estimated blood loss during operation, hemoglobin difference on the third operative day, and the number of cases with transfusion, postpartum infection, and bleeding demonstrated no significant differences between the two groups. At 6 to 8 weeks after operation, transvaginal ultrasound revealed no differences in residual myometrial scar thickness. Conclusion: Knotless barbed suture resulted in significantly shorter uterine closure time and no increased rate of perioperative complications. Our work supports that knotless barbed suture can be effectively used for cesarean section.

Size of Patent Ductus Arteriosus and Echocardiographic Markers of Shunt Volume in Preterm Infants Based on Postnatal Age.

Objective: This study was conducted in order to compare the strength of correlation between echocardiographic markers of shunt volume and patent ductus arteriosus (PDA) diameter based on postnatal age.Methods: This retrospective study focused on preterm infants (aged <32 weeks of gestation) admitted to the Neonatal Intensive Care Unit of Korea University Ansan Hospital, between April 2014 and December 2017, who studied serial targeted neonatal echocardiography (TNE) for PDA during hospitalization. The association between echocardiographic characteristics and duct size was divided into the following days: within 3 days (very early, VE), 4–7 days after birth (early, E), and after 8 days of birth (late, L).Results: We found 113 assessments conducted on 57 infants in the VE period, 92 assessments on 40 infants in the E period, and 342 assessments on 37 infants in the L period. Median gestational age and birth weight were 28+2 weeks of gestation and 1,115 g, respectively. In the univariate regression analysis, we found a statistically significant correlation between PDA diameter and all TNE markers in the E and L days, but not in the VE period. Only ductal velocity [coefficient of determination (R2) = 0.224], antegrade left pulmonary artery diastolic flow velocity (R2 = 0.165), left ventricular output (LVO)/superior vena cava (SVC) flow ratio (R2 = 0.048), and E/A wave ratio (R2 = 0.092) showed weak correlations with PDA diameter in the VE period. The slopes of the regressions showed significant changes based on postnatal age in the maximum ductal velocity, left atrium/aorta ratio, LVO/SVC flow ratio, and LVO.Conclusions: It is difficult to predict the echocardiographic markers of shunt volume based on the PDA diameter in preterm infants younger than 4 days. A better understanding of the changes in the hemodynamic consequences of PDA based on postnatal age is needed when considering treatment.

Cross-Contamination versus Outbreak: Pre-XDR Mycobacterial Strains Confirmed by Whole-Genome Sequencing.

Whole-genome sequencing (WGS) is promising for the quality control of laboratory facilities for Mycobacterium tuberculosis (MTB) strains. We describe the clinical and laboratory characteristics of false positive versus true positive MTB cultures based on WGS, which were experienced in a real clinical setting. Strain harvest and DNA extraction from seven isolates from pre-extensive drug-resistant (pre-XDR) TB patients transferred to the Korea University Ansan Hospital were performed, and epidemiologic links and clinical information, including the phenotypic drug susceptibility test (pDST), were investigated. WGS was performed using Ion GeneStudio with an ION530tm chip (average sequencing depth, ~100-fold). In the phylogenetic tree, identical and different strains were distributed separately. Five of the seven isolates were identical; the remaining two isolates differed from the others. The images of the referred pre-XDR-TB patients with false positive MTB that were analyzed were of regions close to old TB scars. Further, the results of WGS gene mutation analysis for ethambutol, streptomycin, and fluoroquinolone resistance in all six patients were not concordant with the pDST results. WGS and clinical information were useful in differentiating laboratory cross-contamination from true positive TB, thereby avoiding the unnecessary treatment of false positive patients and delay in treating true positive TB patients, with reliable genotypic drug resistance results.

Standardization of the Korean Version of the Patient Health Questionnaire-4 (PHQ-4).

ObjectiveThe Patient Health Questionnaire-4 (PHQ-4) has been used for screening owing to ease of use and brevity. In this study, we developed the Korean version of the PHQ-4 and tested its validity.MethodsOne hundred sixteen new adult outpatients at the Department of Psychiatry of the Korea University Ansan Hospital participated in the study. We simultaneously administered other depression/anxiety scales the Hamilton Rating Scale for Depression, the Hamilton Anxiety Scale, the Beck Depression Inventory, and the Beck Anxiety Inventory.ResultsThe mean PHQ-4 score was 6.52 (standard deviation = 3.45). Cronbach’s α was 0.792, and the intraclass correlation coefficient of test and 2-week interval retest was 0.827 (p < 0.01). The Pearson correlation coefficients between the PHQ-4 total score and other depression/anxiety scales were all over 0.6. Confirmatory factorial analysis showed acceptable convergent validity and reliability but questionable discriminant validity for some model fit values.ConclusionThe Korean version of the PHQ-4 has sufficient internal consistency, test-retest reliability, and construct validity, but its two-factor structure showed incompleteness. However, we suggest that it should be used as a brief screening measure for common psychiatric distress that warrants further detailed assessment, but not to separately assess the severity of depression and anxiety symptoms.

Sex differences in deterioration of sleep properties associated with aging: a 12-year longitudinal cohort study.

The sleep patterns of humans are greatly influenced by age and sex and have various effects on overall health as they change continuously during the lifespan. We investigated age-dependent changes in sleep properties and their relation to sex in middle-aged individuals. We analyzed data from 2,640 participants (mean age of 49.8 ± 6.8 years at baseline, 50.6% women) in the Korean Genome and Epidemiology Study, which assessed sleep habits using the Pittsburgh Sleep Quality Index and other clinical characteristics. We analyzed the sleep habit changes that occurred between baseline and a follow-up point (mean interval: 12.00 ± 0.16 years). Associations of age and sex with 9 sleep characteristics were evaluated. Age was associated with most of the sleep characteristics cross-sectionally and longitudinally (P < .05), except for the time in bed at baseline (P = .455) and change in sleep duration (P = .561). Compared with men, women had higher Pittsburgh Sleep Quality Index scores, shorter time in bed, shorter sleep duration, and longer latency at baseline (P ≤ .001). Longitudinal deterioration in Pittsburgh Sleep Quality Index score, habitual sleep efficiency, duration, and latency was more prominent in women (P < .001). The sex differences in these longitudinal sleep changes were mainly noticeable before age 60 years (P < .05). Worsening of Pittsburgh Sleep Quality Index scores, habitual sleep efficiency, and latency was most evident in perimenopausal women. Men presented with greater advancement of chronotype (P = .006), with the peak sex-related difference occurring when they were in their late 40s (P = .048). Aging is associated with substantial deterioration in sleep quantity and quality as well as chronotype advancement, with the degree and timing of these changes differing by sex.

Perinatal Outcomes Associated with Latency in Late Preterm Premature Rupture of Membranes.

This study aims to evaluate the perinatal outcomes of preterm premature rupture of membrane (PPROM) with latency periods at 33 + 0–36 + 6 weeks of gestation. This retrospective case-control study included women with singleton pregnancies who delivered at 33 + 0–36 + 6 weeks at Korea University Ansan Hospital in South Korea between 2006–2019. The maternal and neonatal characteristics were compared between different latency periods (expectant delivery ≥72 h vs. immediate delivery <72 h). Data were compared among 345 women (expectant, n = 39; immediate delivery, n = 306). There was no significant difference in maternal and neonatal morbidities between the groups, despite the younger gestational age in the expectant delivery group. Stratified by gestational weeks, the 34-week infants showed a statistically significant lower exposure to antenatal steroids (73.4% vs. 20.0%, p < 0.001), while the incidence of respiratory distress syndrome (12.8%) and the use of any respiratory support (36.8%) was higher than those in the 33-week infants, without significance. Our study shows that a prolonged latency period (≥72 h) did not increase maternal and neonatal morbidities, and a considerable number of preterm infants immediately delivered at 34 weeks experienced respiratory complications. Expectant management and antenatal corticosteroids should be considered in late preterm infants with PPROM.

Does Intracorporeal Anastomosis Decrease the Rate of Surgical Site Infection in Laparoscopic Colon Cancer Surgery?

Objective This study aimed to compare the surgical site infection (SSI) rates between intracorporeal anastomosis (ICA) and extracorporeal anastomosis (ECA). Summary of background data Laparoscopic surgery is recommended for colonic malignancies because of its superior clinical outcomes and comparable oncologic results. Laparoscopic colectomy with ICA has the advantages of incision length and free extraction site choice. However, ICA may be associated with a risk of SSI due to enterotomy inside the abdominal cavity. Methods We retrospectively analyzed patients with colon cancer who underwent radical surgery at Korea University Ansan Hospital between January 2017 and June 2020. We compared the SSI rates and other clinical variables between the ICA and ECA groups. Results Of the 502 patients who underwent radical surgery for colorectal cancer during the study period, 234 were eligible for inclusion. ECA and ICA were performed in 62.4% and 37.6% of patients, respectively. There were no statistically significant intergroup differences in clinicopathologic variables. The overall SSI rate did not differ between the groups (P = 0.801), but organ/space SSIs were more common in the ICA group than in the ECA group (P = 0.048). Conclusions There was no significant difference in overall SSI or anastomotic leakage (AL) rates between the ICA and ECA groups, but the organ/space SSI rate was higher in the ICA group when AL cases were excluded. Further high-quality studies are needed to assess the risk of organ/space SSIs in the ICA after colon cancer surgery.

Sarcopenia is Associated With Decreased Brain Volume in the General Population: A Longitudinal Cohort Study

Background: Substantial evidence supports an association between physical activity and cognitive function. However, the role of muscle mass and function in brain structural changes is not well known. This study investigated whether sarcopenia, defined as low muscle mass and strength, affects brain atrophy in the general middle-to-later aged population. Methods: A total of 1284 participants with sarcopenic measurements and baseline and 4-year follow-up brain magnetic resonance images were recruited from the Korean Genome and Epidemiology Study. Muscle mass was represented as appendicular skeletal muscle mass divided by the body mass index. Muscle function was measured by handgrip strength. The low mass and strength groups were defined as being in the lowest quintile of each variable for one’s sex. Sarcopenia was defined as being in the lowest quintile for both muscle mass and strength. Findings: Of the 1284 subjects, 12·6%, 10·8%, and 5·4% were classified as the low mass, low strength, and sarcopenia groups, respectively. All three groups presented greater loss in gray matter (GM) volume than the control group. The adjusted mean changes of GM volume during 4-year follow-up period were -9·6 mL in the control group, whereas -11·6 mL in the other three groups (P < 0·001). For lobar brain volumes, the low mass group presented greater atrophy in frontal and occipital GM. The significantly faster atrophy in parietal GM was observed in the sarcopenia group compared with the control group, mostly driven by atrophy in the left inferior parietal lobule. In a joint regression model, low muscle mass, but not muscle strength, was an independent factor associated with rapid GM atrophy. Interpretation: Sarcopenia is associated with rapid declines in GM volume, especially parietal GM volume, in the general population. Maintaining good levels of muscle mass could be important for brain health in later adulthood. Funding Statement: This research was supported by funds (2009-E71002-00, 2010-E71001-00, 2011-E71004-00, 2012-E71005-00, 2013-E71005-00, 2014-E71003-00, 2015-P71001-00, 2016-E71003-00, 2017-E71001-00, 2018-E7101-00) from the Korean Centers for Disease Control and Prevention, the Bio & Medical Technology Development Program of the National Research Foundation (NRF) funded by the Korean government (MSIT) (NRF-2019M3E5D3073102,2019R1H1A2039682, 2020R1F1A1074265), a Korea University Grant (K1824431), Ansan-Si hidden champion fostering and supporting project funded by Ansan city, and computational resources provided by the University of Iowa, Iowa City, Iowa. . Declaration of Interests: No potential conflicts of interest relevant to this article are reported. Ethics Approval Statement: Each participant signed an informed consent form. This study was performed according to the principles of the Declaration of Helsinki of the World Medical Association and was approved by the Institutional Review Board of Korea University Ansan Hospital.

First-line Helicobacter pylori Eradication Rate of the 10-day Hybrid Therapy

Background/Aims: To improve the eradication rate of a first-line therapy for Helicobacter pylori infection, alternate regimens such as sequential, concomitant, and hybrid therapies have been tried. The aim of this study was to evaluate the eradication rate of the 10-day hybrid therapy as a first-line therapy.Materials and Methods: This retrospective study enrolled 124 patients from the Korea University Ansan Hospital between April 2016 and December 2019. The 10-day hybrid therapy comprised 5 days of dual therapy (proton pump inhibitor [PPI] standard dose and amoxicillin 1 g, twice daily) followed by 5 days of quadruple therapy (PPI, amoxicillin 1 g, clarithromycin 500 mg, and metronidazole 500 mg, twice daily). We compared the 10-day hybrid therapy with the 10-day concomitant therapy comprising PPI, amoxicillin 1 g, clarithromycin 500 mg, and metronidazole 500 mg, twice daily. Eradication was assessed by a &lt;sup&gt;13&lt;/sup&gt;C-urea breath test or gastroscopic biopsy at least 4 weeks after treatment completion.Results: The eradication rates of the 10-day hybrid and concomitant therapies were 74.2% (46/62) and 67.7% (42/62), respectively, in the intention-to-treat (ITT) analysis and 88.5% (46/52) and 82.4% (42/51), respectively, in the per-protocol (PP) analysis. There was no significant difference in the eradication rates between the two groups in the ITT (P=0.429) and PP analysis (P=0.380). Adverse events developed in 75.0% and 70.6% of patients in the hybrid and concomitant groups, respectively, but there was no significant difference (P=0.615).Conclusions: The 10-day hybrid therapy can be an option for a first-line therapy of Helicobacter pylori infection.

An Estimation of the Proper Isolation Period for COVID-19 Patients with Mild Symptoms: A Multicenter Study in South Korea

Background: The objective of the present study was to estimate the proper isolation period for confirmed cases of COVID-19 with mild symptoms. Methods: The multicenter retrospective study was conducted in the three living and treatment support centers (LTSC) of South Korea. Data on the characteristics of the patients and the duration of the LTSC stay were collected. Questionnaires were used for basic medical information and daily self-monitoring for health from the day of admission to March 25, 2020. In addition, the real-time reverse-transcriptase polymerase chain reaction (RT-PCR) results were analyzed. Findings: A total of 572 patients was enrolled in the study. On March 25, 2020, a total of 280 patients (49.0%) maintained isolation, and 292 (51.0%) were released from isolation. Among the patients released from isolation, 75% and 90% of patients were released from isolation within 21 days and 28 days, respectively (range: 7-42 days). A Kaplan-Meier survival curve showed that there was no significant difference in the total isolation period between patients who were isolated in a solitary room and those who were isolated in a cohort room (log-rank, P = 0.714). The receiver operating characteristic curves of the release from isolation based on the total days of isolation showed that the optimal cut-off value was 19 days. Interpretation: Of mild COVID-19 patients, 90% released from the isolation within 4 weeks. As for the RT-PCR test to decide for the discontinuation of isolation, it would be cost-effective to start the test at 19 days after diagnosis. Funding Statement: The study was conducted without special funding. Declaration of Interests: The authors declare that they have no competing interests relevant to this study. Ethics Approval Statement: The study protocol was approved by the Institutional Review Board (IRB) of Korea University Ansan Hospital (IRB number: 2020AS0083). The requirement for written informed consent from patients was waived due to the retrospective nature of the study and its impracticability

Machine Learning Models for Predicting Hearing Prognosis in Unilateral Idiopathic Sudden Sensorineural Hearing Loss.

Objectives.Prognosticating idiopathic sudden sensorineural hearing loss (ISSNHL) is an important challenge. In our study, a dataset was split into training and test sets and cross-validation was implemented on the training set, thereby determining the hyperparameters for machine learning models with high test accuracy and low bias. The effectiveness of the following five machine learning models for predicting the hearing prognosis in patients with ISSNHL after 1 month of treatment was assessed: adaptive boosting, K-nearest neighbor, multilayer perceptron, random forest (RF), and support vector machine (SVM).Methods.The medical records of 523 patients with ISSNHL admitted to Korea University Ansan Hospital between January 2010 and October 2017 were retrospectively reviewed. In this study, we analyzed data from 227 patients (recovery, 106; no recovery, 121) after excluding those with missing data. To determine risk factors, statistical hypothesis tests (e.g., the two-sample t-test for continuous variables and the chi-square test for categorical variables) were conducted to compare patients who did or did not recover. Variables were selected using an RF model depending on two criteria (mean decreases in the Gini index and accuracy).Results.The SVM model using selected predictors achieved both the highest accuracy (75.36%) and the highest F-score (0.74) on the test set. The RF model with selected variables demonstrated the second-highest accuracy (73.91%) and F-score (0.74). The RF model with the original variables showed the same accuracy (73.91%) as that of the RF model with selected variables, but a lower F-score (0.73). All the tested models, except RF, demonstrated better performance after variable selection based on RF.Conclusion.The SVM model with selected predictors was the best-performing of the tested prediction models. The RF model with selected predictors was the second-best model. Therefore, machine learning models can be used to predict hearing recovery in patients with ISSNHL.

Can natural killer cell activity help screen patients requiring a biopsy for the diagnosis of prostate cancer?

ABSTRACTPurpose To evaluate the usefulness of natural killer cell activity (NKA) in diagnosing prostate cancer (PC).Materials and Methods The medical records of patients who underwent transrectal prostate biopsy (TRBx) at Korea University Ansan Hospital between May 2017 and December 2017 were retrospectively reviewed. NKA levels were measured using NK Vue® Tubes (ATgen, Sungnam, Korea). All blood samples were obtained at 8 AM on the day of biopsy. Patients with other malignancies, chronic inflammatory conditions, high prostate-specific antigen (PSA) level (>20ng/mL), or history of taking 5-alpha-reductase inhibitor or testosterone replacement therapy were excluded.Results A total of 102 patients who underwent TRBx for PC diagnosis were enrolled. Among them, 50 were diagnosed with PC. Significant differences in age and NKA level were observed between the PC and no-PC groups. Receiver operating characteristic (ROC) curve analysis showed that the optimal cut-off of NKA level for the prediction of PC was 500pg/dL, with a sensitivity of 68.0% and a specificity of 73.1%. In addition, NKA level (0.630) had the greatest area under the ROC curve compared to those for the ratio of total PSA to free PSA (0.597) and PSA density (0.578).Conclusions The results of this pilot study revealed that low NKA and high PSA levels were likely to be associated with a positive TRBx outcome. NKA detection was easy and improved the diagnostic accuracy of PC.

Timing of antibiotics in septic patients: a prospective cohort study

ObjectivesTo evaluate the effect of timing and appropriateness of antibiotics administration on mortality in patients diagnosed with sepsis according to the Sepsis-3 definition. MethodsThis prospective cohort study was conducted in patients diagnosed with sepsis according to the Sepsis-3 definition at the emergency department of Korea University Ansan Hospital from January 2016 to January 2019. The time to antibiotics was defined as the time in hours from emergency department arrival to the first antibiotic administration. Cox proportional hazards regression analysis was used to estimate the association between time to antibiotics and 7-, 14- and 28-day mortality. ResultsOf 482 patients enrolled onto this study, 203 (42.1%) of 482 and 312 (64.7%) of 482 were diagnosed with septic shock and high-grade infection respectively. The median time to receipt of antibiotic therapy was 115 minutes. Antibiotics were administered within 3 and 6 hours in 340 (70.4%) of 482 and 450 (93.2%) of 482 patients respectively. Initial appropriate empirical antibiotics were administered in 375 (77.8%) of 482 patients. The time to and appropriateness of the initial antibiotics were not associated with 7-, 14- and 28-day mortality in multivariate analysis. The Sequential Organ Failure Assessment (SOFA) score (adjusted hazard ratio (aHR) 1.229, 95% confidence interval (CI) 1.093–1.381, p 0.001) and initial lactate levels (aHR 1.128, 95% CI 1.034–1.230, p 0.007), Charlson comorbidity index (aHR 1.115, 95% CI 1.027–1.210, p 0.014), 2-hour lactate level (aHR 1.115, 95% CI 1.027–1.210, p 0.009) and SOFA score (aHR 1.077, 95% CI 1.013–1.144, p 0.018) affected 7-, 14- and 28-day mortality respectively. Subgroup analysis with septic shock, bacteraemia and high-grade infection did not affect mortality rates. ConclusionsTime to receipt of antibiotics may not affect the prognosis of patients with sepsis if a rapid and well-trained resuscitation is combined with appropriate antibiotic administration within a reasonable time.