Selective Fusion Research Articles

Preoperative factors associated with optimal outcomes of selective thoracic fusion at 5years.

Prior work identified optimal outcomes at 2years following selective thoracic fusion (STF) for adolescent idiopathic scoliosis (AIS) but it is unclear whether these published predictors represent what is required to achieve an optimal outcome with further time and potential growth. The purpose of this study was to determine the preoperative factors associated with optimal outcome of STF for AIS at 5years. Patients with primary thoracic AIS (Lenke 1-4C curves) who underwent a STF and had minimum 5years of follow-up were included. Optimal postoperative outcomes for a STF included a deformity-flexibility quotient (DFQ) < 4, lumbar curve < 26°, lumbar correction > 37%, C7-CSVL < 2cm, lumbar prominence < 5°, and trunk shift < 1.5cm. These outcomes were used to determine whether adhering to published recommendations for STF increased the likelihood of obtaining an optimal outcome at 5years, which included: preoperative lumbar curve < 45°, lumbar bend < 25°, apical vertebral translation ratio > 1.2, and thoracic/thoracolumbar Cobb ratio > 1.2. 127 patients met inclusion. A preoperative lumbar curve < 45° was associated with an increased likelihood of achieving three of the optimal outcomes: DFQ < 4, lumbar curve < 26°, and lumbar prominence < 5°. Following the 25° bend rule resulted in two optimal outcomes, while Cobb ratio > 1.2 was only associated with one optimal outcome. AVT > 1.2 was not significantly associated with any optimal outcome measures. This study found that at 5years, performing a STF when there is a preoperative lumbar Cobb < 45° remained the best guideline for increasing the likelihood of an optimal outcome.

Multi-scale multi-modal fusion for object detection in autonomous driving based on selective kernel

Fusion object detection using camera and LiDAR information in autonomous driving is still a challenging task, the difference between sensor data increases the difficulty of data fusion. To address this issue, we propose a multi-scale selective kernel fusion(MSSKF) method and demonstrate its practical utility by using LiDAR-camera fusion in object detection network. Specifically, a multi-scale feature fusion module that uses multi-scale convolution to separate the feature expression of multi-modal information and calculates the weight of each modal feature channel is proposed. We use the idea of multi-scale convolution and selection kernel to complete multi-modal fusion in object detection, which is conducive to solving the problem that the image and point cloud fusion are difficult to match due to the difference in data structure, and the complementarity of multi-modal information has been fully utilized. To verify the effectiveness of MSSKF, experiments on the KITTI object detection benchmark dataset are conducted. It has been observed that the proposed method achieves more accurate detection for pedestrians and vehicles, with a 1.6% gain in AP50 compared to the values of the original fusion method, reaching a score of 90.1%, and the mAP reached 60.9%. Experiments show that the proposed method introduces a new optimization idea for multi-modal fusion in the field of autonomous driving object detection, and the fusion detection efficiency is at over 12 fps on a single GPU.

Novel Selective Laser Printing Via Powder Bed Fusion of Ionic Liquid Harvested Iron for Martian Additive Manufacturing

Novel Selective Laser Printing Via Powder Bed Fusion of Ionic Liquid Harvested Iron for Martian Additive Manufacturing

Three-Body Abrasion-Corrosion Behavior of As-Printed and Solution-Annealed Additively Manufactured 316L Stainless Steel

Selective laser melting (SLM) or powder bed fusion is a type of additive manufacturing technology with applications in, e.g., the orthopedics, energy, and aerospace industries. Several studies investigated the localized corrosion behavior of SLM-fabricated Type 316L (UNS S31603) stainless steel. However, little is known about the effects of tribocorrosive conditions on the response of stainless steels fabricated by SLM. In this study, the effects of third-body abrasive particles on the tribo-electrochemical behavior of SLM 316L stainless steel produced by SLM were investigated and compared with wrought counterparts (including UNS S31703, 317W) in 0.6 M NaCl. It was found that the presence of Mo played a more decisive role in the tribocorrosion behavior than the manufacturing method, i.e., 317W revealed the best tribocorrosion behavior vis-a-vis wrought 316L and the SLM-fabricated specimens. The improved tribocorrosion behavior contrasted with the much higher breakdown potential of the SLM-fabricated samples. Nano-scale secondary ion mass spectroscopy was used to investigate the effects of Mo on passivity. The implications of passivity and tribocorrosion behavior are discussed.

Investigation of the influence of the parameters of selective laser fusion of metal powders and cold isostatic pressing on the structure and mechanical properties of a nickel-based heat-resistant alloy

The influence of the parameters of selective laser melting (SLM) of metal powders and cold isostatic pressing (CIP) of samples from a high-chromium heat-resistant alloy on the microstructure of the processed alloy, the size and number of formed pores is investigated. The use of HIP for SLM alloys makes it possible to reduce the number and size of pores. The strengthening effect of deformation (increase in hardness) at normal temperature is established. Keywords: selective laser melting, additive manufacturing, track, pores, cold isostatic pressing, mechanical properties, microstructure. remshev@mail.ru, Galina@mail.ru, grisha.afimin@mail.ru, gulyarenko@mail.ru

High quality proposal feature generation for crowded pedestrian detection

Occlusion is a severe problem for pedestrian detection in crowded scenes. Due to the diversity of pedestrian postures and occlusion forms, leading to false detection and missed detection. In this paper, we propose a high quality proposal feature generation pedestrian detection algorithm to improve detection performance. Firstly, Dual-Region Feature Generation (DRFG) is proposed to generate high quality proposal features. Specifically, visible regions with less occlusion are introduced and low-precision proposals are generated for both the full-body and visible regions respectively. Then, proposals are respectively selected from the two kinds of proposals mentioned above to match in pairs, so as to guarantee a strong correspondence in information between the two proposals. Afterwards, the successfully matched proposal features are fused by Selective Kernel Feature Fusion (SKFF) to generate high quality proposal features. Secondly, Paired Multiple Instance Prediction(PMIP) is performed on the fused features to generate multiple prediction branches, and each prediction branch generates full-body and visible prediction box. Finally, Paired Non-Maximum Suppression(PNMS) is applied to the prediction boxes to reduce the false positives. Experiments have been conducted on CrowdHuman [1] and CityPersons [2] datasets. Comparing with baseline, our methods have achieved 5.9% AP and 1.5% MR−2 improvement on the above two datasets, sufficiently verifying the effectiveness of our methods in crowded pedestrian detection.

ОТРАБОТКА ЭФФЕКТИВНОГО РЕЖИМА СПЛАВЛЕНИЯ МЕТАЛЛОПОРОШКОВОЙ КОМПОЗИЦИИ ПР-12Х18Н10Т

The use of metal-powered composition made of steel with chromium and nickel additives 12X18H10T for use in additive 3D printing technologies by selective laser fusion method (SLM) is considered. The effect of laser radia-tion power and the scanning rate of the laser beam on the solidness of the resulting steel in samples was investigated. Features of the for-mation of material in the process of selective laser melting of metal, the technological features of the equipment are determined. Results on the physicomechanical properties of the alloy ob-tained by SLM are obtained: the strength of the samples after quenching at a temperature of 1050 ° C - 660 MPa; The yield strength is 356 MPa, the relative elongation is 40 %.

Subjective perception of spinal deformity after selective versus non-selective fusion of Lenke 1C curves

PurposeTo assess the self-image perception and the Quality-of-Life (QoL) in female adolescents, with Lenke 1C scoliosis curves, treated with selective versus non-selective posterior spinal instrumentation and fusion (PSF).MethodsPatients undergoing PSF for idiopathic adolescent scoliosis (AIS) were recruited and divided into two groups: patients managed with selective thoracic fusion (STF) were included in Group A, whereas patients treated with non-selective fusion (N-STF) in Group B. Each patient completed the Italian version of the Scoliosis Research Society-22R questionnaire (SRS-22R), the Quality-of-Life Profile for Spinal Deformities questionnaire (QLPSD) and the Spinal Appearance Questionnaire (SAQ), before surgery and at 24-month follow-up.ResultsOne hundred and fifty seven female patients (mean age 16.38) were included in this study. 80 patients underwent STF, while 77 patients received N-STF. At 24-month follow-up, patients managed with N-STF showed better SRS-22R self-image mean score (p = .012), SRS-22R satisfaction mean score (p = .033), QLPSD body image mean score (p = .005), but worse SRS-22 function mean score (p = .006) and QLPSD back flexibility mean score (p = .007), compared with patients who underwent STF. In terms of self-image perception, patients undergoing STF showed significantly worse SAQ total mean score (p = .002), SAQ appearance mean score (p = .001) and SAQ expectation (p = .001). We found a significant correlation between SAQ appearance mean score and SRS-22R self-image (R = − 0.721), SRS-22 mental health (R = − 0.8), QLPSD psychosocial functioning (R = 0.7) and QLPSD back flexibility (R = 0.8).ConclusionAlthough the STF of Lenke 1C curves provides better functional outcomes, in the present study, female patients receiving STF revealed a worse perceived body image, compared with patients treated with N-STF, at 24-month follow-up. Particular attention should be addressed to the preoperative patient’s mental health and body image perception, when choosing between STF and N-STF.

Brain metastasis in gastroesophageal adenocarcinoma: Outcomes and molecular features.

347 Background: Brain metastases (BrM) rarely occur in patients with metastatic gastroesophageal adenocarcinoma (GEA) and represent a unique therapeutic challenge. We describe the unique clinical, molecular, and genomic factors associated with mGEA cancer and BrM development in order to help guide future clinical management. Methods: Patients (pts) with GEA seen at Memorial Sloan Kettering Cancer Center (MSKCC) from 2008-2019 and who had consented for genomic tumor profiling with MSK-IMPACT, a capture-based next-generation sequencing platform that detects mutations, copy-number variations, and select fusions, were retrospectively identified. Clinical and pathologic characteristics were reviewed. BrM were identified via International Classification of Diseases (ICD) billing codes and electronic medical record problem lists, and then manually validated. Survival was calculated from the time of BrM diagnosis until date of death or last follow up and estimated using the Kaplan-Meier method. Results: Fifty pts with GEA metastatic to the brain were identified. Most pts were male (86%) and white (80%), with primary tumor of the esophagus/gastroesophageal junction (82%) and intestinal-type Lauren histology (90%). Twenty-three pts (46%) were HER2 positive (defined as IHC 2+/FISH+ or IHC 3+). Frequencies of PTEN (16%) and EGFR (22%) alterations in primary or metastatic sites were enriched in pts with BM compared to that seen across the MSKCC retrospective cohort and the GEA Cancer Genome Atlas (TCGA) cohort. The majority (68%) of pts had stage IV disease at initial diagnosis, and 4 pts were found to have BrM within 1 month (mo) of stage IV diagnosis, while 27 pts developed BrM during therapy. Median time to BrM diagnosis was 18.3 mos (IQR 11.5-28.9) and 15.1 mos from stage IV diagnosis (IQR 4.8-25.5). Median survival was 7.6 mos from BrM diagnosis and 15.6 (95% CI 10.0-NR, n = 19), 7.6 (95% CI 2.5-NR, n = 13), and 4.3 (95% CI 3.5-12.3, n = 18) mos for pts with 1, 2-3, or 4+ BrM, respectively. Conclusions: GEA pts with BrM had increased frequency of HER2 positivity, as well as PTEN and EGFR alterations, compared to GEA pts overall historically. Further correlation between BrM development, molecular characteristics, and survival in a larger cohort will be presented.

Selective laser powder bed fusion for manufacturing of 3D metal-ceramic multi-materials assemblies

The technology of laser fusion on a powder bed is modified for the manufacture in air, of multi-material parts formed in the three dimensions of space, with zones which can be either a dielectric ceramic, based on alumina, or an aluminum - silicon metallic conductor. This additive manufacturing method uses only one powder. This powder is an AlSi12 alloy previously activated by a chemical treatment, which gives it, increased reactivity to oxygen. Therefore, the only adjustment of the manufacturing conditions, through the power and the scanning speed of the laser, as well as the hatch spacing, makes it possible to choose the construction of ceramic or metallic type zones. The modification of AlSi12 powders by chemical activation is the subject of a detailed characterization, allowing the selection of treatment conditions adapted to the multi-material manufacturing process. The ceramic or metallic areas that are produced, are also characterized for subsequent use of this process, for the manufacturing of microwave devices dedicated to the space industry. • Chemical etching of AlSi12 powder to increase its reactivity toward oxygen. • Parts combining in 3D space, dielectrics and electrical conductors, manufactured by laser fusion on powder bed. • Multi-materials obtained from an activated single AlSi12 powder by selective laser oxidation and selective laser melting. • Selective laser oxidation tuned by energy density. • Potential applications to microwave devices for space industry.

Molecular profiling to identify potential therapeutic targets in hepatocellular carcinoma.

471 Background: Hepatocellular carcinoma (HCC) is the most common type of primary liver cancer and is often diagnosed at an advanced stage with limited effective therapies. The advent of comprehensive genomic molecular profiling has the potential to identify novel therapeutic options in the treatment of HCC. In this descriptive clinical utility study, we evaluated the utility of the Oncotype MAP Pan-Cancer Tissue (MAP) test to identify potential therapeutic options, including clinical trials. Methods: Between October 2018 and October 2020, we identified 41 samples from patients with HCC submitted for MAP testing. The MAP test uses tumor tissue to identify single nucleotide variants, indels, copy number alterations, and select structural variants and fusions by next-generation sequencing (NGS) of a 257 gene panel. Tumor mutational burden and microsatellite instability are also determined. A custom immunohistochemical (IHC) panel is also performed. Results from NGS and IHC are run against a proprietary knowledgebase of genomic and proteomic data that includes FDA approvals, NCCN Recommendations and published biomarker data on potentially targetable alterations. Eligibility for ongoing clinical trials is determined using the NCI database. Results: Of the 41 samples, 37 (90%) met minimum tissue requirements (3mm2 and 15% tumor cellularity) for DNA sequencing. Of these 37 samples, 36 (97%) were successfully sequenced for NGS testing. Median turnaround time from the date of accessioning to the date of laboratory report was 5 days (interquartile range (IQR) 4-6 days). The most frequently identified genomic variants characterized as pathogenic, or unknown significance, were TP53, CTNNB1 and FAT1 genes. Thirty-two samples (86%) had a genomic alteration that could potentially direct treatment. Six samples (16%) had findings directing to an approved treatment for HCC, 4 (11%) had findings for an approved treatment of a different tumor type, and 22 (59%) had findings for one or more clinical trials. Conclusions: The Oncotype MAP Pan-Cancer Tissue test identified molecular variants that could be used to inform treatment decisions in 86% of HCC samples, including 16% to an approved HCC therapy. The Oncotype MAP test may be considered as another tool in the management of advanced HCC patients.

On some Aspects of the Refinement of the Temperature Distribution in the Developed Software Toolkit for Modeling Powder Heating by a Laser

Nowadays, the use of additive manufacturing is gaining popularity. The cost of product manufacturing with additive manufacturing (AM) in large-scale production is usually higher than the traditional technologies (for example, casting, milling), therefore AM is often used for prototyping and small-scale production. The paper presents results of powder heating by a laser beam modeling, powder melting modeling, and area-bath melt formation. The presented results are part of the work of solver development for modeling the selective laser melting (SLM) process. SLM technology makes it possible to manufacture a product with a complex shape in a fairly short time. As a material SLM provides for the use of powder metals and alloys, including titanium, aluminum, stain­less steel. In this case, the shape of the manufactured parts is often prone to deformations and cracks. The reason for the deformation is the presence of inherited stresses due to cyclic uneven heating and cooling, arising during layer-by-layer melting of powder particles. To solve this problem, it was decided to develop a software tool that would analyze the manufacturing process and correct it, reducing the occurrence of residual stresses. The article presents the results of modeling and calculating the geometric size of the melt pool in a dynamic setting. A comparison is made between the results obtained in the course of this work and the results of expe­riments. Modeling the selective fusion process entails high computational costs. The article presents a method based on mesh adaptation, which allows to reduce the computational costs of modeling. The analysis of the effectiveness of this method is carried out.

Curve overcorrection predicts coronal imbalance in selective thoracic fusion in adolescent idiopathic scoliosis

Curve overcorrection predicts coronal imbalance in selective thoracic fusion in adolescent idiopathic scoliosis

Investigation of the transition heat treatment temperature of AlSi10Mg manufactured by selective laser melting and the effect on the compressive strength

AlSi10Mg has been widely adopted for manufacture via Selective Laser Melting (SLM) or Laser Beam Powder Bed Fusion (LBPB) process, as it possesses superior mechanical properties as compared with conventional cast method. However, residual stresses in the manufactured part are an inevitable by-product of the SLM process, and post-processing such as heat treatment is commonly used to relieve the stresses. A common heat treatment method for SLM manufactured AlSi10Mg requires the specimen to be heated up to 300 °C. This treatment temperature is capable of relieving residual stresses but also reduces the mechanical strength at the same time. This paper presents an investigation of the effect of different heat treatment temperatures on the compressive mechanical strength of AlSi10Mg manufactured by SLM. It was found that the deterioration of mechanical strength of AlSi10Mg occurs abruptly over a temperature range between 294 °C and 300 °C and is supported by microstructural observation from Scanning Electron Microscope (SEM).

Subjective perception of spinal deformity after selective versus non-selective fusion of Lenke 1C curves

Subjective perception of spinal deformity after selective versus non-selective fusion of Lenke 1C curves

Clinical, radiological and HRQOL outcomes after selective thoracic fusion with minimum 20 years follow up: Assessment of the degenerative changes of unfused lumbar spine with MRI study

Clinical, radiological and HRQOL outcomes after selective thoracic fusion with minimum 20 years follow up: Assessment of the degenerative changes of unfused lumbar spine with MRI study

An Interpretable Deep Transfer Learning-Based Remaining Useful Life Prediction Approach for Bearings With Selective Degradation Knowledge Fusion

This article tries to answer the two questions of bearings’ remaining useful life (RUL) prediction with deep transfer learning: what bearing data in the source domain contribute more to transfer learning and how to quantify such contribution? From the perspective of sample-based interpretability, this article proposes a new deep transfer learning approach of RUL prediction. First, a new time series clustering algorithm based on multiscale degradation similarity is proposed. Comprehensively considering the geometry and tendency characteristics of degradation sequences, this algorithm can divide the source domain into multiple subsource domains. Second, a significance metric, named transfer domain validity index (T-DVI), is built to quantify the contribution of each subsource domain to transfer learning in terms of degradation similarity. Furthermore, a new stacked long short-term memory model with selective transfer learning is constructed. Running with the obtained T-DVIs, this model builds two new transfer strategies of weighted initialization and adaptive freezing to improve the transfer effect of degradation knowledge for RUL prediction. Experimental results on the two bearing datasets prove the effectiveness of the proposed approach. More importantly, the proposed approach makes the transfer learning process interpretable via identifying the significance of the bearings data, which helps transfer useful degradation knowledge and improve the RUL prediction performance as well.

Influence on lowest instrumented vertebra disc wedging by thoracolumbar/lumbar and thoracic Cobb ratio (TL/L:T Cobb ratio) after selective lumbar fusion in patients with adolescent idiopathic scoliosis Lenke type 5

Influence on lowest instrumented vertebra disc wedging by thoracolumbar/lumbar and thoracic Cobb ratio (TL/L:T Cobb ratio) after selective lumbar fusion in patients with adolescent idiopathic scoliosis Lenke type 5

Head Pose Estimation in Complex Environment Based on Four-Branch Feature Selective Extraction and Regional Information Exchange Fusion Network

Under the severe situation of the COVID-19 pandemic, masks cover most of the effective facial features of users, and their head pose changes significantly in a complex environment, which makes the accuracy of head pose estimation in some systems such as safe driving systems and attention detection systems impossible to guarantee. To this end, we propose a powerful four-branch feature selective extraction network (FSEN) structure, in which three branches are used to extract three independent discriminative features of pose angles, and one branch is used to extract composite features corresponding to multiple pose angles. By reducing the dimension of high-dimensional features, our method significantly reduces the amount of computation while improving the estimation accuracy. Our convolution method is an improved spatial channel dynamic convolution (SCDC) that initially enhances the extracted features. Additionally, we embed a regional information exchange network (RIEN) after each convolutional layer in each branch to fully mine the potential semantic correlation between regions from multiple perspectives and learn and fuse this correlation to further enhance feature expression. Finally, we fuse the independent discriminative features of each pose angle and composite features from the three directions of channel, space, and pixel to obtain perfect feature expression for each pose angle, and then obtain the head pose angle. We conducted extensive experiments on the controlled environment datasets and a self-built real complex environment dataset (RCE) and the results showed that our method outperforms state-of-the-art single-modality methods and performs on par with multimodality-based methods. This shows that our network meets the requirements of accurate head-pose estimation in real complex environments such as complex illumination and partial occlusion.

Feature selection and residual fusion segmentation network for liver tumor

Feature selection and residual fusion segmentation network for liver tumor