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.

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.

Shared decision-making is a two-way symmetrical communication process in which clinicians and patients work together to achieve the best outcome. This study aimed to develop self-assessment items as a decision aid for choosing a dialysis modality in patients with chronic kidney disease (CKD) and to assess the construct validity of the newly developed items. Methods: Five focus group interviews were performed to extract specific self-assessment items regarding patient values in choosing a dialysis modality. After survey items were refined, a survey of 330 patients, consisting of 152 hemodialysis (HD) and 178 peritoneal dialysis (PD) patients, was performed to validate the self-assessment items. Results: The self-assessment for the decision aid was refined to 35 items. The structure of the final items appeared to have three dimensions of factors; health, lifestyle, and dialysis environment. The health factor consisted of 12 subscales (α = 0.724), the lifestyle factor contained 11 subscales (α = 0.624), and the dialysis environment factor was represented by 12 subscales (α = 0.694). A structural equation model analysis showed that the relationship between the decision aid factors (health, lifestyle, and dialysis environment), patients’ CKD perception, and cognition of shared decision-making differed between HD patients and PD patients. Conclusion: We developed and validated self-assessment items as part of a decision aid to help patients with CKD. This attempt may assist CKD patients in making informed and shared decisions closely aligned with their values when considering dialysis modality.

Purpose: Acute promyelocytic leukemia (APL) is a rare disease in children and there are some different characteristics between children and adult. We aimed to evaluate incidence, clinical characteristics and treatment outcomes of pediatric APL in Korea. Materials and Methods: Seventy-nine pediatric APL patients diagnosed from January 2009 to December 2016 in 16 tertiary medical centers in Korea were reviewed retrospectively. Results: Of 801 acute myeloid leukemia children, 79 (9.9%) were diagnosed with APL. The median age at diagnosis was 10.6 years (range, 1.3 to 18.0). Male and female ratio was 1:0.93. Thirty patients (38.0%) had white blood cell (WBC) count greater than 10×109/L at diagnosis. All patients received induction therapy consisting of all-trans retinoic acid and chemotherapy. Five patients (6.6%) died during induction chemotherapy and 66 patients (86.8%) achieved complete remission (CR) after induction chemotherapy. The causes of death were three intracranial hemorrhage, one cerebral infarction, and one sepsis. Five patients (7.1%) suffered a relapse during or after maintenance chemotherapy. The estimated 4-year event-free survival and overall survival (OS) rates were 82.1%±4.4%, 89.7%±5.1%, respectively. The 4-year OS was significantly higher in patients with initial WBC < 10×109/L than in those with initial WBC ≥ 10×109/L (p=0.020). Conclusion This study showed that the CR rates and survival outcomes in Korean pediatric APL patients were relatively good. The initial WBC count was the most important prognostic factor and most causes of death were related to serious bleeding in the early stage of treatment.

Deep electrical burn on the hand and forearm causes ischemic tissue damage due to increased compartment pressure by tight eschars. Early detection of ischemia and prompt release of eschars are necessary for prevention of ischemic tissue damage. Early escharotomy is a useful decompressive therapy for prevention of ischemia of the distal upper extremity.

The International IgA Nephropathy Prediction Tool has been recently developed to estimate the progression risk of immunoglobulin A nephropathy (IgAN). This study aimed to evaluate the clinical performance of this prediction tool in a large IgAN cohort in Korea. Methods: The study cohort was comprised of 2,064 patients with biopsy-proven IgAN from four medical centers between March 2012 and September 2021. We calculated the predicted risk for each patient. The primary outcome was occurrence of a 50% decline in estimated glomerular filtration rate (eGFR) from the time of biopsy or end-stage kidney disease. The model performance was evaluated for discrimination, calibration, and reclassification. We also constructed and tested an additional model with a new coefficient for the Korean race. Results: During a median follow-up period of 3.8 years (interquartile range, 1.8–6.6 years), 363 patients developed the primary outcome. The two prediction models exhibited good discrimination power, with a C-statistic of 0.81. The two models generally underestimated the risk of the primary outcome, with lesser underestimation for the model with race. The model with race showed better performance in reclassification compared to the model without race (net reclassification index, 0.13). The updated model with the Korean coefficient showed good agreement between predicted risk and observed outcome. Conclusion: In Korean IgAN patients, International IgA Nephropathy Prediction Tool had good discrimination power but underestimated the risk of progression. The updated model with the Korean coefficient showed acceptable calibration and warrants external validation.