Segmental Fusion Research Articles

The Data Fusion and Semantic Segmentation Method Based on Multi-Sensor System

The design and implementation of a prototype environmental perception system are introduced, which consists of the multi-camera, LiDAR, and IMU sensor. The processing steps of the environmental data collected by the multi-sensor platform include joint calibration, data fusion, and semantic segmentation. First, the optimization method based on the epipolar constraint is proposed for the joint calibration of the multi-camera and the LiDAR system. In the data fusion section, an improved data association method for the point cloud is proposed where the foreground segmentation method is used to reduce scale estimation error due to the scale-sudden-change. Finally, the semantic segmentation method based on the deep-learning is proposed. The convolutional neural network based on the Squeeze-and-Excitation Net is designed to identify and classify the point cloud data accurately. The physical simulation result shows that the system collects and identifies the environmental information accurately.

Changes in the centre of rotation and the anterior bone loss of the vertebral body in Mobi-C artificial disc replacement segments after cervical hybrid surgery: a retrospective study.

To examine the short-term efficacy and imaging results of using the Mobi-C in cervical hybrid surgery on 2-level cervical spondylolisthesis. To observe post-operative changes in the flexion-extension centre of rotation (FE-COR) and anterior bone loss (ABL) of the anterior cervical disc replacement (ACDR) segment. Forty-two patients (20 males and 22 females, aged 42‒67years) who underwent cervical hybrid surgery were retrospectively analysed. Their ACDR segment used Mobi-C, and the fusion segment used ROI-C, with a follow-up of 25‒42months (31.1 ± 4.8months). The modified Japanese Orthopaedic Association (mJOA) score, Neck Disability Index (NDI), and visual analogue scale (VAS) were used to assess clinical outcomes. Pre-operative, 6-month post-operative, and final follow-up radiographs were collected to compare total cervical spine curvature (C2-C7), curvature of the operated segments, range of motion (ROM) in the total cervical spine, operated segmental ROM, ACDR segmental ROM, and operated adjacent segmental ROM. The height of the superior articular process (HSAP), the orientation of zygapophyseal joint spaces (OZJS), and the length of the superior articular surface (LSAS) were measured. The FE-COR of the ACDR segment was measured using the mid-plumb line method. The translation distance of the Mobi-C was measured. The degree of disc degeneration in the adjacent segment, bony fusion of the ACDF segment, and ABL of the upper and lower vertebra of the ACDR segment were observed. In our group, all patients have shown improvements in their postoperative mJOA, NDI, and VAS scores. Overall cervical ROM and surgical segmental ROM decreased (P < 0.05). However, there was no significant decrease in ACDR segmental ROM and upper or lower adjacent segmental ROM compared with pre-operatively (P > 0.05). For FE-COR-X, only the last follow-up compared with pre-surgery showed statistical significance (46.74 ± 7.71% vs. 50.74 ± 6.92%, P < 0.05). For FE-COR-Y, the change was statistically significant at both 6months post-operation and the final follow-up compared to pre-operation (45.37% ± 21.11% vs. 33.82% ± 10.87%, 45. 37% ± 21.11% vs. 27.48% ± 13.58%, P < 0.05). No significant difference in the Mobi-C translation distance was observed (P > 0.05). Moreover, the difference in HSAP was not statistically significant at each node (P > 0.05). The OZJS and LSAS were significantly different at the final follow-up compared to the pre-operative period (P < 0.05). All the ACDF segments were observed in a stable condition at the final follow-up. Furthermore, 9 of the adjacent segments showed imaging ASD (9/82, 10.98%), and all were present at the last follow-up, of which 6 were mild, and 3 were moderate. Twenty of the 42 Mobi-C segments had no significant ABL (grade 0) 6months post-operatively (47.62%). Sixteen cases (38.10%) showed mild ABL (grade 1), and 6 cases (14.28%) showed moderate ABL (grade 2). No severe ABL occurred. The cervical hybrid surgery using Mobi-C artificial cervical discs can achieve satisfactory results. The Mobi-C segmental FE-COR-X shows a slow forward shift trend, and FE-COR-Y drops noticeably within 6months post-surgery before stabilizing. It's common to see mild to moderate ABL after cervical hybrid surgery using Mobi-C, and significant progression is unlikely in the short term. Furthermore, changes in the FE-COR after hybrid surgery in the Mobi-C segment might not affect clinical outcomes.

Forest stand spectrum reconstruction using spectrum spatial feature gathering and multilayer perceptron.

Three-dimensional spectral distributions of forest stands can provide spatial information on the physiological and biochemical status of forests, which is vital for forest management. However, three-dimensional spectral studies of forest stands are limited. In this study, LiDAR and multispectral data were collected from Masson pine stands in southern Fujian Province, China, and a method was proposed for inverting forest spectra using point clouds as a unit. First, multispectral values were mapped to a point cloud, and the isolated forest algorithm combined with K-means clustering was applied to characterize fusion data. Second, five deep learning algorithms were selected for semantic segmentation, and the overall accuracy (oAcc) and mean intersection ratio (mIoU) were used to evaluate the performance of various algorithms on the fusion data set. Third, the semantic segmentation model was used to reconfigure the class 3D spectral distribution, and the model inversion outcomes were evaluated by the peaks and valleys of the curve of the predicted values and distribution gaps. The results show that the correlations between spectral attributes and between spatial attributes were both greater than 0.98, while the correlation between spectral and spatial attributes was 0.43. The most applicable method was PointMLP, highest oAcc was 0.84, highest mIoU was 0.75, peak interval of the prediction curve tended to be consistent with the true values, and maximum difference between the predicted value and the true value of the point cloud spectrum was 0.83. Experimental data suggested that combining spatial fusion and semantic segmentation effectively inverts three-dimensional spectral information for forest stands. The model could meet the accuracy requirements of local spectral inversion, and the NIR values of stands in different regions were correlated with the vertical height of the canopy and the distance from the tree apex in the region. These findings improve our understanding of the precise three-dimensional spectral distribution of forests, providing a basis for near-earth remote sensing of forests and the estimation of forest stand health.

An online surrogate-assisted neighborhood search algorithm based on deep neural network for thermal layout optimization

Thermal layout optimization problems are common in integrated circuit design, where a large number of electronic components are placed on the layout, and a low temperature (i.e., high efficiency) is achieved by optimizing the positions of the electronic components. The operating temperature value of the layout is obtained by measuring the temperature field from the expensive simulation. Based on this, the thermal layout optimization problem can be viewed as an expensive combinatorial optimization problem. In order to reduce the evaluation cost, surrogate models have been widely used to replace the expensive simulations in the optimization process. However, facing the discrete decision space in thermal layout problems, generic surrogate models have large prediction errors, leading to a wrong guidance of the optimization direction. In this work, the layout scheme and its temperature field are represented by images whose relation can be well approximated by a deep neural network. Therefore, we propose an online deep surrogate-assisted optimization algorithm for thermal layout optimization. First, the iterative local search is developed to explore the discrete decision space to generate new layout schemes. Then, we design a deep neural network to build an image-to-image mapping model between the layout and the temperature field as the approximated evaluation. The operating temperature of the layout can be measured by the temperature field predicted by the mapping model. Finally, a segmented fusion model management strategy is proposed to online updates the parameters of the network. The experimental results on three kinds of layout datasets demonstrate the effectiveness of our proposed algorithm, especially when the required computational budget is limited.

Historical sound changes in Central Chadic (Afroasiatic)

Abstract In its entirety as much as with regard to its four branches, the Chadic language family poses challenges to the application of the Neogrammarian-school comparative method, not the least because of the immense time-depth involved and aggravated by certain typological peculiarities of Chadic phonology and morphology as inherited from its Afroasiatic ancestry. This is particularly true for Central Chadic, which—with 80 languages—is the most numerous and most diverse branch of Chadic, which in total counts almost 200 languages and thereby more than half of all known Afroasiatic languages. Both more or less ‘regular’ and ‘sporadic’ sound changes criss-cross the territories currently occupied by speakers of Central Chadic languages in Nigeria, Cameroon, and Chad, which allow confident identification of cognates. Some cognates have survived many millennia of language history practically unchanged with regard to their phonetic realisations, while others differ remarkably from the reconstructed proto-language forms beyond ad hoc recognisability. Recent historical comparative research by Richard Gravina (2014, 2015) and the present author (2022, 2023) involving 66 spoken languages and some 230 cognate lexical items have unearthed much of the linguistic histories behind the massive synchronic diversity of the modern Central Chadic languages. The recent research by the author has added to our understanding of historical Central Chadic phonology by unravelling the phonological processes by which the languages have developed sets of new phonemes, in addition to tracing more or less ‘regular’ and ‘sporadic’ sound changes that the PCC segmental inventories of vowels and consonants underwent. Modern Central Chadic languages show sets of innovative vowels and consonant whose emergence can largely be attributed, besides occasional instances of segmental fusion, to the ‘colouring’ effects of so-called prosodies. This makes the analysis of prosodic features such as palatalisation, labialisation, prenasalisation and glottalisation essential in order to understand the evolution of modern Central Chadic languages.

An effective U-Net and BiSeNet complementary network for spine segmentation

Spine segmentation with MRI plays an important role in the diagnosis and treatment of various spinal diseases, especially for multi-class segmentation of different vertebrae and intervertebral discs. However, due to the similarity between classes and differences within classes of spinal images, it is still a challenge to realize accurate multi-class fusion segmentation of different vertebrae and intervertebral discs. In this paper, we propose an effective U-Net and BiSeNet complementary network for spine segmentation to fully exploit their advantages in feature extraction. The network consists of a residual U-Net, a spatial feature extractor, and a feature fusion extractor. The residual U-Net selects the U-Net framework based on ResNet34, aiming at fully extracting the features of vertebrae and intervertebral discs in MRI. The spatial feature extractor takes the strip pooling (SP) blocks to replace the spatial extraction path in BiSeNet network, aiming at capturing long-range relations among the vertebrae and intervertebral discs and the interconnections of same class. The feature fusion extractor adopts the attention refinement modules (ARM) to aggregate the features from the residual U-Net and the spatial feature extractor, aiming at guiding the network to learn the distinguishing features between vertebrae and intervertebral discs by attention mechanism and making full use of the complementary features of the two paths. Experiments on MRI of 172 subjects show that our proposed network achieves impressive performance with mean Dice similarity coefficients of 82.802 % in all spinal structures. The proposed method has great value in the diagnosis and treatment of spinal diseases.

Clinical and Radiological Outcomes after Reducing Adult Lytic High-Grade Spondylolisthesis Using a Hybrid Technique: Combination of Percutaneous Pedicle Screws with Midline Microscopic Transforaminal Decompression.

Retrospective study. To analyze the results and effectiveness of percutaneous screws (PS) with midline microscopic transforaminal decompression (MTFD) technique in reducing adult stiff lytic high-grade spondylolisthesis (HGSL) and compare it with the conventional technique. Pedicle screw cannulation and segmental kyphosis negotiation are surgical challenges in HGSL. Open reduction is the preferred approach. PS have the advantage of optimized trajectory and minimized soft tissue exposure. The role of minimally invasive surgery in HGSL remains unknown. We propose a hybrid technique combining PS with MTFD for lytic HGSL. This study included 25 patients with adult lytic HGSL (Meyerding grade III and IV) operated using a hybrid technique from 2012 to 2015. Data were compared with retrospective data on conventional open reduction (n=23) operated from 2000 to 2015. The minimum follow-up was 5 years. Clinical outcomes were assessed using the Visual Analog Scale (VAS) score and modified Oswestry Disability Index (m-ODI). The spinopelvic and perioperative parameters were recorded. The inter-body fusion and adjacent segment degeneration (ASD) were assessed on radiographs at the final follow-up. The average age in the MTFD and open groups was 45.84±12.70 years (nine males and 16 females) and 49.26±13.33 years (eight males and 15 females), respectively. Further, 22 and three patients in the MTFD group and 19 and four in the open group had grade III and IV listhesis, respectively. The MTFD group demonstrated less operative time, blood loss, and hospital stays than the open group. Significant improvements were observed in VAS and m-ODI in subsequent follow-ups in both groups. The MTFD group fared better at 3 months but outcomes were comparable at the final follow-up. Both techniques were equally effective in restoring spinopelvic parameters. The incidence of ASD is comparable. The technique was proven effective in reducing HGSL. The long-term clinical and radiological outcomes were favorable and comparable with the conventional approach.

Fractures in the ankylosed spine are associated with poor bone quality and lower hounsfield units

Study designRetrospective study IntroductionPatients with ankylosing spinal disorders have a higher risk of fractures, highlighting the need for bone health surveillance. Bone assessment by dual energy x-ray absorptiometry (DXA) is challenging due to abnormal bone formation but measurements by quantitative computed tomography (qCT) have demonstrated higher sensitivity and specificity. However, no studies have analyzed bone quality using qCT in the ankylosed spine population to assess three-column fracture characteristics and subsequent outcomes. Methods106 patients with 115 three-column fractures were identified from 1999 to 2020. Patient demographics, Charlson comorbidity index, and injury severity score were extracted. Bone quality measured in Hounsfield units (HU), fracture characteristics, neurologic injury, and mortality were obtained. ResultsMost injuries occurred in the thoracic spine (70.4%) following a ground level fall (60.5%). HU adjacent to the fracture (127 HU) was significantly lower than the mobile segments (173 HU) (p < 0.001). Fracture adjacent HU was significantly lower in AS patients compared to DISH (109 vs 150 HU, p = 0.02, respectively) and were lower in fractures that resulted in a non-union or revision surgery (88 vs 137 HU, p = 0.04). Patients with longer fused segments were associated with multilevel and displaced fractures. ConclusionsFracture adjacent HUs within the autofused segments were significantly lower than in the mobile segments, and longer fusion segments were associated with displaced, multilevel fractures. This study reinforces the importance of assessing patients for decreased HUs as well as better understand how the length of fused segments is associated with displaced, multilevel fractures. Level of evidenceLevel III.

Clinical Effects of the End-Love Technique in the Treatment of Recurrent Lumbar Disc Herniation 1 Year After Surgery.

The number of patients with lumbar disc herniation in China is increasing year by year. Percutaneous endoscopic lumbar discectomy (PELD) is currently the main surgical method for treating lumbar disc herniation (LDH). However, with the increase in the number of surgical cases, the number of patients with recurrent lumbar disc herniation (RLDH) is also increasing. Currently, the common method in China is lumbar fusion surgery, but this surgery would cause the loss of fusion segment mobility and considerable postoperative complications. In order to solve the problem above the following technique will be studied: the technique of posterior lumbar laminectomy and nucleus pulposus removal under fully visualized spinal endoscopy (ENDO-LOVE) to treat RLDH. Its clinical effects will be observed in this paper, too. This series includes RLDH patients treated with ENDO-LOVE technology between January 2017 and January 2021. All patients will undergo at least three follow-up visits one year after surgery. The modified MacNab standard, VAS, JOA, and ODI scores will be used to evaluate clinical efficacy, observe for cerebrospinal fluid leak, nerve root injury, and surgical site infection, and evaluate clinical safety. All 29 patients completed the surgery successfully. Three patients had postoperative pain and numbness in the area of nerve root innervation, and all patients had no serious complications. The VAS, JOA scores and ODI indices of back pain and leg pain 1-day, 3-months, and 1-year postoperatively differed statistically significantly from the preoperative scores (p < 0.05). Efficacy evaluated at 1-year postoperatively using the modified MacNab criteria showed an excellent rate of 89.7%. ENDO-LOVE technology has demonstrated good clinical efficacy and safety in the treatment of patients with RLDH. It should be considered for all patients with this condition.

Improving Existing Segmentators Performance with Zero-Shot Segmentators.

This paper explores the potential of using the SAM (Segment-Anything Model) segmentator to enhance the segmentation capability of known methods. SAM is a promptable segmentation system that offers zero-shot generalization to unfamiliar objects and images, eliminating the need for additional training. The open-source nature of SAM allows for easy access and implementation. In our experiments, we aim to improve the segmentation performance by providing SAM with checkpoints extracted from the masks produced by mainstream segmentators, and then merging the segmentation masks provided by these two networks. We examine the "oracle" method (as upper bound baseline performance), where segmentation masks are inferred only by SAM with checkpoints extracted from the ground truth. One of the main contributions of this work is the combination (fusion) of the logit segmentation masks produced by the SAM model with the ones provided by specialized segmentation models such as DeepLabv3+ and PVTv2. This combination allows for a consistent improvement in segmentation performance in most of the tested datasets. We exhaustively tested our approach on seven heterogeneous public datasets, obtaining state-of-the-art results in two of them (CAMO and Butterfly) with respect to the current best-performing method with a combination of an ensemble of mainstream segmentator transformers and the SAM segmentator. The results of our study provide valuable insights into the potential of incorporating the SAM segmentator into existing segmentation techniques. We release with this paper the open-source implementation of our method.

Orthopedic mechanism analysis of growing rod distraction for early-onset scoliosis based on 3D morphological parameters.

Traditional growing rod (TGR) provides a corrective moment for deformed segments to straighten the spine, whose clinical efficacy has proven positive and growth-friendly. However, an insufficient understanding of orthopedic mechanisms can affect the development of clinical strategies. This research attempts to analyze the spine that has undergone four distraction operations: exploring the spinal orthopedic mechanism, including alignment, growth, and morphology. In this study, the spinal morphology curves were illustrated in three human planes to exhibit the changes in spinal alignment. The spinal growth characteristics were measured to discuss the unsynchronized and diminishing growth rate. The spinal deformations were evaluated to indicate asymmetric growth. As a result, the spinal alignment changes indicated the orthopedic process improved, but the re-unbalance occurred after multiple distractions. Then, unsynchronized growth existed in the superior and inferior segments, and the growth rate over every distraction diminished. Finally, asymmetric growth was indicated as the axial/circumferential growth ratio getting greater and the cuneate level approaching normal. Accordingly, a TGR is growth-friendly, but combining the osteotomy fusion of lumbar segments for severe early-onset scoliosis may be an excellent choice to solve the insufficient corrective stimulation. Regarding the distraction process, reshaping before the final fusion can fix the balance loss, and a prolonged distraction frequency fits the law of diminishing return. In conclusion, studying orthopedic mechanisms based on morphological measurement can guide clinical strategy optimization.

Edge-assisted Object Segmentation Using Multimodal Feature Aggregation and Learning

Object segmentation aims to perfectly identify objects embedded in the surrounding environment and has a wide range of applications. Most previous methods of object segmentation only use RGB images and ignore geometric information from disparity images. Making full use of heterogeneous data from different devices has proved to be a very effective strategy for improving segmentation performance. The key challenge of the multimodal fusion-based object segmentation task lies in the learning, transformation, and fusion of multimodal information. In this article, we focus on the transformation of disparity images and the fusion of multimodal features. We develop a multimodal fusion object segmentation framework, termed the Hybrid Fusion Segmentation Network (HFSNet). Specifically, HFSNet contains three key components, i.e., disparity convolutional sparse coding (DCSC), asymmetric dense projection feature aggregation (ADPFA), and multimodal feature fusion (MFF). The DCSC is designed based on convolutional sparse coding. It not only has better interpretability but also preserves the key geometric information of the object. ADPFA is designed to enhance texture and geometric information to fully exploit nonadjacent features. MFF is used to perform multimodal feature fusion. Extensive experiments show that our HFSNet outperforms existing state-of-the-art models on two challenging datasets.

Prediction of thermal comfort indoors and cooling loads based on reasonable zoning using the improved HHO with multi-strategy fusion-FENN algorithm

A comfortable indoor thermal environment helps occupants work healthily and efficiently, and the air-conditioning system, as the most energy-consuming system in building operation, accurate energy modeling while balancing thermal comfort can provide effective control strategies for managers. Metaheuristic algorithms are widely optimized for various types of machine learning models for the study of data-driven. However, the majority of them currently use the default hyperparameter or random and grid search, which are prone to obtaining inaccurate prediction results. In this paper, a sizable commercial building in Shaanxi Province is used as a research sample. The regional per floor firstly is divided with business model distribution, the cross-validation of random forest and correlation analysis is used to extract features, and the input factors of the prediction model are decided, and then the fully elman neural network is optimized using a segmental fusion strategy combining Cauchy operator and reverse learning to promote harris hawks optimization algorithm, and the prediction model the IMFHHO-FENN (improved Harris Hawks Optimization with multi-strategy fusion-Fully Elman Neural Network) is trained and constructed, which is finally used for predicting thermal comfort and cooling load. According to verification results, it can be proved that the IMFHHO-FENN model has excellent performance compared to mainstream single and combined models and can provide accurate results of predicted models in the design phase of building energy efficiency and later operation management, reducing energy consumption waste while satisfying personnel thermal comfort as much as possible and supply marked theoretical and decision reference for building equipment management.

Comparison of Staged vs Same-Day Circumferential Spinal Fusions for Adult Spinal Deformity.

Patients often undergo circumferential (anterior and posterior) spinal fusions to maximize adult spinal deformity (ASD) correction and achieve adequate fusion. Currently, such procedures are performed in staged (ST) or same-day (SD) procedures with limited evidence to support either strategy. This study aims to compare perioperative outcomes and costs of ST vs SD circumferential ASD corrective surgeries. This is a retrospective review of patients undergoing circumferential ASD surgeries between 2013 and 2018 in a single institution. Patient characteristics, preoperative comorbidities, surgical details, perioperative complications, readmissions, total hospital admission costs, and 90-day postoperative care costs were identified. All variables were tested for differences between ST and SD groups unadjusted and after applying inverse probability weighting (IPW), and the results before and after IPW were compared. The entire cohort included a total of 211 (ST = 50, SD = 161) patients, 100 of whom (ST = 44, SD = 56) underwent more than 4 levels fused posteriorly and anterior lumbar interbody fusion (ALIF). Although patient characteristics and comorbidities were not dissimilar between the ST and SD groups, both the number of levels fused in ALIF and posterior spinal fusion (PSF) were significantly different. Thus, using IPW, we were able to minimize the cohort incongruities in the number of levels fused in ALIF and PSF while maintaining comparable patient characteristics. In both the whole cohort and the long segment fusions, postoperative pulmonary embolism was more common in ST procedures. After adjustment utilizing IPW, both groups were not significantly different in disposition, 30-day readmissions, and reoperations. However, within the whole cohort and the long segment fusion cohort, the ST group continued to show significantly increased rates of pulmonary embolism, longer length of stay, and higher hospital admission costs compared with the SD group. Adjusted comparisons between ST and SD groups showed staging associated with significantly increased length of stay, risk of pulmonary embolism, and admission costs.

A Novel Particle Size Detection System Based on RGB-Laser Fusion Segmentation With Feature Dual-Recalibration for Blast Furnace Materials

Particle size detection (PSD) is used to obtain the particle size distribution of materials in the blast furnace charging process, which is significant for optimizing the gas flow distribution and ensuring stable production. However, due to the complex surface texture of the materials and the uneven illumination of the production environment, existing methods have difficulty obtaining the particle size distribution efficiently. This paper proposes an end-to-end PSD system based on image segmentation to obtain the particle size distribution online with high accuracy. First, to further enhance the expression of edge features and reduce the interference of complex textures, an RGB-laser particle segmentation network (RLPNet) is developed to obtain high-precision segmentation images by camera-LiDAR sensor fusion. Moreover, to improve the fusion of RGB and laser features, a feature dual-recalibration (FDR) module was designed and embedded in RLPNet, consisting of independent recalibration and joint recalibration with T-convolution. Finally, to reduce the error caused by missing edge particle pixels, an edge-recognition-based particle size calculation strategy (ERP) is presented. Experimental results demonstrate that the proposed method performs well on the constructed dataset and in industrial applications. With the segmentation accuracy of RLPNet reaching 64.19 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\%$</tex-math></inline-formula> , the similarity between the particle size distribution predicted with ERP and the actual distribution reaches 79.19 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\%$</tex-math></inline-formula> .

Risk Factors for Distal Junctional Problems Following Long Instrumented Fusion for Degenerative Lumbar Scoliosis: Are they Related to the Paraspinal Muscles.

Although the incidence of distal junctional problems (DJPs) following long construct-based treatment for degenerative lumbar scoliosis (DLS) is lower, affected patients are more likely to require revision surgery when they occur. So the aim of this study is to identify risk factors associated with DJPs to avoid its occurrence by at least 1-year follow-up. A total of 182 DLS patients undergoing long instrumented fusion surgery (≥4 levels) between February 2011 and March 2022 were retrospectively analyzed. Patients were placed into the DJP group if a DJP occurred at the final follow-up; patients without mechanical complications were matched 1:2 according to age, sex and BMI as the control group. Patient characteristics, surgical variables, radiographic parameters, lumbar muscularity and fatty degeneration were analyzed statistically. The statistical differences in the results between the two groups (p values <0.05) and other variables selected by experts were entered into a multivariate logistic regression model, and the forwards likelihood ratio method was used to analyze the independent risk factors for DJPs. Twenty-four (13.2%) patients suffered a DJP in the postoperative period and the reoperation rate was 8.8%. On univariate analysis, the lowest instrumented vertebra (LIV) CT value (p = 0.042); instrumented levels (p = 0.030); preoperative coronal vertical axis (CVA) (p = 0.046), thoracolumbar kyphosis (TLK) (p = 0.006), L4-S1 lordosis (p = 0.013), sacral slop (SS) (p = 0.030), pelvic tilt (PT) classification (p = 0.004), and sagittal vertical axis (SVA) (p = 0.021); TLK correction (p = 0.049); post-operative CVA (p = 0.029); Overall, There was no significant difference in the paraspinal muscle parameters between the two groups. On multivariate analysis, instrumented levels (OR = 1.595; p = 0.035), preoperative SVA (OR = 1.016; p = 0.022) and preoperative PT (OR = 0.873; p = 0.001) were identified as significant independent risk factors for DJP. Longer instrumented levels, a greater preoperative SVA and a smaller PT were found to be strongly associated with the presence of DJPs in patients treated for DLS. The degeneration of the paraspinal muscles may not be related to the occurrence of DJPs. For DLS patients, the occurrence of DJP can be reduced by selecting reasonable fusion segments and evaluating the patient's sagittal balance and spino-pelvic parameters before operation.

Early efficacy and safety of mini open (air/water medium) endoscopy assisted anterior cervical discectomy and fusion in the treatment of cervical spondylotic myelopathy

Objective: To investigate the efficacy and safety of mini open (air/water medium) endoscopy assisted anterior cervical discectomy and fusion (MOEA-ACDF) for the treatment of cervical spondylotic myelopathy (CSM). Methods: A follow-up study. The clinical data of 30 patients with CSM treated by MOEA-ACDF from January to December in 2021 in the Henan NO.3 Provincial People's Hospital were retrospectively analyzed. Of the patients, 20 were male and 10 were female, the mean age was (49.8±9.3) years (ranged 28-70 years). The CSM occurred at C3-4 level in 2 cases, at C4-5 level in 3 cases, at C5-6 level in 22 cases and at C6-7 level in 3 cases. Each case was compared at the moment of pre-operation and final follow-up by the Japanese Orthopedic Association (JOA) score, C2-7 Cobb angle, and anterior column height of surgical segment. The postoperative complications were recorded. Prevertebral soft tissue edema and hydrops were assessed. The fusion rate was evaluated. The JOA improvement rate was computed at the final follow-up. Results: All the operations were successfully completed and all the patients received follow-up for (12.7±2.7) months (ranged 9-20 months). The mean operation time was (85.3±11.0) min (ranged 65-110 min). The postoperative drainage volume was (16.7±7.4) ml (ranged 5-35 ml). The JOA score and the C2-7 Cobb angle both improved at the final follow-up when compared with those before the operation (15.3±1.3 vs 12.2±2.3, 15.5°±6.1° vs 12.3°±6.0°, both P<0.001). The anterior column height of surgical segment at the final follow-up was (35.6±2.5) mm, and it was higher than that before the operation [(34.1±2.4) mm](P<0.001). No postoperative complications such as dysphagia, hoarseness, cerebrospinal fluid leakage, nerve injury, hematoma occurred. Postoperative review of cervical MRI revealed 3 cases of prevertebral soft tissue edema and hydrops without obvious symptoms. At the final follow-up, cervical spine X-ray or CT showed that all fusion segments met the criteria for osseous fusion, and the fusion rate was 100%. No complications such as neurological aggravation, internal fixation failure, fusion cage sinking, and adjacent segment degeneration was recorded at the final follow-up. At the final follow-up, the comprehensive efficacy evaluated by JOA improvement rate indicated the excellent and good rate was 90.0%(27/30): 19 cases got an excellent outcome, 8 cases got good and 3 cases got medium outcome. Conclusion: MOEA-ACDF combines the endoscopic system with ACDF technology in the treatment of CSM can achieve satisfactory clinical efficacy with high safety, and effectively restore the cervical intervertebral height and physiological curvature.

A Deep Learning Network for Individual Tree Segmentation in UAV Images with a Coupled CSPNet and Attention Mechanism

Accurate individual tree detection by unmanned aerial vehicles (UAVs) is a critical technique for smart forest management and serves as the foundation for evaluating ecological functions. Existing object detection and segmentation methods, on the other hand, have reduced accuracy when detecting and segmenting individual trees in complicated urban forest landscapes, as well as poor mask segmentation quality. This study proposes a novel Mask-CSP-attention-coupled network (MCAN) based on the Mask R-CNN algorithm. MCAN uses the Cross Stage Partial Net (CSPNet) framework with the Sigmoid Linear Unit (SiLU) activation function in the backbone network to form a new Cross Stage Partial Residual Net (CSPResNet) and employs a convolutional block attention module (CBAM) mechanism to the feature pyramid network (FPN) for feature fusion and multiscale segmentation to further improve the feature extraction ability of the model, enhance its detail information detection ability, and improve its individual tree detection accuracy. In this study, aerial photography of the study area was conducted by UAVs, and the acquired images were used to produce a dataset for training and validation. The method was compared with the Mask Region-based Convolutional Neural Network (Mask R-CNN), Faster Region-based Convolutional Neural Network (Faster R-CNN), and You Only Look Once v5 (YOLOv5) on the test set. In addition, four scenes—namely, a dense forest distribution, building forest intersection, street trees, and active plaza vegetation—were set up, and the improved segmentation network was used to perform individual tree segmentation on these scenes to test the large-scale segmentation ability of the model. MCAN’s average precision (AP) value for individual tree identification is 92.40%, which is 3.7%, 3.84%, and 12.53% better than that of Mask R-CNN, Faster R-CNN, and YOLOv5, respectively. In comparison to Mask R-CNN, the segmentation AP value is 97.70%, an increase of 8.9%. The segmentation network’s precision for the four scenes in multi-scene segmentation ranges from 95.55% to 92.33%, showing that the proposed network performs high-precision segmentation in many contexts.

Bone tumor examination based on FCNN-4s and CRF fine segmentation fusion algorithm

Background and objectiveBone tumor is a kind of harmful orthopedic disease, there are benign and malignant points. Aiming at the problem that the accuracy of the existing machine learning algorithm for bone tumor image segmentation is not high, a bone tumor image segmentation algorithm based on improved full convolutional neural network which consists fully convolutional neural network (FCNN-4s) and conditional random field (CRF). MethodologyThe improved fully convolutional neural network (FCNN-4s) was used to perform coarse segmentation on preprocessed images. Batch normalization layers were added after each convolutional layer to accelerate the convergence speed of network training and improve the accuracy of the trained model. Then, a fully connected conditional random field (CRF) was fused to refine the bone tumor boundary in the coarse segmentation results, achieving the fine segmentation effect. ResultsThe experimental results show that compared with the traditional convolutional neural network bone tumor image segmentation algorithm, the algorithm has a great improvement in segmentation accuracy and stability, the average Dice can reach 91.56%, the real-time performance is better. ConclusionCompared with the traditional convolutional neural network segmentation algorithm, the algorithm in this paper has a more refined structure, which can effectively solve the problem of over-segmentation and under-segmentation of bone tumors. The segmentation prediction has better real-time performance, strong stability, and can achieve higher segmentation accuracy.

Influence of Simulated State of Disc Degeneration and Axial Stiffness of Coupler in a Hybrid Performance Stabilisation System on the Biomechanics of a Spine Segment Model.

Spinal fusion surgery leads to the restriction of mobility in the vertebral segments postoperatively, thereby causing stress to rise at the adjacent levels, resulting in early degeneration and a high risk of adjacent vertebral fractures. Thus, to address this issue, non-fusion surgery applies some pedicle screw-based dynamic stabilisation systems to provide stability and micromotion, thereby reducing stress in the fusion segments. Among these systems, the hybrid performance stabilisation system (HPSS) combines a rigid rod, transfer screw, and coupler design to offer a semi-rigid fixation method that preserves some mobility near the fusion site and reduces the adjacent segment compensatory effects. However, further research and confirmation are needed regarding the biomechanical effects of the dynamic coupler stiffness of the HPSS on the intrinsic degenerated adjacent segment. Therefore, this study utilised the finite element method to investigate the impact of the coupler stiffness of the HPSS on the mobility of the lumbar vertebral segments and the stress distribution in the intervertebral discs under flexion, extension, and lateral bending, as well as the clinical applicability of the HPSS on the discs with intrinsic moderate and severe degeneration at the adjacent level. The analytical results indicated that, regardless of the degree of disc degeneration, the use of a dynamic coupler stiffness of 57 N/mm in the HPSS may reduce the stress concentrations at the adjacent levels. However, for severely degenerated discs, the postoperative stress on the adjacent segments with the HPSS was still higher compared with that of the discs with moderate degeneration. We conclude that, when the discs had moderate degeneration, increasing the coupler stiffness led to a decrease in disc mobility. In the case of severe disc degeneration, the effect on disc mobility by coupler stiffness was less pronounced. Increasing the coupler stiffness ked to higher stress on intervertebral discs with moderate degeneration, while its effect on stress was less pronounced for discs with severe degeneration. It is recommended that patients with severe degeneration who undergo spinal dynamic stabilisation should remain mindful of the risk of accelerated adjacent segment degeneration.