Objective: This study aimed to develop and validate a deep learning (DL) algorithm for the quantitative measurement of thoracolumbar (TL) fracture features, and to evaluate its efficacy across varying levels of clinical expertise. Methods: Using the pretrained Mask Region-Based Convolutional Neural Networks model, originally developed for vertebral body segmentation and fracture detection, we fine-tuned the model and added a new module for measuring fracture metrics—compression rate (CR), Cobb angle (CA), Gardner angle (GA), and sagittal index (SI)—from lumbar spine lateral radiographs. These metrics were derived from six-point labeling by 3 radiologists, forming the ground truth (GT). Training utilized 1,000 nonfractured and 318 fractured radiographs, while validations employed 213 internal and 200 external fractured radiographs. The accuracy of the DL algorithm in quantifying fracture features was evaluated against GT using the intraclass correlation coefficient. Additionally, 4 readers with varying expertise levels, including trainees and an attending spine surgeon, performed measurements with and without DL assistance, and their results were compared to GT and the DL model. Results: The DL algorithm demonstrated good to excellent agreement with GT for CR, CA, GA, and SI in both internal (0.860, 0.944, 0.932, and 0.779, respectively) and external (0.836, 0.940, 0.916, and 0.815, respectively) validations. DL-assisted measurements significantly improved most measurement values, particularly for trainees. Conclusion The DL algorithm was validated as an accurate tool for quantifying TL fracture features using radiographs. DL-assisted measurement is expected to expedite the diagnostic process and enhance reliability, particularly benefiting less experienced clinicians.

Objective: This study aimed to develop and validate a deep learning (DL) algorithm for the quantitative measurement of thoracolumbar (TL) fracture features, and to evaluate its efficacy across varying levels of clinical expertise. Methods: Using the pretrained Mask Region-Based Convolutional Neural Networks model, originally developed for vertebral body segmentation and fracture detection, we fine-tuned the model and added a new module for measuring fracture metrics—compression rate (CR), Cobb angle (CA), Gardner angle (GA), and sagittal index (SI)—from lumbar spine lateral radiographs. These metrics were derived from six-point labeling by 3 radiologists, forming the ground truth (GT). Training utilized 1,000 nonfractured and 318 fractured radiographs, while validations employed 213 internal and 200 external fractured radiographs. The accuracy of the DL algorithm in quantifying fracture features was evaluated against GT using the intraclass correlation coefficient. Additionally, 4 readers with varying expertise levels, including trainees and an attending spine surgeon, performed measurements with and without DL assistance, and their results were compared to GT and the DL model. Results: The DL algorithm demonstrated good to excellent agreement with GT for CR, CA, GA, and SI in both internal (0.860, 0.944, 0.932, and 0.779, respectively) and external (0.836, 0.940, 0.916, and 0.815, respectively) validations. DL-assisted measurements significantly improved most measurement values, particularly for trainees. Conclusion The DL algorithm was validated as an accurate tool for quantifying TL fracture features using radiographs. DL-assisted measurement is expected to expedite the diagnostic process and enhance reliability, particularly benefiting less experienced clinicians.

Objective: This study aimed to develop and validate a deep learning (DL) algorithm for the quantitative measurement of thoracolumbar (TL) fracture features, and to evaluate its efficacy across varying levels of clinical expertise. Methods: Using the pretrained Mask Region-Based Convolutional Neural Networks model, originally developed for vertebral body segmentation and fracture detection, we fine-tuned the model and added a new module for measuring fracture metrics—compression rate (CR), Cobb angle (CA), Gardner angle (GA), and sagittal index (SI)—from lumbar spine lateral radiographs. These metrics were derived from six-point labeling by 3 radiologists, forming the ground truth (GT). Training utilized 1,000 nonfractured and 318 fractured radiographs, while validations employed 213 internal and 200 external fractured radiographs. The accuracy of the DL algorithm in quantifying fracture features was evaluated against GT using the intraclass correlation coefficient. Additionally, 4 readers with varying expertise levels, including trainees and an attending spine surgeon, performed measurements with and without DL assistance, and their results were compared to GT and the DL model. Results: The DL algorithm demonstrated good to excellent agreement with GT for CR, CA, GA, and SI in both internal (0.860, 0.944, 0.932, and 0.779, respectively) and external (0.836, 0.940, 0.916, and 0.815, respectively) validations. DL-assisted measurements significantly improved most measurement values, particularly for trainees. Conclusion The DL algorithm was validated as an accurate tool for quantifying TL fracture features using radiographs. DL-assisted measurement is expected to expedite the diagnostic process and enhance reliability, particularly benefiting less experienced clinicians.

Objective: This study aimed to develop and validate a deep learning (DL) algorithm for the quantitative measurement of thoracolumbar (TL) fracture features, and to evaluate its efficacy across varying levels of clinical expertise. Methods: Using the pretrained Mask Region-Based Convolutional Neural Networks model, originally developed for vertebral body segmentation and fracture detection, we fine-tuned the model and added a new module for measuring fracture metrics—compression rate (CR), Cobb angle (CA), Gardner angle (GA), and sagittal index (SI)—from lumbar spine lateral radiographs. These metrics were derived from six-point labeling by 3 radiologists, forming the ground truth (GT). Training utilized 1,000 nonfractured and 318 fractured radiographs, while validations employed 213 internal and 200 external fractured radiographs. The accuracy of the DL algorithm in quantifying fracture features was evaluated against GT using the intraclass correlation coefficient. Additionally, 4 readers with varying expertise levels, including trainees and an attending spine surgeon, performed measurements with and without DL assistance, and their results were compared to GT and the DL model. Results: The DL algorithm demonstrated good to excellent agreement with GT for CR, CA, GA, and SI in both internal (0.860, 0.944, 0.932, and 0.779, respectively) and external (0.836, 0.940, 0.916, and 0.815, respectively) validations. DL-assisted measurements significantly improved most measurement values, particularly for trainees. Conclusion The DL algorithm was validated as an accurate tool for quantifying TL fracture features using radiographs. DL-assisted measurement is expected to expedite the diagnostic process and enhance reliability, particularly benefiting less experienced clinicians.

Objective: This study aimed to develop and validate a deep learning (DL) algorithm for the quantitative measurement of thoracolumbar (TL) fracture features, and to evaluate its efficacy across varying levels of clinical expertise. Methods: Using the pretrained Mask Region-Based Convolutional Neural Networks model, originally developed for vertebral body segmentation and fracture detection, we fine-tuned the model and added a new module for measuring fracture metrics—compression rate (CR), Cobb angle (CA), Gardner angle (GA), and sagittal index (SI)—from lumbar spine lateral radiographs. These metrics were derived from six-point labeling by 3 radiologists, forming the ground truth (GT). Training utilized 1,000 nonfractured and 318 fractured radiographs, while validations employed 213 internal and 200 external fractured radiographs. The accuracy of the DL algorithm in quantifying fracture features was evaluated against GT using the intraclass correlation coefficient. Additionally, 4 readers with varying expertise levels, including trainees and an attending spine surgeon, performed measurements with and without DL assistance, and their results were compared to GT and the DL model. Results: The DL algorithm demonstrated good to excellent agreement with GT for CR, CA, GA, and SI in both internal (0.860, 0.944, 0.932, and 0.779, respectively) and external (0.836, 0.940, 0.916, and 0.815, respectively) validations. DL-assisted measurements significantly improved most measurement values, particularly for trainees. Conclusion The DL algorithm was validated as an accurate tool for quantifying TL fracture features using radiographs. DL-assisted measurement is expected to expedite the diagnostic process and enhance reliability, particularly benefiting less experienced clinicians.

Fabry disease (FD) is an X-linked lysosomal storage disorder caused by the deficient activity of α-galactosidase (α-Gal A), affecting multiple organs including kidney. In this study, we aimed to determine the prevalence of FD in patients with chronic kidney disease (CKD) including those on renal replacement therapy in Korea. Methods: This is a national, multicenter, observational study performed between August 24, 2017 and February 28, 2020. Patients with the presence of proteinuria or treated on dialysis were screened by measuring the α-Gal A enzyme activity using either dried blood spot or whole blood, and plasma globotriaosylsphingosine (lyso-GL3) concentration. A GLA gene analysis was performed in patients with low α-Gal A enzyme activity or increased plasma lyso-GL3 concentration. Results: Of 897 screened patients, 405 (45.2%) were male and 279 (31.1%) were on dialysis. The α-Gal A enzyme activity was measured in 891 patients (99.3%), and plasma lyso-GL3 concentration was measured in all patients. Ten patients were eligible for a GLA gene analysis: eight with low α-Gal A enzyme activity and two with increased plasma lyso-GL3 concentration. The GLA mutations were analyzed in nine patients and one patient was found with a pathogenic mutation. Therefore, one patient was identified with FD, giving a prevalence of 0.1% (1 of 897) in this CKD population. Conclusion: Although the prevalence of FD in the CKD population was low (0.1%), screening tests are crucial to detect potential diseases in patients with relatives who can benefit from early treatment.

Coronary artery calcification (CAC) is highly prevalent in patients with chronic kidney disease (CKD) and is associated with major adverse cardiovascular events and metabolic disturbances. The triglyceride-glucose index (TyGI), a novel surrogate marker of metabolic syndrome and insulin resistance, is associated with CAC in the general population and in patients with diabetes. This study investigated the association between the TyGI and CAC progression in patients with CKD, which is unknown. Methods: A total of 1,154 patients with CKD (grades 1–5; age, 52.8 ± 11.9 years; male, 688 [59.6%]) were enrolled from the KNOWCKD (KoreaN Cohort Study for Outcomes in Patients With Chronic Kidney Disease). The TyGI was calculated as follows: ln (fasting triglycerides × fasting glucose/2). Patients were classified into tertiles (low, intermediate, high) based on the TyGI. The primary outcome was annualized percentage change in CAC score [(percent change in CAC score + 1)12/follow-up months – 1] of ≥15%, defined as CAC progression. Results: During the 4-year follow-up, the percentage of patients with CAC progression increased across TyGI groups (28.6%, 37.5%, and 46.2% in low, intermediate, and high groups, respectively; p < 0.001). A high TyGI was associated with an increased risk of CAC progression (odds ratio [OR], 2.11; 95% confidence interval [CI], 1.14–3.88; p = 0.02) compared to the low group. Moreover, a 1-point increase in the TyGI was related to increased risk of CAC progression (OR, 1.55; 95% CI, 1.06–1.76; p = 0.02) after adjustment. Conclusion: A high TyGI may be a useful predictor of CAC progression in CKD.

Objective: The aim of this study was to investigate the association between blood levels of tumor necrosis factor-α (TNF-α) and interleukin-6 (IL-6) and cognitive impairments among elderly individuals. Methods: Peripheral concentration of TNF-α and IL-6 were measured in all subjects. To assess individual cognitive function, the Consortium to Establish a Registry for Alzheimer’s Disease Neuropsychological Assessment Battery (CERAD-NP) was used, and standardized scores (z-scores) were calculated for each test. Cytokine levels were compared between the diagnostic groups, and correlations between blood inflammatory factor levels and z-scores were analyzed. Results: The 37 participants included 8 patients with Alzheimer’s disease (AD), 15 subjects with mild cognitive impairment (MCI), and 14 cognitively healthy controls. TNF-α and IL-6 levels were higher in patients with AD than in healthy controls. TNF-α levels were higher in the AD group than in the MCI group. However, after adjusting for age, the associations between diagnosis and TNF-α and IL-6 were not significant. The higher the plasma IL-6 level, the lower the z-scores on the Boston Naming Test, Word List Learning, Word List Recognition, and Constructional Recall. The higher the serum TNF-α level, the lower the z-scores on the Word List Learning and Constructional Recall. Negative correlation between serum TNF-α level and the z-score on Word List Learning remained significant when age was adjusted. Conclusion The difference in the blood levels of TNF-α and IL-6 between the diagnostic groups may be associated with aging. However, elevated TNF-α levels were associated with worse immediate memory performance, even after adjusting for age.

Background: The optimal level of glycosylated hemoglobin (HbA1c) to prevent adverse clinical outcomes is unknown in patients with chronic kidney disease (CKD) and type 2 diabetes mellitus (T2DM). Methods: We analyzed 707 patients with CKD G1-G5 without kidney replacement therapy and T2DM from the KoreaN Cohort Study for Outcome in Patients With Chronic Kidney Disease (KNOW-CKD), a nationwide prospective cohort study. The main predictor was time-varying HbA1c level at each visit. The primary outcome was a composite of development of major adverse cardiovascular events (MACEs) or all-cause mortality. Secondary outcomes included the individual endpoint of MACEs, all-cause mortality, and CKD progression. CKD progression was defined as a ≥50% decline in the estimated glomerular filtration rate from baseline or the onset of end-stage kidney disease. Results: During a median follow-up of 4.8 years, the primary outcome occurred in 129 (18.2%) patients. In time-varying Cox model, the adjusted hazard ratios (aHRs) for the primary outcome were 1.59 (95% confidence interval [CI], 1.01 to 2.49) and 1.99 (95% CI, 1.24 to 3.19) for HbA1c levels of 7.0%–7.9% and ≥8.0%, respectively, compared with <7.0%. Additional analysis of baseline HbA1c levels yielded a similar graded association. In secondary outcome analyses, the aHRs for the corresponding HbA1c categories were 2.17 (95% CI, 1.20 to 3.95) and 2.26 (95% CI, 1.17 to 4.37) for MACE, and 1.36 (95% CI, 0.68 to 2.72) and 2.08 (95% CI, 1.06 to 4.05) for all-cause mortality. However, the risk of CKD progression did not differ between the three groups. Conclusion This study showed that higher HbA1c levels were associated with an increased risk of MACE and mortality in patients with CKD and T2DM.

Purpose: Renal tumors account for approximately 7% of all childhood cancers. These include Wilms tumor (WT), clear cell sarcoma of the kidney (CCSK), malignant rhabdoid tumor of the kidney (MRTK), renal cell carcinoma (RCC), congenital mesoblastic nephroma (CMN) and other rare tumors. We investigated the epidemiology of pediatric renal tumors in Korea. Materials and Methods: From January 2001 to December 2015, data of pediatric patients (0–18 years) newly-diagnosed with renal tumors at 26 hospitals were retrospectively analyzed. Results: Among 439 patients (male, 240), the most common tumor was WT (n=342, 77.9%), followed by RCC (n=36, 8.2%), CCSK (n=24, 5.5%), MRTK (n=16, 3.6%), CMN (n=12, 2.7%), and others (n=9, 2.1%). Median age at diagnosis was 27.1 months (range 0-225.5) and median follow-up duration was 88.5 months (range 0-211.6). Overall, 32 patients died, of whom 17, 11, 1, and 3 died of relapse, progressive disease, second malignant neoplasm, and treatment-related mortality. Five-year overall survival and event free survival were 97.2% and 84.8% in WT, 90.6% and 82.1% in RCC, 81.1% and 63.6% in CCSK, 60.3% and 56.2% in MRTK, and 100% and 91.7% in CMN, respectively (p < 0.001). Conclusion The pediatric renal tumor types in Korea are similar to those previously reported in other countries. WT accounted for a large proportion and survival was excellent. Non-Wilms renal tumors included a variety of tumors and showed inferior outcome, especially MRTK. Further efforts are necessary to optimize the treatment and analyze the genetic characteristics of pediatric renal tumors in Korea.