Prediction of early hematoma expansion of spontaneous intracerebral hemorrhage based on deep learning radiomics features of noncontrast computed tomography.

OBJECTIVES: Aimed to develop a nomogram model based on deep learning features and radiomics features for the prediction of early hematoma expansion. METHODS: A total of 561 cases of spontaneous intracerebral hemorrhage (sICH) with baseline Noncontrast Computed Tomography (NCCT) were included. The metrics of hematoma detection were evaluated by Intersection over Union (IoU), Dice coefficient (Dice), and accuracy (ACC). The semantic features of sICH were judged by EfficientNet-B0 classification model. Radiomics analysis was performed based on the region of interest which was automatically segmented by deep learning. A combined model was constructed in order to predict the early expansion of hematoma using multivariate binary logistic regression, and a nomogram and calibration curve were drawn to verify its predictive efficacy by ROC analysis. RESULTS: The accuracy of hematoma detection by segmentation model was 98.2% for IoU greater than 0.6 and 76.5% for IoU greater than 0.8 in the training cohort. In the validation cohort, the accuracy was 86.6% for IoU greater than 0.6 and 70.0% for IoU greater than 0.8. The AUCs of the deep learning model to judge semantic features were 0.95 to 0.99 in the training cohort, while in the validation cohort, the values were 0.71 to 0.83. The deep learning radiomics model showed a better performance with higher AUC in training cohort (0.87), internal validation cohort (0.83), and external validation cohort (0.82) than either semantic features or Radscore. CONCLUSION: The combined model based on deep learning features and radiomics features has certain efficiency for judging the risk grade of hematoma. CLINICAL RELEVANCE STATEMENT: Our study revealed that the deep learning model can significantly improve the work efficiency of segmentation and semantic feature classification of spontaneous intracerebral hemorrhage. The combined model has a good prediction efficiency for early hematoma expansion. KEY POINTS: • We employ a deep learning algorithm to perform segmentation and semantic feature classification of spontaneous intracerebral hemorrhage and construct a prediction model for early hematoma expansion. • The deep learning radiomics model shows a favorable performance for the prediction of early hematoma expansion. • The combined model holds the potential to be used as a tool in judging the risk grade of hematoma.

MR Lymphangiography for Focal Disruption of the Thoracic Duct in Chylothorax of an Infant: a Case Report and Literature Review.

Chylothorax is a rare cause of pleural effusion in children, and it is usually difficult to identify the location of chyle leakage due to the small size of the thoracic duct in children. Herein we report an infant case with chylothorax whose leakage of the thoracic duct was successfully located by magnetic resonance lymphangiography (MRL) using pre-contrast MR cholangiopancreatography (MRCP) and gadodiamide-enhanced spectral presaturation inversion recovery (SPIR) T1-weighted imaging, which demonstrate the imaging method is easy and effective for detecting the focal disruption of the thoracic duct in children with chylothorax and younger than 8 months old.

Clinical value of spectral CT in diagnosis of negative gallstones and common bile duct stones.

OBJECTIVE: To investigate the clinical value of spectral CT in diagnosis of negative gallstones and common bile duct stones primarily. METHODS: All patients diagnosed with negative biliary stones were analyzed and examined by spectral CT scanner retrospectively. Based on acquired raw imaging data, image series were reconstructed as described below: the optimal contrast-to-noise ratio monochromatic energy images, calcium- and fat- based material decomposition images and spectral curve images. All these imaging series were analyzed quantitatively and qualitatively. RESULTS: The contrast between negative stones and adjacent bile was 6.87 ± 5.48 HU on hybrid energy CT images and 47.30 ± 24.05 HU on optimal monochromatic energy CT images. The mean concentration of calcium in bile and negative stones was 19.36 ± 5.12 and 3.88 ± 6.60 mg/mL, and the fat in bile and negative stones was 998.48 ± 11.79 and 1035.68 ± 15.36 mg/mL. Effective atomic number Z of negative stones (6.60 ± 0.45) was lower than that of bile (7.65 ± 0.13). The slopes of the spectral curves for negative stones were k 90-40KeV = 1.43 ± 0.63 and k 140-90KeV = 0.19 ± 0.08, and for bile, they were k 90-40KeV = -0.27 ± 0.09 and k 140-90KeV = -0.04 ± 0.01. The same stone showed different densities in different imaging groups. The positive rate of conventional CT images was lower than that of other imaging groups. CONCLUSION: Spectral CT has a high diagnostic value for negative gallstones or bile duct stones, and material decomposition CT images and spectral curves can make an accurate diagnosis.

Value of High-Resolution MRI in the Diagnosis of Brachial Plexus Injury in Infants and Young Children.

Purpose: To investigate the value of high-resolution MRI based on 3D-short inversion time inversion recovery sampling perfection with application-optimized contrasts (3D-STIR SPACE) sequence for the diagnosis of brachial plexus injury in infants and young children. Methods: Physical examination, electromyography (EMG) and MRI data of 26 children with brachial plexus injury were retrospectively analyzed. Sensitivity, specificity, and accuracy were calculated for the three tests. The agreement among these examinations was analyzed with the Kappa test. P<0.05 was considered statistically significant. Results: Of the 26 children, 3 cases had normal MRIs, 23 cases had unilateral brachial plexus injury diagnosed with MRI, and a total of 73 nerve roots and/or sheaths were involved. Among the 23 cases with aberrant MRI findings, there were 19 cases of nerve root thickening (42 nerve roots), 4 cases of nerve root sleeve expansion (5 nerve roots), 17 cases of pseudomeningeal cysts (34 nerve roots), 2 cases of nerve root loosening (2 nerve roots), 8 cases of nerve root dissection (11 nerve roots), 19 cases with increased nerve signal (43 nerve roots), and 9 cases with an increased signal of the muscles on the affected side. As for the diagnosis of brachial plexus injury, the sensitivity and the accuracy of physical examination, EMG and MRI were 0.92, 0.86, and 0.88, respectively. The agreement between MRI and physical examination was substantial (κ=0.780, P=0.000), as did the agreement between MRI and EMG (κ=0.611, P=0.005). Conclusion: High-resolution MRI based on 3D-STIR SPACE sequence plays a role in the diagnosis and evaluation of brachial plexus injury in infants and young children. It can accurately identify the injured nerve and characterize related pathological alterations. Besides EMG and physical examination, it can be used as a valuable tool for screening and monitoring of brachial plexus injury in infants and children.

Modified Look-Locker Inverse-Recovery (MOLLI) Sequence of Quantitative Imaging in Dirty Magnetic Resonance Longitudinal Relaxation Time Diagnostic Value of GE Combined with Longitudinal Relaxation Time Quantitative Imaging for Myocardial Amyloidosis.

The pathological changes of myocarditis include degeneration and necrosis of myocardial cells and infiltration of inflammatory cells in the myocardial interstitium, accompanied by obvious myocardial fibrosis. Myocardial fibrosis is a determinant of ventricular remodeling and an important indicator of the classification of clinical risk factors and has an important value in evaluating the prognosis of heart disease. Cardiac magnetic resonance (CMR) is the "gold standard" for evaluating the shape and function of the heart, and it can show the characteristic pathological changes of myocardial tissue. The traditional gadolinium imaging agent delays the enhanced sequence images to visually show the extent of the affected myocardial fibrosis, but it cannot effectively identify small focal fibrosis or widespread diffuse fibrosis. The CMR longitudinal relaxation time quantitative technique can directly measure the relaxation time (T1) determined by the myocardial tissue and does not depend on the signal strength of the reference tissue and can quantitatively analyze the affected myocardium. In this study, the initial and enhanced quantitative imaging techniques of CMR were used to measure the magnetic value of the myocardium in patients with myocarditis, to explore the diagnostic value of myocardial fibrosis, and to analyze the correlation between cardiac fibrosis and cardiac function.

Risk Factors for Different Grades of Lower Respiratory Tract Infections in Children Under Five Years Old with Measles: Based on Chest Radiography.

BACKGROUND: Although measles epidemic has been controlled effectively after measles vaccination being carried out, China is still the focus of measles epidemic. The lower respiratory tract infections (LRTIs) is the most common complication, and also the most common reason for the death of patients with measles. PURPOSE: To analyze what clinical factors could affect the grades of LRTIs in children under five years old with measles. METHODS: Patients' 13 clinical factors and chest radiography were analyzed retrospectively. Based on the chest radiography, the LRTIs were divided into 3 grades. The clinical factors and the grades of LRTIs were analyzed by ordered logistic regression analysis. RESULTS: There were 74 cases in grade 0, 77 cases in grade 1, and 36 cases in grade 2. After univariate logistic regression analysis, there were 5 clinical factors whose P value were no more than 0.1, including the children's course of disease before admission, co-infection other pathogens, vaccination, the first symptoms and gender. After ordinal logistic regression analysis, the difference in the course of disease before admission (OR = 1.134, 95% CI: 1.025 to 1.141), without vaccination (OR = 1.808, 95% CI: 1.065 to 3.065) and co-infected other pathogens other pathogens infections (OR = 1.618, 95% CI: 1.073 to 2.440) during different grades was statistically significant (P < 0.05). CONCLUSION: In our study, there were three clinical factors, including "with a long course of disease before admission", "without measles vaccination" and "with other pathogens infections", as the independent risk factors of the grades of LRTIs in children with measles.