Deep learning is one of the most popular artificial intelligence techniques used in the medical field. Although it is at an early stage compared to deep learning analyses of computed tomography or magnetic resonance imaging, studies applying deep learning to ultrasound imaging have been actively conducted. This review analyzes recent studies that applied deep learning to ultrasound imaging of various abdominal organs and explains the challenges encountered in these applications.

Local ablation therapy is considered as a conventional treatment option for patients with early stage hepatocellular carcinoma (HCC). Although radiofrequency (RF) ablation is widely used for HCC, the use of cryoablation has been increasing as newer and safer cryoablation systems have developed. The thermodynamic mechanism of freezing and thawing used in cryoablation is the Joule-Thomson effect. Cryoablation destroys tissue via direct tissue destruction and vascular-related injury. A few recent comparative studies have shown that percutaneous cryoablation for HCCs is comparable to percutaneous RF ablation in terms of long term therapeutic outcomes and complications. Cryoablation has several advantages over RF ablation such as well visualization of iceball, no causation of severe pain, and lack of severe damage to great vessels and gallbladder. It is important to know the advantages and disadvantages of cryoablation compared with RF ablation for improvement of therapeutic efficacy and safety.
Carcinoma, Hepatocellular/diagnostic imaging/*surgery ; Catheter Ablation ; *Cryosurgery ; Humans ; Liver Neoplasms/diagnostic imaging/*surgery ; Magnetic Resonance Imaging ; Tomography, X-Ray Computed ; Ultrasonography

BACKGROUND: We analyzed the changes in QT dispersion (QTd) in children with Kawasaki disease (KD), and determined the presence of repolarization abnormality in these children even in the absence of coronary artery abnormalities. METHODS: Ninety-one children with KD and 20 healthy controls were enrolled in this retrospective study. Serial echocardiographic and electrocardiographic (ECG) measurements in the beginning of treatment, 2nd month and 6th month after the diagnosis were compared. Fifty-one of 91 children had at least 2 serial ECG data. The number of patients who had 3 consecutive ECG data was 23. RESULTS: Among the 67 KD patients with no coronary artery changes, the consecutive mean QTd values were 41.86 ms, 37.84 ms, and 25.47 ms, respectively (26 ms for controls). In the analysis of changes among KD patients without coronary artery abnormalities, QTd showed a significant decrease with time (p=0.01). Especially, the 1st month and the 6th month QTd values were significantly different (p=0.028). The mean QTd values in KD patients with coronary artery changes were significantly higher than those in KD patients with no coronary artery changes at each time (1st, 2nd, and 6th month exam). CONCLUSION: QTd is significantly increased in children during the early stage of KD. Repolarization abnormality may exist during the acute stage of KD, regardless of the echocardiographic changes.
Child* ; Coronary Vessels ; Diagnosis ; Echocardiography ; Electrocardiography ; Humans ; Mucocutaneous Lymph Node Syndrome* ; Retrospective Studies

Recently, considerable progress has been made in interpreting perceptual information through artificial intelligence, allowing better interpretation of highly complex data by machines. Furthermore, the applications of artificial intelligence, represented by deep learning technology, to the fields of medical and biomedical research are increasing exponentially. In this article, we will explain the stages of deep learning algorithm development in the field of medical imaging, namely topic selection, data collection, data exploration and refinement, algorithm development, algorithm evaluation, and clinical application; we will also discuss the latest trends for each stage.

Artificial Intelligence ; Diagnostic Imaging ; Learning

No abstract available.
Adenoma, Bile Duct/pathology/*radiography ; Adult ; Aged ; Angiomyolipoma/pathology/*radiography ; Bile Duct Neoplasms/pathology/*radiography ; Bile Ducts, Intrahepatic ; Carcinoma, Hepatocellular/radiography ; Diagnosis, Differential ; Female ; Gadolinium DTPA/*chemistry ; Humans ; Liver Diseases/pathology/*radiography ; Liver Neoplasms/pathology/*radiography ; *Magnetic Resonance Imaging ; Male ; Middle Aged ; Pseudolymphoma/pathology/*radiography ; Tomography, X-Ray Computed

No abstract available.
Adult ; Cerebroside-Sulfatase/*metabolism ; Child, Preschool ; *Chromatography, High Pressure Liquid ; Enzyme Assays/instrumentation/*methods ; Female ; Humans ; Kinetics ; Leukocytes/*enzymology ; Leukodystrophy, Metachromatic/diagnosis/enzymology ; Male ; Middle Aged ; Reference Standards ; Substrate Specificity ; Sulfoglycosphingolipids/analysis/metabolism/standards ; *Tandem Mass Spectrometry/standards

Generally, osteoporosis is caused by the age-related physiologic bone loss. In cases of severe osteoporosis, the reduced bone mass and increaed skeletal fragility make the risk of fracture high. Recently, the relationship between the mandible and the condition of the skeletal bone, the molphometric data about the quality and quantity of the mandible are very important in the diagnosis and treatment of osteoporosis. This study was undertaken to clarify the correlations of the mineral density between mandible and other bones, to predict the bone mineral density (BMD) of the mandible. Cervical (2nd~5th) and lumbar (1st~4th) vertebrae, the bones of forearm and proximal femurs, mandibles from 42-embalmed cadavers (22/M, 20/F, mean age: 66.95 years) were examined. The BMD was measured by dual energy X-ray absorptiometry. The data were analysed with SPSS 12.0 program according to age, gender, to verify the statistical significance. 1. The BMD in the mandibular areas were variable. The BMD in the angle of the mandible (mean 0.40 g/cm2) was similar with that of ultra-distal part of the forearm (mean 0.34 g/cm2). 2. The BMD in the mandible was highly correlated with that of around the forearm. The correlation coefficient between density in the angle, the part between 1st and 2nd molar, the part between 1st molar and 2nd premolar and incisor portion of the mandible and density in the 1/3-distal part of the forearm was high (r=0.742, 0.697, 0.618, 0.652). The correlation coefficient between density in the part between 1st and 2nd premolar and density in the total femur was high (r=0.542). 3. The BMD in the mandible was correlated with that of around the femur neck (r=0.607). 4. The correlation coefficient between the BMD in the 1st and 2nd molar of the mandible and the BMD in the 5th cervical vertebra was high (r=0.543). And also, the correlation coefficient between the BMD in the incisor area of the mandible and the BMD in the 3rd lumbar vertebra was high (r=0.561). 5. Bony status of the mandible was classified into normal, osteopenia, osteoporosis by BMD of WHO criteria. The data represented that BMD of the mandible was decreased as BMD of the skeletal bones decreased. It is concluded that it could be possible to predict BMD of the mandible by the measurement of BMD in the forearm.
Absorptiometry, Photon ; Bicuspid ; Bone Density ; Bone Diseases, Metabolic ; Cadaver ; Femur ; Femur Neck ; Forearm ; Incisor ; Mandible ; Molar ; Osteoporosis ; Spine

BACKGROUND: To prevent complications of diabetes mellitus, diabetic patients should test blood glucose level frequently. In these days, glucometers are widely used for self-monitoring and many kinds of products are introduced. We performed the present study to evaluate the performance of glucophone (GlucoPack(TM), Infopia Co. Ltd., Anyang, Korea) as a point-of-care testing glucometer. METHODS: Glucometers including glucophone and Finetest(TM) (Infopia Co. Ltd., Anyang, Korea) were evaluated for precision, linearity, and accuracy. The interpersonal variation by different operators, reagent stability, comparison capillary blood with venous blood, and user acceptability were also evaluated. RESULTS: Glucophone and Finetest glucometer showed excellent precisions wtih less than 5% of CVs of within-run and total precision. Linearity was also satisfactory from 24 to 517 mg/dL for glucophone and Finetest glucometer. Comparison with routine chemistry autoanalyzer, TBA-200FR showed close concordance over the entire range of evaluated concentrations (y = 0.8397x + 3.8351, x = TBA-200FR, y=glucophone, R2=0.9523). There were no significant changes in test results during exposure period at room temperature after opening the reagents. Generally random users expressed high satisfaction to glucophone with the exception of complicated operating method. CONCLUSION: Glucophone showed excellent precision, linearity, and correlation with the reference method. Because POCT glucometers are influenced by operator and multiple external factors, it is important that users recognize interfering factors and preservation conditions of test strips. It is hoped that glucophone is a good POCT glucose meter by establishment continuous quality control system and improvement of operation.
Blood Glucose ; Capillaries ; Chemistry ; Diabetes Complications ; Diabetes Mellitus ; Glucose ; Gyeonggi-do ; Hope ; Humans ; Indicators and Reagents ; Quality Control