Objectives: Coronavirus disease 2019 (COVID-19) is sweeping the globe and has resulted in infections in millions of people. Patients with COVID-19 face a high fatality risk once symptoms worsen; therefore, early identification of severely ill patients can enable early intervention, prevent disease progression, and help reduce mortality. This study aims to develop an artificial intelligence-assisted tool using computed tomography (CT) imaging to predict disease severity and further estimate the risk of developing severe disease in patients suffering from COVID-19. Materials and Methods: Initial CT images of 408 confirmed COVID-19 patients were retrospectively collected between January 1, 2020 and March 18, 2020 from hospitals in Honghu and Nanchang. The data of 303 patients in the People's Hospital of Honghu were assigned as the training data, and those of 105 patients in The First Affiliated Hospital of Nanchang University were assigned as the test dataset. A deep learning based-model using multiple instance learning and residual convolutional neural network (ResNet34) was developed and validated. The discrimination ability and prediction accuracy of the model were evaluated using the receiver operating characteristic curve and confusion matrix, respectively. Results: The deep learning-based model had an area under the curve (AUC) of 0.987 (95% confidence interval [CI]: 0.968-1.00) and an accuracy of 97.4% in the training set, whereas it had an AUC of 0.892 (0.828-0.955) and an accuracy of 81.9% in the test set. In the subgroup analysis of patients who had non-severe COVID-19 on admission, the

Study Design. Retrospective study. Objective. To evaluate the outcomes and safety of endoscopic laminectomy for central lumbar canal spinal stenosis. Summary of Background Data. Spinal endoscopy is mostly used in the treatment of lumbar disc herniation, while endoscopic laminectomy for lumbar spinal stenosis is rarely reported. Methods. From January 2016 to June 2017, 38 patients with central lumbar canal spinal stenosis were treated with endoscopic laminectomy. Clinical symptoms were evaluated at 1, 3, 6, and 12 months and the last follow-up after surgery. Functional outcomes were assessed by using the Japanese Orthopedic Association Scores (JOA) and Oswestry Disability Index (ODI). The decompression effect was assessed by using the dural sac cross-sectional area (DSCA). Lumbar stability was evaluated using lumbar range of motion (ROM), ventral intervertebral space height (VH), and dorsal intervertebral space height (DH). Results. The mean age of the cases was 60.8 years, the mean operation time was 66.3 minutes, the blood loss was 38.8 mL, and the length of incision was 19.6 mm. The mean time in bed was 22.3 hours, and the mean hospital stay was 8.8 days. JOA scores were improved from 10.9 to 24.1 (P < 0.05), ODI scores were improved from 79.0 to 27.9 (P < 0.05), DSCA was improved from 55.7 to 109.5 mm 2 (P < 0.05), ROM scores were improved from 5.68 to 5.78 (P < 0.05), and DH scores were reduced from 6.6 to 6.5 mm (P < 0.05). There was no significant difference in VH before and after operation (P > 0.05). There were no serious complications during the follow-ups. Conclusion. Endoscopic laminectomy had the advantage of a wider view, which was effective, safe, and less invasive for lumbar spinal stenosis.

BACKGROUND: Lumbar disc herniation (LDH) can affect lower limb muscle function resulting in an abnormal gait. This study aims to use surface electromyography (SEMG) to evaluate patients with L4/L5 and L5/S1 LDH throughout muscle movement. METHODS: Twenty L4/L5 LDH patients (L5 Group), twenty L5/S1 LDH patients (S1 Group), and twenty healthy controls (Healthy) were recruited for the study. SEMG of bilateral tibialis anterior (TA) and lateral gastrocnemius (LG) muscles of patients were recorded using the DELSYS Wireless EMG System (Trigno TM Wireless Systems, Delsys Inc., USA). Root-mean-square (RMS), mean power frequency (MPF), and median frequency (MF) were compared between bilateral limbs in each participant. RESULTS: Reduced MPF and MF was found in TA measurements of the L5 Group and LG measurements of the S1 Group. The MPF and MF of the TA of symptomatic limbs of the L5 Group were reduced when compared to asymptomatic limbs (p = 0.006, p = 0.012, p < 0.05), and there were no significant differences in LG measurements (p > 0.05). The LG MPF and MF of the S1 Group in symptomatic limbs were reduced when compared to asymptomatic limbs (p = 0.006, p = 0.017, p < 0.05), and there were no significant differences in TA measurements (p > 0.05). Although there were no significant differences in RMS between bilateral limbs of the L5 and S1 Groups, we found some changes in RMS curves. First, compared to asymptomatic limbs of L4/L5LDH patients, β-peaks in the TA of symptomatic limbs appeared earlier. Second, two peaks in the LG of symptomatic limbs were found in L5/S1 LDH patients. CONCLUSION: TA is affected in patients with LDH of L4/L5, and LG is affected in patients with LDH of L5/S1. As demonstrated, SEMG can identify LDH-related muscle dysfunction.

Introduction: The pathological rare category of thyroid is a type of lesion with a low incidence rate and is easily misdiagnosed in clinical practice, which directly affects a patient’s treatment decision. However, it has not been adequately investigated to recognize the rare, benign, and malignant categories of thyroid using the deep learning method and recommend the rare to pathologists.Methods: We present an empirical decision tree based on the binary classification results of the patch-based UNet model to predict rare categories and recommend annotated lesion areas to be rereviewed by pathologists.Results: Applying this framework to 1,374 whole-slide images (WSIs) of frozen sections from thyroid lesions, we obtained an area under a curve of 0.946 and 0.986 for the test datasets with and without WSIs, respectively, of rare types. However, the recognition error rate for the rare categories was significantly higher than that for the benign and malignant categories (p &lt; 0.00001). For rare WSIs, the addition of the empirical decision tree obtained a recall rate and precision of 0.882 and 0.498, respectively; the rare types (only 33.4% of all WSIs) were further recommended to be rereviewed by pathologists. Additionally, we demonstrated that the performance of our framework was comparable to that of pathologists in clinical practice for the predicted benign and malignant sections.Conclusion: Our study provides a baseline for the recommendation of the uncertain predicted rare category to pathologists, offering potential feasibility for the improvement of pathologists’ work efficiency.