A fully automated rib fracture detection system on chest CT images and its impact on radiologist performance.

OBJECTIVE: To compare rib fracture detection and classification by radiologists using CT images with and without a deep learning model. MATERIALS AND METHODS: A total of 8529 chest CT images were collected from multiple hospitals for training the deep learning model. The test dataset included 300 chest CT images acquired using a single CT scanner. The rib fractures were marked in the bone window on each CT slice by experienced radiologists, and the ground truth included 861 rib fractures. We proposed a heterogeneous neural network for rib fracture detection and classification consisting of a cascaded feature pyramid network and a classification network. The deep learning-based model was evaluated based on the external testing data. The precision rate, recall rate, F1-score, and diagnostic time of two junior radiologists with and without the deep learning model were computed, and the Chi-square, one-way analysis of variance, and least significant difference tests were used to analyze the results. RESULTS: The use of the deep learning model increased detection recall and classification accuracy (0.922 and 0.863) compared with the radiologists alone (0.812 vs. 0.850). The radiologists achieved a higher precision rate, recall rate, and F1-score for fracture detection when using the deep learning model, at 0.943, 0.978, and 0.960, respectively. When using the deep learning model, the radiologist's reading time was decreased from 158.3 ± 35.7 s to 42.3 ± 6.8 s. CONCLUSION: Radiologists achieved the highest performance in diagnosing and classifying rib fractures on CT images when assisted by the deep learning model.

[Cluster analysis of Dendrobium officinale from different populations through agronomy characterization].

Through resources investigation and sample collection,a total number of 392 Dendrobium officinale from 38 different populations,9 provinces were processed for measuring and observing.Fourteen agronomy characterizations like stem height,stem diameter,number of node in stem were selected for further classification.The cluster analysis was performed using Ward and Euclidean method.The results showed that the threshold of genetic distance was 2.77.Thirtyeight populations were divided into 6 groups.The factor analysis showed that,the leaf shape,stem shape,pitch length and leaf color were very important factors for classification.This study establish the foundation for analyzing the genetic relationship of D.officinale from different populations.

[Study on identification of Dendrobium officinale and related species by bidirectional PCR amplification of mismatched and specific alleles].

Based on rDNA ITS sequences of Dendrobium officinale and the other 69 species of Dendrobium, a pair of dismatched allele-specific diagnostic primers, TPSH-AS1F and TPSH-AS1R were designed to authenticate D. officinale from the other species. Thebidirectional PCR amplification were performed using the diagnostic primers with the total DNAs of the original plants or processing products as a template. When the annealing temperature was raised to 60 ℃, only the template DNA of D. officinale could be amplified whereas the diagnostic PCRs of the other Dendrobium species were all negative. Compared with the other authentification methods, the bidirectional PCR amplifications is not only simpler and time-saving but practical and effective.

The clinical value of using magnetic resonance contrast-enhanced three-dimensional nerve view in the diagnosis and management of sacral fracture accompanied by sacral plexus injury.

BACKGROUND: The treatment of sacral fractures accompanied by nerve injury is complex and often leads to an unsatisfactory prognosis and poor quality of life in patients. OBJECTIVE: The present study aimed to investigate the clinical value of using 3.0T magnetic resonance contrast-enhanced three-dimensional (MR CE-3D) nerve view magnetic resonance neurography (MRN) in the diagnosis and management of a sacral fracture accompanied by a sacral plexus injury. METHODS: Thirty-two patients with a sacral fracture accompanied by a sacral plexus injury, including 24 cases of Denis spinal trauma type II and 8 cases of type III, were enrolled in the study. All patients had symptoms or signs of lumbosacral nerve injury, and an MRN examination was performed to clarify the location and severity of the sacral nerve injury. Segmental localization of the sacral plexus was done to indicate the site of the injury as being intra-spinal (IS), intra-foraminal (IF), or extra-foraminal (EF), and the severity of the nerve injury was determined as being mild, moderate, or severe. Surgical nerve exploration was then conducted in six patients with severe nerve injury. The location and severity of the nerve injury were recorded using intra-operative direct vision, and the results were statistically compared with the MRN examination results. RESULTS: MRN showed that 81 segments had mild sacral plexus injuries (8 segments of IS, 20 segments of IF, 53 segments of EF), 78 segments had moderate sacral plexus injuries (8 segments of IS, 37 segments of IF, and 33 segments of EF), and 19 segments had severe sacral plexus injuries (7 segments of IS, 9 segments of IF, and 3 segments of EF). The six patients who underwent surgery had the following intra-operative direct vision results: 3 segments of moderate injury (IF) and 20 segments of severe injury (7 segments of IS, 10 segments of IF, 3 segments of EF). There was no statistically significant difference in the results between the intra-operative direct vision and those of the MRN examination (p> 0.05). CONCLUSION: MR CE-3D nerve view can clearly and accurately demonstrate the location and severity of sacral nerve injury accompanied by a sacral fracture, and has the potential for being the first choice of examination method for this kind of injury, which would be of important clinical value.

CT-based Morphometric Analysis of Approach of Percutaneous Transforaminal Endoscopic Lumbar Interbody Fusion.

OBJECTIVES: A radiographic study was designed to measure the relationship of the exiting nerve root and its surroundings to the corresponding intervertebral disc for percutaneous transforaminal endoscopic lumbar interbody fusion to better understand the regional anatomy and to improve clinical applications. METHODS: A retrospective study from January 2017 to October 2017 was conducted at Tianjin Hospital. CT images were obtained from patients presenting low back pain (110 patients), and analysis was performed bilaterally from L2-3 to L5 S1 . In the rotating coronal plane we analyzed: the nerve root-dural sac distance at the superior and inferior margins of the disc (Js, Ji); the nerve root-pedicle distance at the medial, middle, and lateral borders of the pedicle (Pa, Pb, Pc); the pedicle width (W); and the safe working zone, defined as a trapezoid bounded by the inferior pedicle and the exiting nerve root (S). In the transverse plane, the nerve root-articular process and the shortest distance for the nerve root-articular process joint surface were analyzed at the superior and inferior margins of the disc (Gs, Gi), respectively. The groups were analyzed using ANOVA, and paired t-tests were used to compare the left and right sides. RESULTS: From L2-3 to L5 S1 , the distance of the nerve root to the dural sac was larger at the inferior margin of the disc. From L2-3 to L5 S1 , each segment of the vertebral nerve root-pedicle distance gradually decreased from medial to lateral. From L2-3 to L5 S1 , the distance from the exiting nerve root to the middle and lateral margins of the pedicle gradually decreased, with L5 S1 being the minimum. Some significant differences were observed between the left and right sides for L4-5 and L5 S1 . The pedicle width of the vertebral body and the mean area for the safe working zone gradually increased from L2-3 to L5 S1 . In the axial plane, the shortest distance between the nerve root and articular process joint surface at the inferior margin of the disc was greater than the distance for the nerve root to the articular process at the superior margin of the disc from L2-3 to L5 S1 . There were no significant differences between the two sides. CONCLUSIONS: It is more difficult to implant a cage with a width of 10 mm above the L3-4 level. By removing part of the superior articular process, the safe working area can be expanded, and damage to the nerve or other structures can be avoided when implanting a cage.