Background: The Dejour classification is widely accepted for grading the severity of trochlear dysplasia. However, whether it serves as a reliable prognostic marker for patellar instability surgery or a guide for clinical decision-making has yet to be established. Medial patellofemoral ligament (MPFL) reconstruction, with or without tibial tubercle osteotomy (TTO), yields comparable outcomes across Dejour grades. Nonetheless, we hypothesized that specific trochlear parameters would be more significant predictors of surgical results. Methods: This study included patients who underwent MPFL reconstruction, with or without TTO, for recurrent patellar dislocation at a single institution between 2010 and 2023. Patients with at least 1 year of follow-up were included. Preoperative demographics, radiographic measurements such as sulcus angle, lateral trochlear inclination (LTI), and trochlear depth, patient-reported outcome measures (PROMs), and intraoperative findings including cartilage status were analyzed. Severe trochlear dysplasia (STD) was categorized as Dejour B/D, while non-STD encompassed all other cases, and these 2 groups were compared. Postoperative PROMs were assessed at 1 year postoperatively and at the final follow-up. A noninferiority test was performed using Kujala, Lysholm, and International Knee Documentation Committee (IKDC) scores, with the inferiority margin defined by the minimal clinically important difference (MCID). Linear relationships between LTI, a widely used indicator of trochleoplasty, and PROMs were evaluated. Additionally, differences in the proportion of patients achieving MCID, substantial clinical benefit, and patient-acceptable symptom state were analyzed using odds ratios. Results: Forty-two knees (41 patients) were finally enrolled. Radiographic measurements, except for LTI at the most proximal trochlea, showed no significant differences between the STD and non-STD groups. Clinical outcomes also demonstrated no significant differences between the 2 groups, and noninferiority was confirmed. When analyzed based on LTI, no differences in PROMs were observed at the postoperative 1 year. However, at the final follow-up (mean, 37.27 months), the most proximal trochlear LTI showed significant correlations with both Kujala and Knee Injury and Osteoarthritis Outcome Score over the follow-up period. Conclusions MPFL reconstruction, with or without TTO, leads to similar clinical improvements for STD classified by Dejour criteria. However, when assessed using proximal LTI, dysplasia’s role as a prognostic factor should be considered.

Background: Enzyme immunoassays (EIAs), which detect glutamate dehydrogenase (GDH) and toxin A/B, are widely used to screen for Clostridioides difficile infection (CDI); however, their sensitivity is lower than that of molecular assays. This study aimed to evaluate the performance of two EIAs, C. Diff Quik Chek Complete (QCC) and RIDASCREEN (RIDA), and investigate the cycle threshold (Ct) values from two real-time polymerase chain reaction (PCR) assays (Allplex GI–Bacteria(I) and Xpert C. difficile) in EIA-discordant samples. Methods: A total of 180 clinical stool samples were tested using QCC, RIDA, and Allplex GI– Bacteria(I) PCR assays. The Xpert C. difficile assay was used to analyze discordant results. Results: QCC and RIDA showed high sensitivities for GDH detection, 100.0% and 94.4%, respectively. QCC was significantly more sensitive than RIDA for toxin detection (51.4% vs. 28.6%, p = 0.007). In 25 EIA-discordant, Xpert positive samples, the Ct values of the toxin B gene ranged from 31.5 to 44.8 (mean, 38.1) for Allplex PCR and from 23.7 to 36.3 (mean, 30.4) for Xpert PCR. The Ct values of the two PCR assays were not significantly correlated (r = 0.201, p = 0.324). Conclusion QCC is a suitable initial immunological test for diagnosing CDI. The lack of correlation in the Ct values between the two real-time PCR assays suggests that assay-specific validation is necessary for cutoff level interpretation.

Rectal cancer, which accounts for approximately 40% of colorectal cancers, remains a major clinical concern. Recent advances in diagnostic imaging, surgical techniques, radiotherapy, and systemic treatment have steadily improved rectal cancer outcomes. Considering this, the Korean Rectal Cancer Multidisciplinary (KRCM) Committee has aimed to provide clinicians and policymakers with up-to-date, evidence-based clinical practice guidelines to support optimal decision-making, reflecting current evidence, the Korean healthcare context, and patient values and preferences. The Clinical Practice Guidelines for Rectal Cancer version 2.0 were developed through multidisciplinary collaboration with related academic societies, building upon and updating the KRCM Clinical Practice Guidelines version 1.0 (titled “Multidisciplinary guidelines for the management of rectal cancer”). These consensus guidelines of the KRCM were established based on a comprehensive literature review, evidence synthesis, with recommendation development guided by the GRADE (Grading of Recommendations Assessment, Development and Evaluation) methodology, and consideration of applicability in real-world clinical practice under the national health insurance system. Each recommendation has been presented with its strength and level of evidence.

Background: Rosacea is a chronic inflammatory skin disorder characterized by erythema, papules, ocular symptoms, and heightened sensitivity. Patients with neurogenic symptoms such as burning or stinging remain particularly difficult to manage. Current guidelines often underrepresent energy-based devices (EBDs), pigmentary sequelae, psychosocial burden, and ocular comorbidities. Objective: To examine Korean dermatologists’ expert perspectives on rosacea management, focusing on skin sensitivity, neurogenic symptoms, pigmentary changes, psychosocial impact, ocular involvement, and EBD use. Methods: A web-based, 29-item survey was administered to 25 board-certified Korean dermatologists (May–June 2025). Quantitative and qualitative responses were analyzed. Results: Erythematotelangiectatic and papulopustular phenotypes with sensitivity skin predominated. EBDs (pulsed dye laser, intense pulsed light) were frequently used but limited by cost and sensitivity issues. Neurogenic symptoms were recognized but rarely treated with neuromodulators. Post-inflammatory hyperpigmentation was infrequent, yet monitoring was inconsistent.Psychosocial and ocular aspects were acknowledged but seldomly systematically addressed.Respondents expressed interest in emerging adjunctive treatments such as cold plasma, skin boosters, and holistic care approaches. Conclusion Korean dermatologists adopt individualized strategies for rosacea, yet practice gaps remain regarding neurogenic symptoms, pigmentary complications, and psychosocial and ocular comorbidities. Findings support the need for updated multidisciplinary, phenotype-driven guidelines aligned with real-world practice.

Recent advancements in artificial intelligence have led to increased interest in predictive modeling across various domains, including medicine. Although numerous metrics have been established for binary classification, the growing adoption of multi-class systems necessitates robust evaluation measures. However, comprehensive simulation studies comparing the performance of existing multi-class metrics under diverse data conditions remain limited. In this study, we first provide a concise overview of commonly used accuracy metrics for multi-class classification. Then, we report a simulation study that systematically evaluates several diagnostic accuracy measures under a wide range of scenarios, including three- and five-class settings, balanced and imbalanced sample sizes, and different distributional assumptions for predictors. We assessed each metric’s performance in terms of bias and 95% confidence interval coverage. Under balanced conditions, most metrics demonstrated stable and unbiased performance, closely approximating the true values. However, under imbalanced conditions, greater bias was observed, with the M-index and polytomous discrimination index exhibiting comparatively more stable performance across various scenarios. The micro-averaged receiver operating characteristic curve area consistently showed higher bias under class imbalance. Finally, we applied these metrics to a glioma tumor grading task using external datasets. This study provides a systematic comparison of commonly used metrics and offers practical guidance for selecting appropriate measures in multi-class classification tasks.

Purpose: Neoadjuvant treatment (NAT) is now the standard for borderline resectable pancreatic cancer (RPC) and is being considered for RPC. Early recurrence after curative surgery in RPC is often seen as a treatment failure, prompting considerations for NAT. Our goal was to develop an artificial intelligence (AI)-based predictive model utilizing preoperatively available factors to forecast early recurrences of resected RPC. Methods: This study included 469 patients who underwent surgery for RPC between 2011 and 2019. Clinicopathologic and oncologic data were retrospectively reviewed. Preoperative variables, including laboratory data and imaging findings, were collected. Early recurrence was defined as recurrence occurring within a year after surgery. Deep neural networks were then used to select variables by assessing their importance. A new model predicting early recurrence of RPC was subsequently developed. Results: Of the patients evaluated, 199 (42.4%) experienced early recurrence. The predictive model included 14 preoperative variables: CA 19-9, preoperative pancreatitis, serum albumin, platelet count, lymphocyte count, the American Society of Anesthesiologists physical status classification, tumor size, monocyte count, age, body mass index, CRP, hemoglobin, WBC count, and CEA. The area under the curve for the model was 0.786 in the training set and 0.734 in the test set. Conclusion We developed an AI-based model to predict the early recurrence of RPC using preoperative parameters. By identifying patients at risk of early recurrence, optimal individualized treatments such as NAT can be considered. Future prospective studies are crucial to establish clear indications for NAT in RPC.

Purpose: High-cost novel therapies for multiple myeloma (MM) require evaluation of efficacy and cost-effectiveness. Materials and Methods: This study developed a methodology to assess cost-effectiveness using nationwide data from 11,450 newly diagnosed MM patients. A novel algorithm was applied to identify lines of therapy (LoT). Results: The number of newly diagnosed MM patients increased significantly, from 873 in 2010 to 1,464 in 2019 (p < 0.001). Advancing LoT was associated with shorter time to next treatment (TTNT) and overall survival (OS) (p < 0.001), while all-cause medical costs increased with each LoT (p < 0.001). Bortezomib-melphalan-prednisolone was the most common frontline regimen for transplant-ineligible patients (29.2%), while bortezomib-thalidomide-dexamethasone was most used for transplant-eligible patients (11.3%). Daratumumab monotherapy demonstrated superior second TTNT (7.8 vs. 5.2 months) and OS (8.5 vs. 5.3 months) compared to standard care in heavily treated MM patients, with statistical significance maintained after cost adjustment. For subsequent therapies following daratumumab, a methodology was developed to estimate required medical costs using the incremental cost-effectiveness ratio (ICER): Expected cost ($)=ICER×(Expected life expectancy–0.567)+35,601. Conclusion This study provides a novel cost-effectiveness framework linking treatment efficacy and real-world costs, supporting predictions of societal costs for future MM therapies.

Pharmacogenomic variations may significantly influence responses to commonly prescribed stroke medications. Despite accumulating evidence, genetic testing has not yet been widely integrated into stroke care. This review summarizes current evidence and provides practical guidance for clinical implementation. Pharmacogenomic studies and clinical guidelines related to antiplatelet agents, anticoagulants, and statins were reviewed, with particular emphasis on East Asian populations. Substantial evidence supports genotype-guided use of clopidogrel (CYP2C19), warfarin (CYP2C9, VKORC1, CYP4F2), and statins (SLCO1B1, ABCG2). For aspirin, PTGS1/2 and PEAR1 variants have been investigated; however, current data remain insufficient for clinical application. Regarding direct oral anticoagulants (DOACs), candidate genes such as ABCB1 and CES1 demonstrate pharmacokinetic associations, though robust clinical outcome data are lacking. Distinct allele frequencies in East Asians—such as higher prevalence of CYP2C19 and ABCG2 variants—underscore the need for population-specific strategies. Beyond single-gene approaches, polygenic risk scores, pharmacogenomic panels, and integration with multi-omics data and artificial intelligence represent promising directions for personalized therapy. Pharmacogenomic testing can enhance stroke pharmacotherapy, particularly in populations with high frequencies of actionable variants. Broader implementation requires rapid testing platforms, clinician education, tailored clinical guidelines, and real-world validation of aspirin, DOACs, and multi-gene approaches. Future research should expand population-specific studies and integrate pharmacogenomics within the broader framework of precision medicine to ensure equitable clinical benefit.

Background: Left bundle branch block (LBBB) significantly increases the risk of left ventricular systolic dysfunction (LVSD) due to cardiac dyssynchrony. Although artificial intelligence–enabled electrocardiography (AI-ECG) mod‑ els show promise in detecting LVSD, their performance in LBBB patients remains underexplored. We hypothesized that an AI-ECG model clinically validated for detecting LVSD would accurately detect LVSD and predict future clinical outcomes in LBBB patients. Methods: In this retrospective multicenter study, 5,689 expert-validated LBBB ECGs collected from 2,813 patients between 2016 and 2024 were analyzed using a previously developed and validated AI-ECG model. LVSD was defined as an ejection fraction of ≤ 40%. Model performance was assessed using the area under the receiver operating char‑ acteristic curve (AUROC), the area under the precision-recall curve (AUPRC), sensitivity, and specificity. Patients were stratified into high- and low-risk groups based on a threshold that achieved 90% sensitivity. A Kaplan–Meier analysis was used to compare clinical outcomes. Results: Among the 2,813 LBBB patients (mean age, 70.7 years; male sex, 43.7%), hypertension and a history of heart failure were common. The AiTiALVSD model showed strong diagnostic performance for LVSD (AUROC, 0.930 [95% CI, 0.924–0.937]; AUPRC, 0.913 [95% CI, 0.902–0.923]; sensitivity, 0.979; specificity, 0.473). During the mean follow-up of 4.1 years, high-risk patients had significantly higher hazards than low-risk patients for all-cause mortality (adjusted hazard ratio [HR], 1.87; 95% CI, 1.53–2.28), implantable cardioverter defibrillator/cardiac resynchronization therapy implantation (adjusted HR, 15.2; 95% CI, 7.51–30.77), and cardiovascular hospitalization (adjusted HR, 1.11; 95% CI, 0.96–1.28). Conclusions AiTiALVSD effectively detects LVSD and stratifies long-term cardiovascular risk in LBBB patients, support‑ ing its clinical utility for early detection and patient management.