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.

Odontogenic keratocyst (OKC) accounts for 3-11% of all odontogenic cysts. OKC is a benign intra-osseous odontogenic tumor, but what makes this cyst special is its aggressive behavior and high recurrence rate. OKC is relatively aggressive compared to other odontogenic cysts, but its malignant transformation is considered extremely rare. Squamous cell carcinoma associated with odontogenic keratocysts have rarely been reported in the medical literature. We recently experienced a case of a 63-year-old man finally confirmed with squamous cell carcinoma of the mandible, which was initially diagnosed as a benign odontogenic keratocyst. Surgical resection was performed as definitive treatment. Therefore, we present this unique case with a review of the literature.

OBJECTIVE: Sleep disturbance is common in the elderly, which is result from multi-factorial causes encompassing socio-demographic, behavioral, and clinical factors. We aimed to identify factors associated with insomnia among the elderly in a rural community in South Korea, a country with a rapidly growing aged population. METHODS: This cross-sectional study used the data from the second wave of the Korean Social life, Health and Ageing Project, which is a cohort study of individuals living in a typical rural community in South Korea. Socio-demographic, behavioral, and clinical characteristics were obtained through face-to-face interviews. Various factors suspected to be associated with insomnia were compared between elderly participants with and without insomnia, and multiple logistic regression analyses were conducted to identify independent risk factors for insomnia. RESULTS: We found that 32.4% of 509 participants (72.8±7.7 years old) had insomnia. Female sex [odds ratio (OR)=2.19], low education level (OR=2.44), current smoking (OR=2.26), number of chronic diseases (OR=2.21 for 2–3 chronic diseases; OR=2.06 for 4 or more chronic diseases), and depression (OR=2.53) were independently associated with insomnia. CONCLUSION: We found that sex, education, chronic disease, and depression independently increase the risk of insomnia of the elderly in a Korean rural community. To overcome the elderly's insomnia, interventions should target modifiable factors such as depression. To promote active aging, longitudinal studies of factors associated with insomnia among the elderly should be performed in different regions and communities.
Aged* ; Aging ; Chronic Disease ; Cohort Studies ; Cross-Sectional Studies ; Depression ; Education ; Female ; Humans ; Korea ; Logistic Models ; Longitudinal Studies ; Republic of Korea ; Risk Factors ; Rural Population* ; Sleep Initiation and Maintenance Disorders* ; Smoke ; Smoking

A beneficial radioprotective agent has been used to treat the radiation-induced lung injury. This study was performed to investigate whether curcumin, which is known to have anti-inflammatory and antioxidant properties, could ameliorate radiation-induced pulmonary inflammation and fibrosis in irradiated lungs. Rats were given daily doses of intragastric curcumin (200 mg/kg) prior to a single irradiation and for 8 weeks after radiation. Histopathologic findings demonstrated that macrophage accumulation, interstitial edema, alveolar septal thickness, perivascular fibrosis, and collapse in radiation-treated lungs were inhibited by curcumin administration. Radiation-induced transforming growth factor-beta1 (TGF-beta1), connective tissue growth factor (CTGF) expression, and collagen accumulation were also inhibited by curcumin. Moreover, western blot analysis revealed that curcumin lowered radiation-induced increases of tumor necrosis factor-alpha (TNF-alpha), TNF receptor 1 (TNFR1), and cyclooxygenase-2 (COX-2). Curcumin also inhibited the nuclear translocation of nuclear factor-kappa B (NF-kappaB) p65 in radiation-treated lungs. These results indicate that long-term curcumin administration may reduce lung inflammation and fibrosis caused by radiation treatment.
Animals ; Blotting, Western ; Collagen ; Connective Tissue Growth Factor ; Curcumin ; Cyclooxygenase 2 ; Edema ; Fibrosis ; Inflammation ; Lung ; Lung Injury ; Macrophages ; Pneumonia ; Rats ; Receptors, Tumor Necrosis Factor ; Tumor Necrosis Factor-alpha

Metastases to the heart are rarely diagnosed before the patient dies. A 26-year-old man was admitted with multiple metastasis of a testicular embryonal carcinoma and he was found to have intracardiac metastasis. Echocardiography showed that he had a mass rising from the interventricular septum and it was floating through the right ventricular outflow tract. The histology of the mass we removed from the right ventricle was consistent with testicular embryonal carcinoma. The patient made a smooth recovery after surgical intervention and chemotherapy. We believe this is the first reported case of testicular embryonal carcinoma that metastasized to the heart and that was successfully removed via surgery in Korea.
Adult ; Carcinoma, Embryonal ; Echocardiography ; Heart ; Heart Neoplasms ; Heart Ventricles ; Humans ; Korea ; Neoplasm Metastasis ; Testicular Neoplasms

PURPOSE: In spite of medical advances, empyema is a serious complication of pneumonia in children. Vaccination practices and antibiotic prescribing practices promote the change of clinical manifestations of empyema and causative organisms. So we made a nationwide clinical observation of 122 cases of empyema in children from 32 hospitals during the 5 year period from September 1999 to August 2004. METHODS: Demographic data, and clinical information on the course and management of empyema patients were collected retrospectively from medical records in secondary and tertiary hospitals in Korea. RESULTS: One hundred twenty two patients were enrolled from 35 hospitals. The most frequent age group was 1-3 years, accounting for 48 percent of all cases. The male to female sex ratio was 1.2:1. The main symptoms were cough, fever, respiratory difficulty, lethargy and chest pain in order of frequency. Hematologic findings on admission revealed decreased hemoglobin levels (10.4+/-1.6 g/dL) and increased leukocyte counts (16,234.3+/-10,601.8/microliter). Pleural fluid obtained from patients showed high leukocyte counts (30,365.8+/-64,073.0/microliter), high protein levels (522.3+/-1582.3 g/dL), and low glucose levels (88.1+/-523.5 mg/dL). Findings from pleural fluid cultures were positive in 80 cases(65.6 percent). The most common causative agent was Streptococcus pneumoniae. The majority of patients were treated with antibiotics and closed drainage. Some patients needed open drainage (16.4 percent) or decortication (3.3 percent). The mean duration of hospitalization was 28.6+/-15.3 days. CONCLUSION: We analyzed childhood empyema patients during a period of 5 years in Korean children. The most frequent age group was 1-3 years and the most common causative agent was Streptococcus pneumoniaeiae. The majority of patients were treated with antibiotics and close drainage.
Anti-Bacterial Agents ; Chest Pain ; Child ; Cough ; Drainage ; Empyema* ; Female ; Fever ; Glucose ; Hospitalization ; Humans ; Korea* ; Lethargy ; Leukocyte Count ; Male ; Medical Records ; Pneumonia ; Retrospective Studies ; Sex Ratio ; Streptococcus ; Streptococcus pneumoniae ; Tertiary Care Centers ; Vaccination