Synthetic data, generated using advanced artificial intelligence (AI) techniques, replicates the statistical properties of real-world datasets while excluding identifiable information.Although synthetic data does not consist of actual data points, it is derived from original datasets, thereby enabling analyses that yield results comparable to those obtained with real data. Synthetic datasets are evaluated based on their utility—a measure of how effectively they mirror real data for analytical purposes. This paper presents the generation of synthetic datasets through the Healthcare Big Data Showcase Project (2019–2023). The original dataset comprises comprehensive multi-omics data from 400 individuals, including cancer survivors, chronic disease patients, and healthy participants. Synthetic data facilitates efficient access and robust analyses, serving as a practical tool for research and education. It addresses privacy concerns, supports AI research, and provides a foundation for innovative applications across diverse fields, such as public health and precision medicine.

Synthetic data, generated using advanced artificial intelligence (AI) techniques, replicates the statistical properties of real-world datasets while excluding identifiable information.Although synthetic data does not consist of actual data points, it is derived from original datasets, thereby enabling analyses that yield results comparable to those obtained with real data. Synthetic datasets are evaluated based on their utility—a measure of how effectively they mirror real data for analytical purposes. This paper presents the generation of synthetic datasets through the Healthcare Big Data Showcase Project (2019–2023). The original dataset comprises comprehensive multi-omics data from 400 individuals, including cancer survivors, chronic disease patients, and healthy participants. Synthetic data facilitates efficient access and robust analyses, serving as a practical tool for research and education. It addresses privacy concerns, supports AI research, and provides a foundation for innovative applications across diverse fields, such as public health and precision medicine.

Synthetic data, generated using advanced artificial intelligence (AI) techniques, replicates the statistical properties of real-world datasets while excluding identifiable information.Although synthetic data does not consist of actual data points, it is derived from original datasets, thereby enabling analyses that yield results comparable to those obtained with real data. Synthetic datasets are evaluated based on their utility—a measure of how effectively they mirror real data for analytical purposes. This paper presents the generation of synthetic datasets through the Healthcare Big Data Showcase Project (2019–2023). The original dataset comprises comprehensive multi-omics data from 400 individuals, including cancer survivors, chronic disease patients, and healthy participants. Synthetic data facilitates efficient access and robust analyses, serving as a practical tool for research and education. It addresses privacy concerns, supports AI research, and provides a foundation for innovative applications across diverse fields, such as public health and precision medicine.

Synthetic data, generated using advanced artificial intelligence (AI) techniques, replicates the statistical properties of real-world datasets while excluding identifiable information.Although synthetic data does not consist of actual data points, it is derived from original datasets, thereby enabling analyses that yield results comparable to those obtained with real data. Synthetic datasets are evaluated based on their utility—a measure of how effectively they mirror real data for analytical purposes. This paper presents the generation of synthetic datasets through the Healthcare Big Data Showcase Project (2019–2023). The original dataset comprises comprehensive multi-omics data from 400 individuals, including cancer survivors, chronic disease patients, and healthy participants. Synthetic data facilitates efficient access and robust analyses, serving as a practical tool for research and education. It addresses privacy concerns, supports AI research, and provides a foundation for innovative applications across diverse fields, such as public health and precision medicine.

Synthetic data, generated using advanced artificial intelligence (AI) techniques, replicates the statistical properties of real-world datasets while excluding identifiable information.Although synthetic data does not consist of actual data points, it is derived from original datasets, thereby enabling analyses that yield results comparable to those obtained with real data. Synthetic datasets are evaluated based on their utility—a measure of how effectively they mirror real data for analytical purposes. This paper presents the generation of synthetic datasets through the Healthcare Big Data Showcase Project (2019–2023). The original dataset comprises comprehensive multi-omics data from 400 individuals, including cancer survivors, chronic disease patients, and healthy participants. Synthetic data facilitates efficient access and robust analyses, serving as a practical tool for research and education. It addresses privacy concerns, supports AI research, and provides a foundation for innovative applications across diverse fields, such as public health and precision medicine.

Objective: To develop and evaluate a deep learning model for automated segmentation and detection of bone metastasis on spinal MRI. Materials and Methods: We included whole spine MRI scans of adult patients with bone metastasis: 662 MRI series from 302 patients (63.5 ± 11.5 years; male:female, 151:151) from three study centers obtained between January 2015 and August 2021 for training and internal testing (random split into 536 and 126 series, respectively) and 49 MRI series from 20 patients (65.9 ± 11.5 years; male:female, 11:9) from another center obtained between January 2018 and August 2020 for external testing. Three sagittal MRI sequences, including non-contrast T1-weighted image (T1), contrast-enhanced T1-weighted Dixon fat-only image (FO), and contrast-enhanced fat-suppressed T1-weighted image (CE), were used. Seven models trained using the 2D and 3D U-Nets were developed with different combinations (T1, FO, CE, T1 + FO, T1 + CE, FO + CE, and T1 + FO + CE). The segmentation performance was evaluated using Dice coefficient, pixel-wise recall, and pixel-wise precision. The detection performance was analyzed using per-lesion sensitivity and a free-response receiver operating characteristic curve. The performance of the model was compared with that of five radiologists using the external test set. Results: The 2D U-Net T1 + CE model exhibited superior segmentation performance in the external test compared to the other models, with a Dice coefficient of 0.699 and pixel-wise recall of 0.653. The T1 + CE model achieved per-lesion sensitivities of 0.828 (497/600) and 0.857 (150/175) for metastases in the internal and external tests, respectively. The radiologists demonstrated a mean per-lesion sensitivity of 0.746 and a mean per-lesion positive predictive value of 0.701 in the external test. Conclusion The deep learning models proposed for automated segmentation and detection of bone metastases on spinal MRI demonstrated high diagnostic performance.

Musculoskeletal ultrasonography (US) has unique advantages, such as excellent spatial resolution for superficial structures, the capability for dynamic imaging, and the ability for direct correlation and provocation of symptoms. For these reasons, US is increasingly used to evaluate problems in small joints, such as the foot and ankle. However, it is almost impossible to evaluate every anatomic structure within a limited time. Therefore, US examinations can be faster and more efficient if radiologists know where to look and image patients with typical symptoms. In this review, common etiologies of heel pain are discussed in a problem-based manner. Knowing the common pain sources and being familiar with their US findings will help radiologists to perform accurate and effective US examinations.

PURPOSE: The purpose of this study was to present a technique for measuring physiologic distal tibiofibular syndesmosis widening using 3-dimensional ultrasonography (3D-US) with an evaluation of its reliability, and to determine whether there were differences in the measurements between different dynamic stress tests. METHODS: We retrospectively evaluated 3D-US of 20 subjects with normal ankle syndesmosis. 3D-US was performed in neutral (N), dorsiflexion with external rotation (DFER), and weight-bearing standing (WB) positions at the anterior inferior tibiofibular ligament level in both ankles for comparison. Using 3D-US volume data, axial images were reconstructed at the level of the lateral prominence of the anterior tibial tubercle to ensure consistent measurements of the tibiofibular clear space (TFCS) by two radiologists. RESULTS: There was a wide range of TFCS values among the subjects (N, 1.2 to 4.2 mm; DFER, 2.3 to 4.8 mm; WB, 1.7 to 4.6 mm). When both ankles of each subject were evaluated, the side-to-side differences were less than 1 mm in all positions, with high intraclass correlation coefficient (ICC) values between both ankles (ICC, 0.85 to 0.93). The inter-rater agreement for all TFCS measurements between the two radiologists was excellent (ICC, 0.81 to 0.96). In comparisons between the two dynamic stress tests, the TFCS was significantly wider in the DFER position than in the WB position (DFER vs. WB, 3.3 mm vs. 2.9 mm; P<0.001). CONCLUSION: Using 3D-US, we were able to consistently evaluate the TFCS with good reliability. In a comparison of the two dynamic tests, there was more significant widening of the TFCS in the DFER position than in the WB position.
Ankle Joint ; Ankle ; Diagnosis ; Exercise Test ; Lateral Ligament, Ankle ; Retrospective Studies ; Ultrasonography ; Weight-Bearing

BACKGROUND AND PURPOSE: Decreasing the time delay for thrombolysis, including intravenous thrombolysis (IVT) with tissue plasminogen activator and intra-arterial thrombectomy (IAT), is critical for decreasing the morbidity and mortality of patients experiencing acute stroke. We aimed to decrease the in-hospital delay for both IVT and IAT through a multidisciplinary approach that is feasible 24 h/day. METHODS: We implemented the Stroke Alert Team (SAT) on May 2, 2016, which introduced hospital-initiated ambulance prenotification and reorganized in-hospital processes. We compared the patient characteristics, time for each step of the evaluation and thrombolysis, thrombolysis rate, and post-thrombolysis intracranial hemorrhage from January 2014 to August 2016. RESULTS: A total of 245 patients received thrombolysis (198 before SAT; 47 after SAT). The median door-to-CT, door-to-MRI, and door-to-laboratory times decreased to 13 min, 37.5 min, and 8 min, respectively, after SAT implementation (P<0.001). The median door-to-IVT time decreased from 46 min (interquartile range [IQR] 36–57 min) to 20.5 min (IQR 15.8–32.5 min; P<0.001). The median door-to-IAT time decreased from 156 min (IQR 124.5–212.5 min) to 86.5 min (IQR 67.5–102.3 min; P<0.001). The thrombolysis rate increased from 9.8% (198/2,012) to 15.8% (47/297; P=0.002), and the post-thrombolysis radiological intracranial hemorrhage rate decreased from 12.6% (25/198) to 2.1% (1/47; P=0.035). CONCLUSIONS: SAT significantly decreased the in-hospital delay for thrombolysis, increased thrombolysis rate, and decreased post-thrombolysis intracranial hemorrhage. Time benefits of SAT were observed for both IVT and IAT and during office hours and after-hours.
Ambulances ; Cerebral Infarction ; Humans ; Intracranial Hemorrhages ; Mortality ; Stroke* ; Thrombectomy ; Thrombolytic Therapy ; Tissue Plasminogen Activator

BACKGROUND: Aldosterone antagonists are reported to have beneficial effects on diabetic nephropathy by effective blocking of the renin-angiotensin-aldosterone system. We investigated the renoprotective effect of the selective aldosterone receptor blocker eplerenone, the angiotensin converting enzyme inhibitor lisinopril, and combined eplerenone and lisinopril treatment in type 2 diabetic rats. METHODS: Animals were divided into six groups as follows: Otsuka Long-Evans Tokushima Fatty (OLETF) rat control, OLETF rats treated with a low dose of eplerenone (50 mg/kg/day), OLETF rats treated with a high dose of eplerenone (200 mg/kg/day), OLETF rats treated with lisinopril (10 mg/kg/day), OLETF rats treated with a combination of both drugs (eplerenone 200 mg/kg/day and lisinopril 10 mg/kg/day), and obese non-diabetic Long-Evans Tokushima Otsuka rats for 26 weeks. RESULTS: Urinary albumin excretion was significantly lower in the lisinopril group, but not in the eplerenone group. Urinary albumin excretion was decreased in the combination group than in the lisinopril group. Glomerulosclerosis and renal expression of type I and type IV collagen, plasminogen activator inhibitor-1, transforming growth factor-beta1, connective tissue growth factor, and fibronectin mRNA were markedly decreased in the lisinopril, eplerenone, and combination groups. CONCLUSION: Eplerenone and lisinopril combination showed additional benefits on type 2 diabetic nephropathy compared to monotherapy of each drug.
Aldosterone ; Animals ; Collagen Type IV ; Connective Tissue Growth Factor ; Diabetic Nephropathies ; Fibronectins ; Lisinopril ; Mineralocorticoid Receptor Antagonists ; Peptidyl-Dipeptidase A ; Plasminogen Activators ; Rats ; Rats, Inbred OLETF ; Receptors, Mineralocorticoid ; Renin-Angiotensin System ; RNA, Messenger ; Spironolactone