Mid-term results of cyanoacrylate closure for the treatment of incompetent great and small saphenous veins: Findings from a Japanese prospective consecutive multi-center registry: Mid-term results of cyanoacrylate closure.

OBJECTIVES: We evaluated the efficacy and safety of cyanoacrylate closure (CAC) for endovascular treatment of varicose veins with cyanoacrylate adhesive (VenaSeal® closure system) in Japan. METHODS: A multicenter prospective consecutive registry study was conducted at 12 centers in Japan on 125 patients with primary varicose veins who underwent CAC. The patients were evaluated on target vein occlusion, postoperative complications, Visual Analogue Scale (VAS) for pain, revised Venous Clinical Severity Score (rVCSS), Aberdeen Varicose Vein Questionnaire (AVVQ), and EuroQol 5 dimensions 5-level (EQ-5D-5L) for 1-year after the surgery. RESULTS: The closure rate was 92.6% at 1 year postoperatively, and 95.0% and 90.2% for GSV and SSV respectively with little difference (p = .491). The mean VAS in the immediate postoperative period was 18.9 ± 23.4. Postoperative complications were observed in 20 patients (16%). Hypersensitivity-type phlebitis occurred in 7 patients (5.6%). Infection of the treated vein resulted in resection of GSV. The rVCSS and AVVQ improved significantly after 90 days and 1 year postoperatively (p < .001), while the EQ-5D-5L have not changed. CONCLUSION: Cyanoacrylate Closure was considered generally a safe and minimally invasive treatment with good mid-term outcomes including SSV. However further study is required for some CAC specific complications.

Interlaboratory Study Toward Combining Gait Kinematics Data Sets of Long-Distance Runners.

The limited sample size in gait studies has hampered progress in the field. This challenge could be addressed through multicenter studies, thereby leveraging data sets from different laboratories. This study compared 3-dimensional lower-extremity running kinematics between the Biomechanics and Motor Control Laboratory, Federal University of ABC (Brazil), and the Running Injury Clinic, University of Calgary (Canada). Three-dimensional lower-extremity kinematics from 23 male runners were collected from each laboratory using comparable instrumentation and experimental procedures. The 3-dimensional hip, knee, and ankle angles were compared within and between centers using root-mean-square deviation. Two-sample t tests Statistical Parametric Mapping tested the hypothesis that the data from both laboratories were not different. The sagittal plane hip, knee, and ankle angles were similar between laboratories, while notable differences were observed for frontal (hip and ankle) and transverse (hip and knee) plane angles. The average interlaboratory root-mean-square deviation (2.6°) was lower than the intralaboratory root-mean-square deviation (Biomechanics and Motor Control = 4.8°, Running Injury Clinic = 5.6°), with the ankle transverse angle displaying the smallest, and the knee transverse angle displaying the largest variability. This study demonstrates the potential of combining gait kinematics data from different laboratories to increase sample size, but frontal and transverse plane data should be considered with caution.

A multicenter study on deep learning for glioblastoma auto-segmentation with prior knowledge in multimodal imaging.

A precise radiotherapy plan is crucial to ensure accurate segmentation of glioblastomas (GBMs) for radiation therapy. However, the traditional manual segmentation process is labor-intensive and heavily reliant on the experience of radiation oncologists. In this retrospective study, a novel auto-segmentation method is proposed to address these problems. To assess the method's applicability across diverse scenarios, we conducted its development and evaluation using a cohort of 148 eligible patients drawn from four multicenter datasets and retrospective data collection including noncontrast CT, multisequence MRI scans, and corresponding medical records. All patients were diagnosed with histologically confirmed high-grade glioma (HGG). A deep learning-based method (PKMI-Net) for automatically segmenting gross tumor volume (GTV) and clinical target volumes (CTV1 and CTV2) of GBMs was proposed by leveraging prior knowledge from multimodal imaging. The proposed PKMI-Net demonstrated high accuracy in segmenting, respectively, GTV, CTV1, and CTV2 in an 11-patient test set, achieving Dice similarity coefficients (DSC) of 0.94, 0.95, and 0.92; 95% Hausdorff distances (HD95) of 2.07, 1.18, and 3.95 mm; average surface distances (ASD) of 0.69, 0.39, and 1.17 mm; and relative volume differences (RVD) of 5.50%, 9.68%, and 3.97%. Moreover, the vast majority of GTV, CTV1, and CTV2 produced by PKMI-Net are clinically acceptable and require no revision for clinical practice. In our multicenter evaluation, the PKMI-Net exhibited consistent and robust generalizability across the various datasets, demonstrating its effectiveness in automatically segmenting GBMs. The proposed method using prior knowledge in multimodal imaging can improve the contouring accuracy of GBMs, which holds the potential to improve the quality and efficiency of GBMs' radiotherapy.

Repeatability and reproducibility of deep learning features for lung adenocarcinoma subtypes with nodules less than 10 mm in size: a multicenter thin-slice computed tomography phantom and clinical validation study.

Background: Deep learning features (DLFs) derived from radiomics features (RFs) fused with deep learning have shown potential in enhancing diagnostic capability. However, the limited repeatability and reproducibility of DLFs across multiple centers represents a challenge in the clinically validation of these features. This study thus aimed to evaluate the repeatability and reproducibility of DLFs and their potential efficiency in differentiating subtypes of lung adenocarcinoma less than 10 mm in size and manifesting as ground-glass nodules (GGNs). Methods: A chest phantom with nodules was scanned repeatedly using different thin-slice computed tomography (TSCT) scanners with varying acquisition and reconstruction parameters. The robustness of the DLFs was measured using the concordance correlation coefficient (CCC) and intraclass correlation coefficient (ICC). A deep learning approach was used for visualizing the DLFs. To assess the clinical effectiveness and generalizability of the stable and informative DLFs, three hospitals were used to source 275 patients, in whom 405 nodules were pathologically differentially diagnosed as GGN lung adenocarcinoma less than 10 mm in size and were retrospectively reviewed for clinical validation. Results: A total of 64 DLFs were analyzed, which revealed that the variables of slice thickness and slice interval (ICC, 0.79±0.18) and reconstruction kernel (ICC, 0.82±0.07) were significantly associated with the robustness of DLFs. Feature visualization showed that the DLFs were mainly focused around the nodule areas. In the external validation, a subset of 28 robust DLFs identified as stable under all sources of variability achieved the highest area under curve [AUC =0.65, 95% confidence interval (CI): 0.53-0.76] compared to other DLF models and the radiomics model. Conclusions: Although different manufacturers and scanning schemes affect the reproducibility of DLFs, certain DLFs demonstrated excellent stability and effectively improved diagnostic the efficacy for identifying subtypes of lung adenocarcinoma. Therefore, as the first step, screening stable DLFs in multicenter DLFs research may improve diagnostic efficacy and promote the application of these features.

Developing a Machine-Learning Prediction Model for Infliximab Response in Crohn's Disease: Integrating Clinical Characteristics and Longitudinal Laboratory Trends.

BACKGROUND: Achieving long-term clinical remission in Crohn's disease (CD) with antitumor necrosis factor α (anti-TNF-α) agents remains challenging. AIMS: This study aims to establish a prediction model based on patients' clinical characteristics using a machine-learning approach to predict the long-term efficacy of infliximab (IFX). METHODS: Three cohorts comprising 746 patients with CD were included from 3 inflammatory bowel disease (IBD) centers between June 2013 and January 2022. Clinical records were collected from baseline, 14-, 30-, and 52-week post-IFX treatment. Three machine-learning approaches were employed to develop predictive models based on 23 baseline predictors. The SHapley Additive exPlanations (SHAP) algorithm was used to dissect underlying predictors, and latent class mixed model (LCMM) was applied for trajectory analysis of the longitudinal change of blood routine tests along with long-term IFX therapy. RESULTS: The XGBoost model exhibited the best discrimination between long-term responders and nonresponders. In the internal training and testing set, the model achieved an AUC of 0.91 (95% CI, 0.86-0.95) and 0.71 (95% CI, 0.66-0.87), respectively. Moreover, it achieved a moderate predictive performance in the independent external cohort, with an AUC of 0.68 (95% CI, 0.59-0.77). The SHAP algorithm revealed disease-relevant laboratory measurements, notably hemoglobin (HB), white blood cells (WBC), erythrocyte sedimentation rate (ESR), albumin (ALB), and platelets (PLT), alongside age at diagnosis and the Montreal classification, as the most influential predictors. Furthermore, 2 distinct patient clusters based on dynamic laboratory tests were identified for monitoring the long-term remission. CONCLUSIONS: The established prediction model demonstrated remarkable discriminatory power in distinguishing long-term responders from nonresponders to IFX therapy. The identification of distinct patient clusters further emphasizes the need for tailored therapeutic approaches in CD management.

The study developed a machine-learning model using clinical data to predict long-term efficacy of IFX in Crohn's disease. The XGBoost model demonstrated strong discriminatory power, revealing influential predictors and distinct patient clusters, emphasizing the importance of tailored therapeutic approaches in CD management.

Ocular Disease Detection with Deep Learning (Fine-Grained Image Categorization) Applied to Ocular B-Scan Ultrasound Images.

INTRODUCTION: The aim of this work is to develop a deep learning (DL) system for rapidly and accurately screening for intraocular tumor (IOT), retinal detachment (RD), vitreous hemorrhage (VH), and posterior scleral staphyloma (PSS) using ocular B-scan ultrasound images. METHODS: Ultrasound images from five clinically confirmed categories, including vitreous hemorrhage, retinal detachment, intraocular tumor, posterior scleral staphyloma, and normal eyes, were used to develop and evaluate a fine-grained classification system (the Dual-Path Lesion Attention Network, DPLA-Net). Images were derived from five centers scanned by different sonographers and divided into training, validation, and test sets in a ratio of 7:1:2. Two senior ophthalmologists and four junior ophthalmologists were recruited to evaluate the system's performance. RESULTS: This multi-center cross-sectional study was conducted in six hospitals in China. A total of 6054 ultrasound images were collected; 4758 images were used for the training and validation of the system, and 1296 images were used as a testing set. DPLA-Net achieved a mean accuracy of 0.943 in the testing set, and the area under the curve was 0.988 for IOT, 0.997 for RD, 0.994 for PSS, 0.988 for VH, and 0.993 for normal. With the help of DPLA-Net, the accuracy of the four junior ophthalmologists improved from 0.696 (95% confidence interval [CI] 0.684-0.707) to 0.919 (95% CI 0.912-0.926, p < 0.001), and the time used for classifying each image reduced from 16.84 ± 2.34 s to 10.09 ± 1.79 s. CONCLUSIONS: The proposed DPLA-Net showed high accuracy for screening and classifying multiple ophthalmic diseases using B-scan ultrasound images across mutiple centers. Moreover, the system can promote the efficiency of classification by ophthalmologists.

One-year outcomes of a novel venous stent for symptomatic iliofemoral venous obstruction: prospective cohort study.

BACKGROUND: A stent with characteristics of a hybrid design may have advantages in improving the patency of symptomatic iliofemoral vein obstruction. This study assessed the safety and effectiveness of the V-Mixtent Venous Stent in treating symptomatic iliofemoral outflow obstruction. METHODS: Eligible patients had a Clinical-Etiologic-Anatomic-Physiologic (CEAP) C classification of ≥ 3 or a Venous Clinical Severity Score (VCSS) pain score of ≥ 2. The primary safety endpoint was the rate of major adverse events within 30 days. The primary effectiveness endpoint was the 12-month primary patency rate. Secondary endpoints included changes in VCSS from baseline to 6 and 12 months, alterations in CEAP C classification, Chronic Venous Disease Quality of Life Questionnaire (CIVIQ-14) scores at 12 months, and stent durability measures. RESULTS: Between December 2020 and November 2021, 171 patients were enrolled across 15 institutions. A total of 185 endovenous stents were placed, with 91.81% of subjects receiving one stent and 8.19% receiving 2 stents. Within 30 days, only two major adverse events occurred (1.17%; 95% confidence interval [CI], 0.14-4.16%), below the literature-defined performance goal of 11% (P < .001). The 12-month primary patency rate (91.36%; 95% CI, 85.93-95.19%; P < .001) exceeded the literature-defined performance goal. VCSS changes from baseline demonstrated clinical improvement at 6 months (- 4.30 ± 3.66) and 12 months (- 4.98 ± 3.67) (P < .001). Significant reduction in symptoms, as measured by CEAP C classification and CIVIQ-14, was observed from pre-procedure to 12 months (P < .001). CONCLUSIONS: The 12-month outcomes confirm the safety and effectiveness of the V-Mixtent Venous Stent in managing symptomatic iliofemoral venous outflow obstruction, including clinical symptom improvement compared to before treatment.

Robot-assisted radical nephroureterectomy using the KangDuo Surgical Robot-01 Sys-tem versus the da Vinci System: a multicenter prospective randomized controlled trial.

INTRODUCTION: We aim to compare the safety and effectiveness of the KangDuo (KD)-Surgical Robot-01 (KD-SR-01) system and the da Vinci (DV) system for robot-assisted radical nephroureterectomy (RARNU). MATERIALS AND METHODS: This multicenter prospective randomized controlled trial was conducted between March 2022 and September 2023. Group 1 included 29 patients undergoing KD-RARNU. Group 2 included 29 patients undergoing DV-RARNU. Patient demographic and clinical characteristics, perioperative data, and follow-up outcomes were collected prospectively and compared between the two groups. RESULTS: There were no significant differences in patient baseline demographic and preoperative characteristics between the two groups. The success rates in both groups were 100% without conversion to open or laparoscopic surgery or positive surgical margins. No significant difference was observed in docking time [242 (120-951) s vs 253 (62-498) s, P = 0.780], console time [137 (55-290) min vs 105 (62-220) min, P = 0.114], operative time [207 (121-460) min vs 185 (96-305) min, P = 0.091], EBL [50 (10-600) mL vs 50 (10-700) mL, P = 0.507], National Aeronautics and Space Administration Task Load Index scores, and postoperative serum creatinine levels between the two groups. None of the patients showed evidence of distant metastasis, local recurrence, or equipment-related adverse events during the four-week follow-up. One (3.4%) patient in Group 2 experienced postoperative enterovaginal and enterovesical fistulas (Clavien-Dindo grade III). CONCLUSIONS: The KD-SR-01 system is safe and effective for RARNU compared to the DV Si or Xi system. Further randomized controlled studies with larger sample sizes and longer durations are required.

Oncologic Outcomes of Patients with Immune Checkpoint Inhibitor Resistant Urothelial Carcinoma Treated with Enfortumab Vedotin and the Impact of Neutrophil-to-Lymphocyte Ratio and Dysgeusia on Overall Survival: A Retrospective Multicenter Cohort Study in Japan.

Randomized phase III trial results have demonstrated enfortumab vedotin (EV), an antibody-drug conjugate (ADC) consisting of an anti-Nectin-4 human IgG1 monoclonal antibody and monomethyl auristatin E, is a useful treatment for patients with locally advanced or metastatic urothelial carcinoma (la/mUC) that progressed after immune checkpoint inhibitor (ICI) therapies. This multicenter retrospective cohort study aimed to identify predictive factors for the efficacy of EV therapy and prolonged overall survival (OS) of patients in clinical practice. This study included patients with la/mUC who received ICI treatment. Patients who subsequently received EV treatment, those who received non-EV chemotherapy, and those who received no treatment were defined as EV, non-EV, and best supportive care (BSC) groups, respectively. The median OS was 20, 15, and 7 months in the EV, non-EV, and BSC groups, respectively (p < 0.001). Patients with la/mUC who had a complete or partial response after EV treatment had a significantly prolonged OS compared with those with stable or progressive disease. Univariate analysis showed age, neutrophil-to-lymphocyte ratio (NLR), dysgeusia, and rash as independent predictors of OS improvement. NLR and dysgeusia were independent predictors of OS after EV in multivariate analysis. Patients without these factors had a significantly prolonged OS compared to those with both factors. In real-world practice, EV therapy is an effective treatment for patients with la/mUC after ICI treatment.

How to customize common data models for rare diseases: an OMOP-based implementation and lessons learned.

BACKGROUND: Given the geographical sparsity of Rare Diseases (RDs), assembling a cohort is often a challenging task. Common data models (CDM) can harmonize disparate sources of data that can be the basis of decision support systems and artificial intelligence-based studies, leading to new insights in the field. This work is sought to support the design of large-scale multi-center studies for rare diseases. METHODS: In an interdisciplinary group, we derived a list of elements of RDs in three medical domains (endocrinology, gastroenterology, and pneumonology) according to specialist knowledge and clinical guidelines in an iterative process. We then defined a RDs data structure that matched all our data elements and built Extract, Transform, Load (ETL) processes to transfer the structure to a joint CDM. To ensure interoperability of our developed CDM and its subsequent usage for further RDs domains, we ultimately mapped it to Observational Medical Outcomes Partnership (OMOP) CDM. We then included a fourth domain, hematology, as a proof-of-concept and mapped an acute myeloid leukemia (AML) dataset to the developed CDM. RESULTS: We have developed an OMOP-based rare diseases common data model (RD-CDM) using data elements from the three domains (endocrinology, gastroenterology, and pneumonology) and tested the CDM using data from the hematology domain. The total study cohort included 61,697 patients. After aligning our modules with those of Medical Informatics Initiative (MII) Core Dataset (CDS) modules, we leveraged its ETL process. This facilitated the seamless transfer of demographic information, diagnoses, procedures, laboratory results, and medication modules from our RD-CDM to the OMOP. For the phenotypes and genotypes, we developed a second ETL process. We finally derived lessons learned for customizing our RD-CDM for different RDs. DISCUSSION: This work can serve as a blueprint for other domains as its modularized structure could be extended towards novel data types. An interdisciplinary group of stakeholders that are actively supporting the project's progress is necessary to reach a comprehensive CDM. CONCLUSION: The customized data structure related to our RD-CDM can be used to perform multi-center studies to test data-driven hypotheses on a larger scale and take advantage of the analytical tools offered by the OHDSI community.

Effect and Safety of Apatinib as Neoadjuvant Therapy in Locally Advanced Differentiated Thyroid Cancer: A Phase 2 Trial.

Context: Presently, there is a paucity of prospective clinical trials investigating neoadjuvant therapy for locally advanced thyroid cancer. Objective: This study was a multicenter, open-label, single-arm, phase II trial evaluating the efficacy and safety of apatinib as neoadjuvant therapy in patients with local advanced differentiated thyroid cancer (DTC). Methods: Patients were treated with preoperative apatinib over a course of 2 to 4 cycles, culminating in surgical resection. The primary endpoints were objective response rate (ORR) and disease control rate (DCR); the secondary endpoints were the rate of R0 surgery, alterations in serum thyroglobulin levels, disease-free survival, and adverse events (AEs). Results: A total of 14 patients who met the inclusion criteria were administered neoadjuvant apatinib. Among these, 13 patients underwent surgical procedures following apatinib treatment and were enrolled in the ITT population. The ORR was 53.8% and the DCR was 100%. Of the patients, 84.6% received R0 surgery, while the remaining 15.4% underwent R1 resection. Predominant among the observed AEs were hypertension, hand-foot syndrome, hepatic dysfunction, proteinuria, and hypothyroidism, with no instances of grade 4 or 5 AEs reported. Subsequent to surgery, patients were followed up for a median period of 34 months, during which disease progression occurred in 5 individuals (35.7%), encompassing 3 cases of locoregional recurrences and 2 cases of distant metastases. Conclusion: Apatinib may be an effective agent in the use of neoadjuvant therapy for locally advanced DTC. Patients may therefore benefit from surgical outcomes and their long-term prognosis.

Effects of high- versus low-intensity lipid-lowering treatment in patients undergoing serial coronary computed tomography angiography: results of the multi-center LOCATE study.

AIM: To evaluate the effects of lipid-lowering medications of different intensities on total, calcified, and non-calcified plaque volumes in patients undergoing serial cardiac computed tomography angiography (CCTA). METHODS: Individuals with chronic coronary syndromes from 11 centers were included in a retrospective registry. Total, calcified, and non-calcified plaque volumes were quantified and the relative difference in plaque volumes between baseline and follow-up CCTA was calculated. The intensity of lipid-lowering treatment was designated as low, moderate, or high, based on current recommendations. RESULTS: Of 216 patients (mean age 63.1 ± 9.7 years), undergoing serial CCTA (median timespan = 824.5 [IQR = 463.0-1323.0] days), 89 (41.2%) received no or low-intensity lipid-lowering medications, and 80 (37.0%) and 47 (21.8%) moderate- and high-intensity lipid-lowering agents, respectively. Progression of total and non-calcified plaque was attenuated in patients on moderate-/high- versus those on no/low-intensity treatment and arrested in patients treated with high-intensity statins or PCSK9 inhibitors (p < 0.001). Halted increase of non-calcified plaque was associated with LDL-cholesterol reduction (p < 0.001), whereas calcified plaque mass and Agatston score increased irrespective of the lipid-lowering treatment (p = NS). The intensity of lipid-lowering therapy robustly predicted attenuation of non-calcified plaque progression as a function of the time duration between the two CCTA scans, and this was independent of age and cardiovascular risk factors (HR = 3.83, 95% CI = 1.81-8.05, p < 0.001). CONCLUSION: The LOCATE multi-center observational study shows that progression of non-calcified plaques, which have been previously described as precursors of acute coronary syndromes, can be attenuated with moderate-intensity, and arrested with high-intensity lipid-lowering therapy. GERMAN CLINICAL TRIALS REGISTER: DRKS00031954.

Long-term Durability Between Parent and Child Patient Reported Outcomes in Eosinophilic Esophagitis.

BACKGROUND: Because young children cannot self-report symptoms, there is a need for parent surrogate reports. While early work suggested parent child alignment for eosinophil esophagitis (EoE) patient reported outcomes (PROs), the longitudinal alignment is unclear. OBJECTIVE: To assess the agreement and longitudinal stability of PROs between children with EoE and their parents. METHODS: 292 parent-child respondents completed 723 completed questionnaires over 5 years in an observational trial in the Consortium of Eosinophilic Gastrointestinal Disease Researchers. The change in and agreement between parent and child Pediatric Eosinophilic Esophagitis Symptom Score version 2 (PEESSv2.0) and Pediatric Quality of Life Eosinophilic Esophagitis Module (PedsQL-EoE) PROs over time were assessed using Pearson correlation and Bland-Altman analyses. Clinical factors influencing PROs and their agreement were evaluated using linear mixed models. RESULTS: The cohort had a median disease duration equalling 3.7 years and was predominantly male (73.6%) and white (85.3%). Child and parent PEESSv2.0 response groups were identified and were stable over time. There was strong correlation between child and parent report (PEESSv2.0 0.83, PedsQL-EoE 0.74) with minimal pairwise differences for symptoms. Longitudinally, parent-reported PedsQL-EoE scores were stable (p ≥ 0.32), whereas child-reported PedsQL-EoE scores improved (p = 0.026). A larger difference in parent and child PedsQL-EoE reports was associated with younger age (p < 0.001) and differences were driven by psychosocial PRO domains. CONCLUSION: There is strong longitudinal alignment between child and parent report using EoE PROs. These data provide evidence that parent report is a stable proxy for objective EoE symptoms in their children.

Non-linear relationship between red blood cell distribution width and gastrointestinal bleeding risk in stroke patients: results from multi-center ICUs.

Background: The red blood cell distribution width (RDW) is closely linked to the prognosis of multiple diseases. However, the connection between RDW and gastrointestinal bleeding (GIB) in stroke patients is not well understood. This study aimed to clarify this association. Methods: This retrospective study involved 11,107 hospitalized patients from 208 hospitals in the United States, admitted between January 1, 2014, and December 31, 2015. We examined clinical data from 7,512 stroke patients in the intensive care unit (ICU). Multivariate logistic regression assessed the link between RDW and in-hospital GIB in stroke patients. Generalized additive model (GAM) and smooth curve fitting (penalty spline method) were utilized to explore the non-linear relationship between RDW and GIB in stroke patients. The inflection point was calculated using a recursive algorithm, and interactions between different variables were assessed through subgroup analyses. Results: Among the 11,107 screened stroke patients, 7,512 were included in the primary analysis, with 190 identified as having GIB. The participants had a mean age of (61.67 ± 12.42) years, and a median RDW of 13.9%. Multiple logistic analysis revealed RDW as a risk factor for in-hospital GIB in stroke patients (OR = 1.28, 95% CI 1.21, 1.36, p < 0.05). The relationship between RDW and in-hospital GIB in stroke patients was found to be non-linear. Additionally, the inflection point of RDW was 14.0%. When RDW was ≥14.0%, there was a positive association with the risk of GIB (OR: 1.24, 95% CI: 1.16, 1.33, p < 0.0001). Conversely, when RDW was <14.0%, this association was not significant (OR: 1.02, 95% CI: 0.97-1.07, p = 0.4040). Conclusion: This study showed a substantial non-linear link between RDW and the risk of GIB in stroke patients. Maintaining the patient's RDW value below 14.0% could lower the risk of in-hospital GIB.

The Cancer Clinical Library Database (CCLD) from the Korea-Clinical Data Utilization Network for Research Excellence (K-CURE) Project.

The common data model (CDM) has found widespread application in healthcare studies, but its utilization in cancer research has been limited. This article describes the development and implementation strategy for Cancer Clinical Library Databases (CCLDs), which are standardized cancer-specific databases established under the Korea-Clinical Data Utilization Network for Research Excellence (K-CURE) project by the Korean Ministry of Health and Welfare. Fifteen leading hospitals and fourteen academic associations in Korea are engaged in constructing CCLDs for 10 primary cancer types. For each cancer type-specific CCLD, cancer data experts determine key clinical data items essential for cancer research, standardize these items across cancer types, and create a standardized schema. Comprehensive clinical records covering diagnosis, treatment, and outcomes, with annual updates, are collected for each cancer patient in the target population, and quality control is based on six-sigma standards. To protect patient privacy, CCLDs follow stringent data security guidelines by pseudonymizing personal identification information and operating within a closed analysis environment. Researchers can apply for access to CCLD data through the K-CURE portal, which is subject to Institutional Review Board and Data Review Board approval. The CCLD is considered a pioneering standardized cancer-specific database, significantly representing Korea's cancer data. It is expected to overcome limitations of previous CDMs and provide a valuable resource for multicenter cancer research in Korea.

Performance of federated learning-based models in the Dutch TAVI population was comparable to central strategies and outperformed local strategies.

Background: Federated learning (FL) is a technique for learning prediction models without sharing records between hospitals. Compared to centralized training approaches, the adoption of FL could negatively impact model performance. Aim: This study aimed to evaluate four types of multicenter model development strategies for predicting 30-day mortality for patients undergoing transcatheter aortic valve implantation (TAVI): (1) central, learning one model from a centralized dataset of all hospitals; (2) local, learning one model per hospital; (3) federated averaging (FedAvg), averaging of local model coefficients; and (4) ensemble, aggregating local model predictions. Methods: Data from all 16 Dutch TAVI hospitals from 2013 to 2021 in the Netherlands Heart Registration (NHR) were used. All approaches were internally validated. For the central and federated approaches, external geographic validation was also performed. Predictive performance in terms of discrimination [the area under the ROC curve (AUC-ROC, hereafter referred to as AUC)] and calibration (intercept and slope, and calibration graph) was measured. Results: The dataset comprised 16,661 TAVI records with a 30-day mortality rate of 3.4%. In internal validation the AUCs of central, local, FedAvg, and ensemble models were 0.68, 0.65, 0.67, and 0.67, respectively. The central and local models were miscalibrated by slope, while the FedAvg and ensemble models were miscalibrated by intercept. During external geographic validation, central, FedAvg, and ensemble all achieved a mean AUC of 0.68. Miscalibration was observed for the central, FedAvg, and ensemble models in 44%, 44%, and 38% of the hospitals, respectively. Conclusion: Compared to centralized training approaches, FL techniques such as FedAvg and ensemble demonstrated comparable AUC and calibration. The use of FL techniques should be considered a viable option for clinical prediction model development.

Ruyi Zhenbao Tablet and Baimai Ointment Therapy in Acute Ischemic Stroke: A Study Protocol for a Multi-Center, Randomized, Double-blinded, and Parallel-Controlled Trial.

Introduction: Stroke is characterized by high incidence, recurrence rate, and mortality. Patients with acute ischemic stroke (AIS) who are ineligible for acute revascularization therapy require more effective medication treatments. A previous clinical study showed that Ruyi Zhenbao tablets and Baimai ointments might be effective against AIS; however, high-quality clinical evidence supporting their application in AIS is lacking. To explore the efficacy of the two classic Tibetan medicines in the treatment of AIS, a randomized clinical trial will be conducted in patients with AIS who are not eligible for thrombolytic treatment. Methods: A prospective, randomized, multiple-center, double-blinded, placebo-controlled, and parallel-group trial will be conducted. We shall randomize 480 eligible participants to either the intervention or the control group. The distribution ratio of each group will be 1:1:1:1, including 120 patients each in the dual-medication group, the Baimai ointment group, the Ruyi Zhenbao tablet group, and the placebo group. Participants will be treated with medication for 8 weeks, and they will receive three follow-up visits: at 4 weeks (D29), 8 weeks (D56), and 90 days (D90) after commencing treatment. The primary outcome will be D90 change in the simplified Fugl-Meyer score from baseline to posttreatment. The secondary outcomes are as follows: D29 change of simplified Fugl-Meyer score from baseline to posttreatment; proportion of participants whose D29 NIHSS scores decreased by four or more points from baseline D90 proportion of subjects with mRS score of 0-2 (inclusive); D90 proportion of subjects with Barthel index score ≥95; D90 incidence of cardiovascular and cerebrovascular events. Safety endpoint includes mortality within 90 days; proportion of subjects with adverse events/serious adverse events within 90 days. Conclusion: This research protocol lays a solid groundwork for its practical execution. This study is poised to serve as a reference for other Tibetan medicine researchers, contributing to the reduction of stroke-related expenditures globally and, in turn, benefiting a broader population of stroke patients.

Automated interpretation of retinal vein occlusion based on fundus fluorescein angiography images using deep learning: A retrospective, multi-center study.

Purpose: Fundus fluorescein angiography (FFA) is the gold standard for retinal vein occlusion (RVO) diagnosis. This study aims to develop a deep learning-based system to diagnose and classify RVO using FFA images, addressing the challenges of time-consuming and variable interpretations by ophthalmologists. Methods: 4028 FFA images of 467 eyes from 463 patients were collected and annotated. Three convolutional neural networks (CNN) models (ResNet50, VGG19, InceptionV3) were trained to generate the label of image quality, eye, location, phase, lesions, diagnosis, and macular involvement. The performance of the models was evaluated by accuracy, precision, recall, F-1 score, the area under the curve, confusion matrix, human-machine comparison, and Clinical validation on three external data sets. Results: The InceptionV3 model outperformed ResNet50 and VGG19 in labeling and interpreting FFA images for RVO diagnosis, achieving 77.63%-96.45% accuracy for basic information labels and 81.72%-96.45% for RVO-relevant labels. The comparison between the best CNN and ophthalmologists showed up to 19% accuracy improvement with the inceptionV3. Conclusion: This study developed a deep learning model capable of automatically multi-label and multi-classification of FFA images for RVO diagnosis. The proposed system is anticipated to serve as a new tool for diagnosing RVO in places short of medical resources.

Time Series AI Model for Acute Kidney Injury Detection Based on a Multicenter Distributed Research Network: Development and Verification Study.

Background: Acute kidney injury (AKI) is a marker of clinical deterioration and renal toxicity. While there are many studies offering prediction models for the early detection of AKI, those predicting AKI occurrence using distributed research network (DRN)-based time series data are rare. Objective: In this study, we aimed to detect the early occurrence of AKI by applying an interpretable long short-term memory (LSTM)-based model to hospital electronic health record (EHR)-based time series data in patients who took nephrotoxic drugs using a DRN. Methods: We conducted a multi-institutional retrospective cohort study of data from 6 hospitals using a DRN. For each institution, a patient-based data set was constructed using 5 drugs for AKI, and an interpretable multivariable LSTM (IMV-LSTM) model was used for training. This study used propensity score matching to mitigate differences in demographics and clinical characteristics. Additionally, the temporal attention values of the AKI prediction model's contribution variables were demonstrated for each institution and drug, with differences in highly important feature distributions between the case and control data confirmed using 1-way ANOVA. Results: This study analyzed 8643 and 31,012 patients with and without AKI, respectively, across 6 hospitals. When analyzing the distribution of AKI onset, vancomycin showed an earlier onset (median 12, IQR 5-25 days), and acyclovir was the slowest compared to the other drugs (median 23, IQR 10-41 days). Our temporal deep learning model for AKI prediction performed well for most drugs. Acyclovir had the highest average area under the receiver operating characteristic curve score per drug (0.94), followed by acetaminophen (0.93), vancomycin (0.92), naproxen (0.90), and celecoxib (0.89). Based on the temporal attention values of the variables in the AKI prediction model, verified lymphocytes and calcvancomycin ium had the highest attention, whereas lymphocytes, albumin, and hemoglobin tended to decrease over time, and urine pH and prothrombin time tended to increase. Conclusions: Early surveillance of AKI outbreaks can be achieved by applying an IMV-LSTM based on time series data through an EHR-based DRN. This approach can help identify risk factors and enable early detection of adverse drug reactions when prescribing drugs that cause renal toxicity before AKI occurs.