Previous Fusion Research Articles

Transforaminal lumbar interbody fusion subsidence: computed tomography analysis of incidence, associated risk factors, and impact on outcomes.

The aims of this study were to 1) define the incidence of transforaminal lumbar interbody fusion (TLIF) interbody subsidence; 2) determine the relative importance of preoperative and intraoperative patient- and instrumentation-specific risk factors predictive of postoperative subsidence using CT-based assessment; and 3) determine the impact of TLIF subsidence on postoperative complications and fusion rates. All adult patients who underwent one- or two-level TLIF for lumbar degenerative conditions at a multi-institutional academic center between 2017 and 2019 were retrospectively identified. Patients with traumatic injury, infection, malignancy, previous fusion at the index level, combined anterior-posterior procedures, surgery with greater than two TLIF levels, or incomplete follow-up were excluded. Interbody subsidence at the superior and inferior endplates of each TLIF level was directly measured on the endplate-facing surface of both coronal and sagittal CT scans obtained greater than 6 months postoperatively. Patients were grouped based on the maximum subsidence at each operative level classified as mild, moderate, or severe based on previously documented < 2-mm, 2- to 4-mm, and ≥ 4-mm thresholds, respectively. Univariate and regression analyses compared patient demographics, medical comorbidities, preoperative bone quality, surgical factors including interbody cage parameters, and fusion and complication rates across subsidence groups. A total of 67 patients with 85 unique fusion levels met the inclusion and exclusion criteria. Overall, 28% of levels exhibited moderate subsidence and 35% showed severe subsidence after TLIF with no significant difference in the superior and inferior endplate subsidence. Moderate (≥ 2-mm) and severe (≥ 4-mm) subsidence were significantly associated with decreases in cage surface area and Taillard index as well as interbody cages with polyetheretherketone (PEEK) material and sawtooth surface geometry. Severe subsidence was also significantly associated with taller preoperative disc spaces, decreased vertebral Hounsfield units (HU), the absence of bone morphogenetic protein (BMP) use, and smooth cage surfaces. Regression analysis revealed decreases in Taillard index, cage surface area, and HU, and the absence of BMP use predicted subsidence. Severe subsidence was found to be a predictor of pseudarthrosis but was not significantly associated with revision surgery. Patient-level risk factors for TLIF subsidence included decreased HU and increased preoperative disc height. Intraoperative risk factors for TLIF subsidence were decreased cage surface area, PEEK cage material, bullet cages, posterior cage positioning, smooth cage surfaces, and sawtooth surface designs. Severe subsidence predicted TLIF pseudarthrosis; however, the causality of this relationship remains unclear.

MTFR: An universal multimodal fusion method through Modality Transfer and Fusion Refinement

Multimodal learning has made great achievements in recent years. However, most of the existing modal fusion methods capture cross-modal correlations in a single stage and are designed for specific tasks, with poor generalizability and portability. In this work, we propose the Modality Transfer and Fusion Refinement method (MTFR), which contains two feature-level fusion modules and can be used for several applications. Specifically, the Modality Transfer greatly shortens the distance between modalities through transferring non-text information into text-like representations. Fusion Refinement corrects and complements the previous fusion process to make the model focus more on task-specific knowledge by introducing uni-modal processing. To illustrate the validity of MTFR, we design three experiments for sarcasm detection, sentiment analysis, and offensiveness detection. The experimental results validate that our method achieves competitive performance in all three applications and outperforms both task-specific models and multimodal fusion baselines, which demonstrate the good generalizability and robustness of our method. Thanks to the promising performance, the method will facilitate the development of multimodal fusion and can promote multimodal applications in practice such as social sentiment analysis and multimodal conversational robots.

Modified S1 Pedicle Subtraction Osteotomy.

Surgical technique video. To report a surgical technique to revise patients with previous fusions at L4-S1 leading to an iatrogenic flat back and sagittal imbalance using L5-S1 transforaminal interbody fusion combined with a small S1 corner osteotomy. This is a case of a woman (51 y old) with a history of multiple lumbar surgeries, severe back pain, sagittal imbalance, and loss of lordosis. We describe a feasible revision technique in a complex patient with the goal of attaining optimal distribution of lumbar lordosis and sagittal balance through a modified S1 pedicle subtraction osteotomy, and the use of an interbody cage to enhance the fusion rate and facilitate closure of the 3-column osteotomy. The preoperative patient lordosis angle of 31 degrees at L1-L4 and 16 degrees at L4-S1 became 12 degrees at L1-L4 and 44 degrees at L4-S1 postoperatively. The combination of L5-S1 transforaminal interbody fusion and S1 corner osteotomy is a feasible technique for the restoration of lumbar lordosis in patients with previous fusion and consequent loss of lordosis.

Benchmark Values for Construct Survival and Complications by Type of ASD Surgery.

The aim of this study was to provide benchmarks for the rates of complications by type of surgery performed. Prospective multicenter database. We have previously examined overall construct survival and complication rates for ASD surgery. However, the relationship between type of surgery and construct survival warrants more detailed assessment. Eight surgical scenarios were defined based on the levels treated, previous fusion status [primary (P) vs. revision (R)], and three-column osteotomy use (3CO): short lumbar fusion, LT-pelvis with 5 to 12 levels treated (P, R, or 3CO), UT-pelvis with 13 levels treated (P, R, or 3CO), and thoracic to lumbar fusion without pelvic fixation, representing 92.4% of the case in the cohort. Complication rates for each type were calculated and Kaplan-Meier curves with multivariate Cox regression analysis was used to evaluate the effect of the case characteristics on construct survival rate, while controlling for patient profile. A total of 1073 of 1494 patients eligible for 2-year follow-up (71.8%) were captured. Survival curves for major complications (with or without reoperation), while controlling for demographics differed significantly among surgical types ( P <0.001). Fusion procedures short of the pelvis had the best survival rate, while UT-pelvis with 3CO had the worst survival rate. Longer fusions and more invasive operations were associated with lower 2-year complication-free survival, however, there were no significant associations between type of surgery and renal, cardiac, infection, wound, gastrointestinal, pulmonary, implant malposition, or neurological complications (all P >0.5). This study suggests that there is an inherent increased risk of complication for some types of ASD surgery independent of patient profile. The results of this paper can be used to produce a surgery-adjusted benchmark for ASD surgery with regard to complications and survival. Such a tool can have very impactful applications for surgical decision-making and more informed patient counseling. Level III.

Completely Minimally Invasive Implant Removal and Transforaminal Lumbar Interbody Fusion for Adjacent Segment Disease: Case Series and Operative Video.

Adjacent segment disease is a relatively common late complication after lumbar fusion. If symptomatic, certain patients require fusion of the degenerated adjacent segment. Currently, there are no posterior completely minimally invasive techniques described for fusion of the adjacent segment above or below a previous fusion. We describe here a novel minimally invasive technique for both implant removal (MIS-IR) and adjacent level transforaminal lumbar interbody fusion (MIS-TLIF) for lumbar stenosis. Demographic, surgical, and radiographic outcome data were collected for patients with lumbar stenosis and previous lumbar fusion, who were treated with MIS-IR and MIS-TLIF through the same incision. Radiographic outcomes were assessed postoperatively and complications were assessed at the primary end point of 3 months. A total of 14 patients (7 female and 7 male), with average age 64.6 years (SD 13.4), were included in this case series. Nine patients had single-level MIS-IR with single-level MIS-TLIF. Three patients had 2-level MIS-IR with single-level MIS-TLIF. Two patients had single-level MIS-IR with 2-level MIS-TLIF. Only 1 patient had a postoperative complication-hematoma requiring same-day evacuation. There were no other complications at the primary end point and no fusion failure at the hardware removal levels to date (average follow-up, 11 months). Average increases in posterior disk height and foraminal height after MIS-TLIF were 4.44, and 2.18 mm, respectively. Minimally invasive spinal IR can be successfully completed along with adjacent level TLIF through the same incisions, via an all-posterior approach.

Single position, prone oblique lateral interbody fusion (OLIF)-case illustration and technical considerations.

Oblique lateral interbody fusion (OLIF) is a powerful method to treat various spinal conditions and is frequently combined with posterior instrumentation. This is traditionally performed in dual positions, with the patient first in lateral then turned prone. Single position lateral surgery (SPS-L) has been studied in a bid to improve surgical efficiency and reduce operative costs, but various limitations have been identified. More recently, the single position prone surgery (SPS-P) has been described as an alternative to address some of these limitations. This case illustrates a patient who underwent SPS-P using an OLIF corridor with subsequent posterior decompression and instrumentation. The benefits and limitations of this procedure compared to the conventional techniques are highlighted in this case. We present the case of a 75-year-old female presenting with thoracic myelopathy over T11/12 and concurrent L2-4 spinal stenosis. She underwent OLIF of L2/3 and L3/4, posterior decompression of T11/12 and L2/3, and posterior instrumented fusion from T10-L4 via a single prone position. We aim to describe the advantages of this approach and the challenges encountered through our experience. SPS-P offers numerous benefits compared to the already powerful SPS-L. In the upper levels of the lumbar spine, a pre-psoas approach may also be feasible. However, the prone lateral technique does not replace all patients suited for a lateral interbody fusion but should be seen as a viable option for selected cases such as those with previous fusion at the L5/S1 with adjacent degeneration requiring extension and posterior fixation.

Comparison of multilevel low-grade techniques versus three-column osteotomies in adult spinal deformity surgery: does harmonious correction matter?

Recent debate has arisen between whether to use a three-column osteotomy (3CO) or multilevel low-grade (MLG) techniques to treat severe sagittal malalignment in adult spinal deformity (ASD) surgery. The goal of this study was to compare the outcomes of 3CO and MLG techniques performed in corrective surgeries for ASD. ASD patients who had a baseline PI-LL > 30° and 2-year follow-up data were included. Patients underwent either 3CO or MLG (thoracolumbar posterior column osteotomies at ≥ 3 levels or anterior lumbar interbody fusion at ≥ 3 levels with no 3CO). The segmental utility ratio was used to assess relative segmental correction (segmental correction divided by overall correction in lordosis divided by the number of thoracolumbar interventions [interbody fusion, thoracolumbar posterior column osteotomies, and 3CO]). The paired t-test was used to assess lordotic distribution by differences in lordosis between adjacent lumbar disc spaces (e.g., L1-2 to L2-3). Multivariate analysis, controlling for age, sex, BMI, osteoporosis, baseline pelvic incidence, and T1 pelvic angle, was used to evaluate the complication rates and radiographic and patient-reported outcomes between the groups. A total of 93 patients were included, 53% of whom underwent MLG and 47% of whom underwent 3CO. The MLG group had a lower BMI (p < 0.05). MLG patients received fewer previous fusions than 3CO patients (31% vs 80%, p < 0.001). MLG patients had 24% less blood loss but a 22% longer operative time (565 vs 419 minutes, p = 0.008). Using adjusted analysis, the 3CO group had greater segmental and relative correction at each level (segmental utility ratio mean 69% for 3CO vs 23% for MLG, p < 0.001). However, the 3CO group had lordotic differences between two adjacent lumbar disc pairs (range -0.5° to 9.0°, p = 0.009), while MLG was more harmonious (range 2.2°-6.5°, p > 0.4). MLG patients were more likely to undergo realignment to age-adjusted standards (OR 5.6, 95% CI 1.2-46.4; p = 0.033). MLG patients were less likely to develop neurological complications or undergo reoperation (OR 0.4, 95% CI 0.1-0.9; p = 0.041). Adjusted analysis revealed that MLG patients more often met a substantial clinical benefit in the Oswestry Disability Index score (OR 5.3, 95% CI 1.1-26.8; p = 0.043). MLG techniques showed better utility in lumbar distribution and age-adjusted global correction while minimizing neurological complications and reoperation rates by 2 years postoperatively. In selected instances, these techniques may offer the spine deformity surgeon a safer alternative when correcting severe adult spinal deformity.

Axial Concentration of Dissolved Deuterium by Dispersion in a LiPb Experimental Loop

The axial concentration of deuterium by dispersion in a circulating liquid lithium-lead (LiPb) loop was analyzed and experimentally verified. In previous fusion blanket studies, the tritium transport rate in flowing LiPb was treated by convection a priori; i.e., the dispersion effect was negligible. In contrast, Taylor dispersion theory shows conflicting results, exhibiting axial transport enhancement via convective flow. In the current paper, the experimental setup consists of a deuterium dissolving tube that substitutes for tritium breeding and a deuterium concentration monitor by LiPb droplets in a vacuum with four nozzles of ϕ = 1.0 mm. The released deuterium mass flux from the droplets was measured using a quadrupole mass spectrometer. An electromagnetic pump circulated 49 L of LiPb at 350°C at a rate between 0.15 and 0.3 L∙min–1 with the corresponding Re number between 600 and 1000, i.e., in the laminar flow range. The dispersion coefficient was analyzed by measuring the temporal distortion of the deuterium concentration profile. The obtained axial dispersion coefficients of dissolved deuterium in LiPb were between 4.6 × 10–2 and 1.2 × 10–1 (m2∙s–1) and approximately seven orders of magnitude greater than those under static conditions. The results agreed with the Taylor dispersion theory, which studied the mass transport enhancement by convection. The applicability of Taylor’s theory to the deuterium flow in liquid LiPb is suggested, whereas the Prandtl number was three orders of magnitude lower and the Schmidt number was one order of magnitude higher than that of the water.

Combined Anterior–Posterior vs. Posterior-Only Approach in Adult Spinal Deformity Surgery: Which Strategy Is Superior?

Introduction: Whether a combined anterior–posterior (AP) approach offers additional benefits over the posterior-only (P) approach in adult spinal deformity (ASD) surgery remains unknown. In a cohort of patients undergoing ASD surgery, we compared the combined AP vs. the P-only approach in: (1) preoperative/perioperative variables, (2) radiographic measurements, and (3) postoperative outcomes. Methods: A single-institution, retrospective cohort study was performed for patients undergoing ASD surgery from 2009 to 2021. Inclusion criteria were ≥5-level fusion, sagittal/coronal deformity, and 2-year follow-up. The primary exposure was the operative approach: a combined AP approach or P alone. Postoperative outcomes included mechanical complications, reoperation, and minimal clinically important difference (MCID), defined as 30% of patient-reported outcome measures (PROMs). Multivariable linear regression was controlled for age, BMI, and previous fusion. Results: Among 238 patients undergoing ASD surgery, 34 (14.3%) patients underwent the AP approach and 204 (85.7%) underwent the P-only approach. The AP group consisted mostly of anterior lumbar interbody fusion (ALIF) at L5/S1 (73.5%) and/or L4/L5 (38.0%). Preoperatively, the AP group had more previous fusions (64.7% vs. 28.9%, p &lt; 0.001), higher pelvic tilt (PT) (29.6 ± 11.6° vs. 24.6 ± 11.4°, p = 0.037), higher T1 pelvic angle (T1PA) (31.8 ± 12.7° vs. 24.0 ± 13.9°, p = 0.003), less L1-S1 lordosis (−14.7 ± 28.4° vs. −24.3 ± 33.4°, p &lt; 0.039), less L4-S1 lordosis (−25.4 ± 14.7° vs. 31.6 ± 15.5°, p = 0.042), and higher sagittal vertical axis (SVA) (102.6 ± 51.9 vs. 66.4 ± 71.2 mm, p = 0.005). Perioperatively, the AP approach had longer operative time (553.9 ± 177.4 vs. 397.4 ± 129.0 min, p &lt; 0.001), more interbodies placed (100% vs. 17.6%, p &lt; 0.001), and longer length of stay (8.4 ± 10.7 vs. 7.0 ± 9.6 days, p = 0.026). Radiographically, the AP group had more improvement in T1PA (13.4 ± 8.7° vs. 9.5 ± 8.6°, p = 0.005), L1-S1 lordosis (−14.3 ± 25.6° vs. −3.2 ± 20.2°, p &lt; 0.001), L4-S1 lordosis (−4.7 ± 16.4° vs. 3.2 ± 13.7°, p = 0.008), and SVA (65.3 ± 44.8 vs. 44.8 ± 47.7 mm, p = 0.007). These outcomes remained statistically significant in the multivariable analysis controlling for age, BMI, and previous fusion. Postoperatively, no significant differences were found in mechanical complications, reoperations, or MCID of PROMs. Conclusions: Preoperatively, patients undergoing the combined anterior–posterior approach had higher PT, T1PA, and SVA and lower L1-S1 and L4-S1 lordosis than the posterior-only approach. Despite increased operative time and length of stay, the anterior–posterior approach provided greater sagittal correction without any difference in mechanical complications or PROMs.

Multi-modal Physiological Signal Fusion for Emotion Classification: A Multi-Head Attention Approach

In this essay, a model-level fusion technique of multi-modal physiological signals using Multi-Head Attention is studied. A framework that utilizes multi-model physiological signals for the task of emotion classification is proposed. First, the GCRNN model, which combines the Graph Convolutional Network (GCN) and the Long and Short Term Memory (LSTM), captures the unique features of electroencephalogram (EEG) signals. The spatial and temporal information that makes up impulses from the EEG can be captured precisely by such a technique. The CCRNN model, which combines the Convolutional Neural Network (CNN) integrated with the Channel-wise Attention and the LSTM, is used for peripheral physiological signals. The model can extract useful features from peripheral physiological signals and automatically learn to weigh the importance of various channels. Finally, Multi-head Attention is employed to fuse the output of the GCRNN and CCRNN methods. The Multi-head Attention can automatically learn the relevance and importance of different modal signals and weigh them accordingly. Emotion classification is implemented by adding a component of Softmax to map what the model produced to discrete emotion categories. The DEAP dataset was utilized in this study for experimental verification, and the results indicate that the method using multi-modal physiological signal fusion is substantially greater in precision than the technique using simply EEG signals. Additionally, the Multi-head Attention fusion method performs better than previous fusion techniques.

Interactive Multi-Scale Fusion of 2D and 3D Features for Multi-Object Vehicle Tracking

Multiple Object Tracking (MOT) is a significant task in autonomous driving. Nonetheless, relying on one single sensor is not robust enough, because one modality tends to fail in some challenging situations. Texture information from RGB cameras and 3D structure information from Light Detection and Ranging (LiDAR) have respective advantages under different circumstances. Therefore, feature fusion from multiple modalities contributes to the learning of discriminative features. However, it is nontrivial to achieve effective feature fusion due to the completely distinct information modality. Previous fusion methods usually fuse the top-level features after the backbones extract the features from different modalities. The feature fusion happens solely once, which limits the information interaction between different modalities. In this paper, we propose multi-scale interactive query and fusion between pixel-wise and point-wise features to obtain more discriminative features. In addition, an attention mechanism is utilized to conduct soft feature fusion between multiple pixels and points to avoid inaccurate match problems of previous single pixel-point fusion methods. We introduce PointNet <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> <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> to obtain multi-scale deep representations of point clouds and make it adaptive to our proposed interactive feature fusion between multi-scale features of images and point clouds. Through the interaction module, each modality can integrate more complementary information from the other modality. Besides, we explore the effectiveness of pre-training on each single modality and fine-tuning on the fusion-based model. Our method can achieve 90.32% MOTA and 72.44% HOTA on the KITTI benchmark and outperform other approaches without using multi-scale soft feature fusion.

FusionCPP: Cooperative fusion of infrared and visible light images based on PCNN and PID control systems

The imaging results of infrared and visible light sensors are highly complementary. Thus, combining the two images can compensate for the shortcomings of a single sensor. However, previous fusion methods have not fully utilized the collaborative guidance of infrared and visible images in the fusion process, leading to a poor correlation between extracted features and cannot guarantee feature fusion accuracy. In order to address these challenges, this paper proposes a collaborative fusion method for infrared and visible images based on pulse-coupled neural networks (PCNN) and proportional-integral-derivative (PID) control systems, called FusionCPP. First, a coupled neural network with a dual pulse structure is designed for the feature extraction section. Unlike traditional PCNN, this network includes two pulse generators, one of which is utilized to extract the salient features of the image, while the other is employed to extract the common pulse layer of the source image and multiply the pulse layer by the coupling iteration value to obtain the base layer. Second, a detail layer is obtained by subtracting the base layer from the source image. This method accurately extracts salient and detailed features based on the coupling between source images. Finally, a closed-loop PID control system is utilized as the fusion strategy in the image fusion task, which judges the difference between the fusion result and the source image in real-time through the feedback effect of the system. The controller adaptively adjusts the fusion weight based on the difference to fuse image features into the new image accurately. Experimental results indicate that the proposed FusionCPP can maintain significant contrast and rich textures. Besides, the proposed FusionCPP has extended to multi-focus image fusion and target detection tasks to verify its effectiveness.

Transforaminal Endoscopic Lumbar Foraminotomy for Juxta-Fusional Foraminal Stenosis.

Adjacent segment foraminal stenosis is a significant adverse event of lumbar fusion. Conventional revision surgery with an extended fusion segment may result in considerable surgical morbidity owing to extensive tissue injury. Transforaminal endoscopic lumbar foraminotomy (TELF) is a minimally invasive surgical approach for symptomatic foraminal stenosis. This study aimed to demonstrate the surgical technique and clinical outcomes of TELF for the treatment of juxta-fusional foraminal stenosis. Full-scale foraminal decompression was performed via a transforaminal endoscopic approach under local anesthesia. A total of 22 consecutive patients who had undergone TELF were evaluated. The included patients had unilateral foraminal stenosis at the juxta-fusional level of the previous fusion surgery, intractable lumbar radicular pain despite at least six months of non-operative treatment, and verified pain focus by imaging and selective nerve root block. The visual analog scale and Oswestry Disability Index scores significantly improved after the two-year follow-up period. The modified MacNab criteria were excellent in six patients (27.27%), good in 12 (55.55%), fair in two (9.09%), and poor in two (9.09%), with a 90.91% symptomatic improvement rate. No significant surgical complications were observed. The minimally invasive TELF is effective for juxta-fusional foraminal stenosis.

Multiresolution generative adversarial networks with bidirectional adaptive-stage progressive guided fusion for remote sensing image

ABSTRACT Remote sensing image (RSI) with concurrently high spatial, temporal, and spectral resolutions cannot be produced by a single sensor. Multisource RSI fusion is a convenient technique to realize high spatial resolution multispectral (MS) images (spatial spectral fusion, i.e. SSF) and high temporal and spatial resolution MS images (spatiotemporal fusion, i.e. STF). Currently, deep learning-based fusion models can only implement SSF or STF, lacking models that perform both SSF and STF. Multiresolution generative adversarial networks with bidirectional adaptive-stage progressive guided fusion (BAPGF) for RSI are proposed to implement both SSF and STF, namely BPF-MGAN. A bidirectional adaptive-stage feature extraction architecture in fine-scale-to-coarse-scale and coarse-scale-to-fine-scale modes is introduced. The designed BAPGF introduces a previous fusion result-oriented cross-stage-level dual-residual attention fusion strategy to enhance critical information and suppress superfluous information. Adaptive resolution U-shaped discriminators are implemented to feed multiresolution context into the generator. A generalized multitask loss function unlimited by no-reference images is developed to strengthen the model via constraints on the multiscale feature, structural, and content similarities. The BPF-MGAN model is validated on SSF datasets and STF datasets. Compared with the state-of-the-art SSF and STF models, results demonstrate the superior performance of the proposed BPF-MGAN model in both subjective and objective evaluations.

ID: 200814 Open Dorsal Root Ganglion Stimulator Lead Placement in the Setting of Previous Fusion: Case Report

Dorsal Root Ganglion (DRG) Stimulation is a relatively new neuromodulation technique that has been showing increasing promise in chronic pain management. Due to unique anatomical attributes, DRG stimulation allows for posture independent, precise, and targeted neuromodulation with lower energy use, leading to the need for fewer generator changes.

Foot &amp; Ankle

The April 2023 Foot &amp; Ankle Roundup360 looks at: Outcomes following a two-stage revision total ankle arthroplasty for periprosthetic joint infection; Temporary bridge plate fixation and joint motion after an unstable Lisfranc injury; Outcomes of fusion in type II os naviculare; Total ankle arthroplasty versus arthrodesis for end-stage ankle osteoarthritis; Normal saline for plantar fasciitis: placebo or therapeutic?; Distraction arthroplasty for ankle osteoarthritis: does it work?; Let there be movement: ankle arthroplasty after previous fusion; Morbidity and mortality after diabetic Charcot foot arthropathy.

Adaptive Multiview Graph Difference Analysis for Video Summarization

Adapting detection to different shot types is a significant challenge for video summarization methods based on shot boundary detection. In our recent work, a new graph model was introduced in the feature modelling of frames and analysed for changes in graph structure to improve the detection of shot boundaries. In this paper, we further explore the potential of graph models and propose a more general framework for online, real-time automatic video summarization. The framework develops a novel adaptive multiview graph difference analysis method to improve the algorithm’s robustness in detecting different shot transitions. Previous fusion methods typically used a priori knowledge to assign weights to the various feature differences from videos. In contrast, our framework can weigh and fuse the resulting differences by learning the importance of various video features from the structural changes of the corresponding multiview graphs. Additionally, we propose a new threshold-based adaptive decision method which can dynamically select the most accurate shot boundary decision threshold by analysing a small number of historical frames and learning the tolerance factor in the current shot. The experimental results show that the proposed method outperforms state-of-the-art methods in terms of precision and F-score on the VSUMM and YouTube datasets.

Full Endoscopic Decompression for a Misdiagnosed Far-Out Syndrome (L5/S1 Extraforaminal Stenosis) in an Elderly Patient with L5 Radiculopathy: Case Report

We present a case of a L5/S1 extraforaminal stenosis (EFS), also known as far-out syndrome, in which ineffective multiple previous fusion surgeries were performed because of misdiagnosis. An 83-year-old man, complaining of back and left sided leg pain along the L5 dermatome for 5 years visited our institute claiming persistent pain even after 2 fusion surgeries at another hospital. A new magnetic resonance image (MRI) scan with a half coronal T1-weighted sequence revealed left L5/S1 EFS and resultant L5 nerve root compression. Because of the patient’s old age, history of angina and anti-platelet medications, a minimally invasive full-endoscopic decompression was performed. Afterwards the patient experienced immediate relief of the radicular pain, satisfied with the surgery. Diagnosis of L5/S1 EFS can be challenging especially when only routine sagittal/axial MRI scans are provided. Half coronal plane T1-weighted image MRI can be extremely helpful in diagnosis of this distinct pathology. Through this case report, we would like to emphasize the significance of half coronal MRI in diagnosis of L5/S1 EFS and also show a representative successful full-endoscopic decompression for the disease.

An Image Fusion Method of SAR and Multispectral Images Based on Non-Subsampled Shearlet Transform and Activity Measure.

Synthetic aperture radar (SAR) is an important remote sensing sensor whose application is becoming more and more extensive. Compared with traditional optical sensors, it is not easy to be disturbed by the external environment and has a strong penetration. Limited by its working principles, SAR images are not easily interpreted, and fusing SAR images with optical multispectral images is a good solution to improve the interpretability of SAR images. This paper presents a novel image fusion method based on non-subsampled shearlet transform and activity measure to fuse SAR images with multispectral images, whose aim is to improve the interpretation ability of SAR images easily obtained at any time, rather than producing a fused image containing more information, which is the pursuit of previous fusion methods. Three different sensors, together with different working frequencies, polarization modes and spatial resolution SAR datasets, are used to evaluate the proposed method. Both visual evaluation and statistical analysis are performed, the results show that satisfactory fusion results are achieved through the proposed method and the interpretation ability of SAR images is effectively improved compared with the previous methods.

EMM-LC Fusion: Enhanced Multimodal Fusion for Lung Cancer Classification

Lung cancer (LC) is the most common cause of cancer-related deaths in the UK due to delayed diagnosis. The existing literature establishes a variety of factors which contribute to this, including the misjudgement of anatomical structure by doctors and radiologists. This study set out to develop a solution which utilises multiple modalities in order to detect the presence of LC. A review of the existing literature established failings within methods to exploit rich intermediate feature representations, such that it can capture complex multimodal associations between heterogenous data sources. The methodological approach involved the development of a novel machine learning (ML) model to facilitate quantitative analysis. The proposed solution, named EMM-LC Fusion, extracts intermediate features from a pre-trained modified AlignedXception model and concatenates these with linearly inflated features of Clinical Data Elements (CDE). The implementation was evaluated and compared against existing literature using F1 score, average precision (AP), and area under curve (AUC) as metrics. The findings presented in this study show a statistically significant improvement (p &lt; 0.05) upon the previous fusion method, with an increase in F-Score from 0.402 to 0.508. The significance of this establishes that the extraction of intermediate features produces a fertile environment for the detection of intermodal relationships for the task of LC classification. This research also provides an architecture to facilitate the future implementation of alternative biomarkers for lung cancer, one of the acknowledged limitations of this study.