Deep learning algorithm for automatically measuring Cobb angle in patients with idiopathic scoliosis.

PURPOSE: The Cobb angle is a standard measurement to qualify and track the progression of scoliosis. However, the Cobb angle has high inter- and intra-observer variability. Consequently, its measurement varies with vertebrae and may even differ when the same vertebra is measured. Therefore, it is not constant and differs with measurements. This study aimed to develop a deep learning model that automatically measures the Cobb angle. The deep learning model for identifying vertebrae on spine radiographs was developed. METHODS: The dataset consisted of 297 images that were divided into two subsets for training and validation. Two hundred and twenty-seven images (76.4%) were used to train the model, while 70 images (23.6%) were used as the validation dataset. Absolut error between the measurements by the observer and developed deep learning model and intraclass correlation coefficient (ICC). RESULTS: The average absolute error between the measurements was 1.97° with a standard deviation of 1.57°. In addition, 95.9% of the angles had an absolute error of less than 5°. The ICC was calculated to assess the model's reliability further. The ICC was 0.981, indicating excellent reliability. CONCLUSIONS: The authors believe the model will be useful in clinical practice by relieving clinicians of the burden of having to manually compute the Cobb angle. Further studies are needed to enhance the accuracy and versatility of this deep learning model.

Erratum: Development of diagnostic molecular markers for marker-assisted breeding against bacterial wilt in tomato.

[This corrects the article DOI: 10.1270/jsbbs.20027.].

A practical evaluation of machine learning for classification of ultrasound images of ovarian development in channel catfish (Ictalurus punctatus).

Machine learning is a powerful tool to improve efficiency of industrial processes, but it has not yet been well utilized in aquacultural and hatchery applications. The goal of the present study was to evaluate the feasibility of using a broad array of machine learning approaches (testing of > 200 vectorization and model combinations, reporting on 20) to classify ultrasound images spanning annual ovarian development (i.e., from undeveloped to mature) of channel catfish (Ictalurus punctatus). The specific objectives were to: 1) establish dataset preprocessing to standardize image features; 2) develop and train image classification models with deep learning methods; 3) develop and train models with traditional machine learning methods; 4) compare performance of deep learning and traditional methods on two classification problems (2-class and 5-class), and 5) propose insights to deploy models in practical aquaculture applications for research and hatchery use. A total of 931 ultrasound images of catfish ovaries were used to train and evaluate models for a 2-class problem (as a 'yes' or 'no' answer) to support hormone-injection decisions for spawning management in hatcheries, and a 5-class problem for classifying gonadal development stages for research. By using feature extraction, cropping, dimension reduction, and histogram normalization, a preprocessing method was created to standardize images to develop traditional (i.e., vector input), and deep learning convolutional neural network (CNN) (i.e., image input) approaches. Traditional machine learning models with image classification achieved 100% median accuracy on the 2-class problem (with the models RN-50 and RN-152), and 96% median accuracy for the 5-class problem (with VGG-19 image vectorization). The deep learning approach for the 2-class problem had a median accuracy of > 98% for 15models. The 5-class deep learning models produced a steady increase in median accuracy with training net size, achievable through expansion of the dataset. These models can be developed further, but traditional models (using CNN architectures to simply calculate image vectors) outperformed the deep learning approach. These models can be directly applicable to aquaculture in hatcheries and reproductive biology research, in addition to a wide variety of other image-based applications.

A Fabricated Force Glove That Measures Hand Forces during Activities of Daily Living.

Understanding hand and wrist forces during activities of daily living (ADLs) are pertinent when modeling prosthetics/orthotics, preventing workplace-related injuries, and understanding movement patterns that make athletes, dancers, and musicians elite. The small size of the wrist, fingers, and numerous joints creates obstacles in accurately measuring these forces. In this study, 14 FlexiForce sensors were sewn into a glove in an attempt to capture forces applied by the fingers. Participants in this study wore the glove and performed grasp and key turn activities. The maximal forces produced in the study were 9 N at the distal middle finger phalanx and 24 N at the distal thumb phalanx, respectively, for the grasp and key turn activities. Results from this study will help in determining the minimal forces of the hand during ADLs so that appropriate actuators may be placed at the appropriate joints in exoskeletons, orthotics, and prosthetics.

An Impedimetric Sensing Probe Based on Printed Circuit Board Technology for Monitoring in Cryobiology Applications.

Cryopreservation of living cells is an effective tool for protection, maintenance, and distribution of genetic resources, which involves exposure to cryogenic temperatures and requires precise control over various parameters to avoid potential cell damages. Hundreds of protocols have been reported for cryopreservation of aquatic species, but replicating them is challenging without a reliable monitoring technique during a cryopreservation process. In this work, we aim to use electrical impedance as a monitoring parameter to assist standardization of cryopreservation processes and reporting. Specifically, this paper reports an impedance sensing probe compatible with cryogenic temperatures and conventional containers in cryopreservation of aquatic species based on printed circuit board technology its characterization in cryopreservation conditions including different sperm extenders (buffer) compositions and concentrations, presence of cryoprotectant, and multiple cooling rates. The developed probe based on printed circuit board (PCB) technology shows a capability of measuring conditions during cryopreservation differentiating among samples with different buffer contents and cryoprotectants. The probe also demonstrates the capability to distinguish different cooling regimes and detect phase change phenomena. This PCB-based sensing platform provides quantitative impedance measurement data during the cryopreservation process at sample preparation, cooling, and while frozen. Technology such as this offers opportunities for improving the reproducibility of protocols generated by the aquatic species community and can be made widely available as open hardware.

Combined the SMAC mimetic and BCL2 inhibitor sensitizes neoadjuvant chemotherapy by targeting necrosome complexes in tyrosine aminoacyl-tRNA synthase-positive breast cancer.

BACKGROUND: Chemotherapy is the standard treatment for breast cancer; however, the response to chemotherapy is disappointingly low. Here, we investigated the alternative therapeutic efficacy of novel combination treatment with necroptosis-inducing small molecules to overcome chemotherapeutic resistance in tyrosine aminoacyl-tRNA synthetase (YARS)-positive breast cancer. METHODS: Pre-chemotherapeutic needle biopsy of 143 invasive ductal carcinomas undergoing the same chemotherapeutic regimen was subjected to proteomic analysis. Four different machine learning algorithms were employed to determine signature protein combinations. Immunoreactive markers were selected using three common candidate proteins from the machine-learning algorithms and verified by immunohistochemistry using 123 cases of independent needle biopsy FFPE samples. The regulation of chemotherapeutic response and necroptotic cell death was assessed using lentiviral YARS overexpression and depletion 3D spheroid formation assay, viability assays, LDH release assay, flow cytometry analysis, and transmission electron microscopy. The ROS-induced metabolic dysregulation and phosphorylation of necrosome complex by YARS were assessed using oxygen consumption rate analysis, flow cytometry analysis, and 3D cell viability assay. The therapeutic roles of SMAC mimetics (LCL161) and a pan-BCL2 inhibitor (ABT-263) were determined by 3D cell viability assay and flow cytometry analysis. Additional biologic process and protein-protein interaction pathway analysis were performed using Gene Ontology annotation and Cytoscape databases. RESULTS: YARS was selected as a potential biomarker by proteomics-based machine-learning algorithms and was exclusively associated with good response to chemotherapy by subsequent immunohistochemical validation. In 3D spheroid models of breast cancer cell lines, YARS overexpression significantly improved chemotherapy response via phosphorylation of the necrosome complex. YARS-induced necroptosis sequentially mediated mitochondrial dysfunction through the overproduction of ROS in breast cancer cell lines. Combination treatment with necroptosis-inducing small molecules, including a SMAC mimetic (LCL161) and a pan-BCL2 inhibitor (ABT-263), showed therapeutic efficacy in YARS-overexpressing breast cancer cells. CONCLUSIONS: Our results indicate that, before chemotherapy, an initial screening of YARS protein expression should be performed, and YARS-positive breast cancer patients might consider the combined treatment with LCL161 and ABT-263; this could be a novel stepwise clinical approach to apply new targeted therapy in breast cancer patients in the future.

Peripheral blood transcriptome analysis and development of classification model for diagnosing antibody-mediated rejection vs accommodation in ABO-incompatible kidney transplant.

The major obstacle to successful ABO blood group-incompatible kidney transplantation (ABOi KT) is antibody-mediated rejection (AMR). This study aimed to investigate transcriptional profiles through RNA sequencing and develop a minimally invasive diagnostic tool for discrimination between accommodation and early acute AMR in ABOi KT. Twenty-eight ABOi KT patients were selected: 18 with accommodation and 10 with acute AMR at the 10th day posttransplant protocol biopsy. Complete transcriptomes of their peripheral blood were analyzed by RNA sequencing. Candidate genes were selected by bioinformatics analysis, validated with quantitative polymerase chain reaction, and used to develop a classification model to diagnose accommodation. A total of 1385 genes were differentially expressed in accommodation compared with in AMR with P-adjusted < .05. Functional annotation and gene set enrichment analysis identified several immune-related and immunometabolic pathways. A 5-gene classification model including COX7A2L, CD69, CD14, CFD, and FOXJ3 was developed by logistic regression analysis. The model was further validated with an independent cohort and discriminated between accommodation and AMR with 92.7% sensitivity, 85.7% specificity, and 91.7% accuracy. Our study suggests that a classification model based on peripheral blood transcriptomics may allow minimally invasive diagnosis of acute AMR vs accommodation and subsequent patient-tailored immunosuppression in ABOi KT.

Cardiac Manifestations of Coronavirus Disease 2019 (COVID-19): a Multicenter Cohort Study.

BACKGROUND: This study aimed to investigate the cardiac manifestations of coronavirus disease 2019 (COVID-19). METHODS: From February to March 2020, we prospectively and retrospectively enrolled consecutive patients diagnosed with COVID-19. Patient's data such as the demographic characteristics, symptoms, vital signs, laboratory and radiologic findings, electrocardiographic, and echocardiographic data, including the global longitudinal strain (GLS) of both ventricles, were obtained. RESULTS: Forty patients (median age, 58 years; 50% men) were enrolled in the initial analysis. Patients were classified into severe and nonsevere groups based on the current guidelines. The 13 patients in the severe group were significantly older, had a greater prevalence of bilateral pneumonia and leukocytosis, and higher aspartate transaminase levels than patients in the nonsevere group. Patients in the severe group had a slightly lower left ventricular ejection fraction (LVEF) than those in the nonsevere group (median [interquartile range], 61.0% [58.5%, 62.3%] vs. 66.7% [60.6%, 69.8%], P = 0.015). In a subgroup of 34 patients in whom GLS could be analyzed, patients in the severe group had a significantly impaired left ventricular GLS (LVGLS) than those in the nonsevere group (-18.1% [-18.8%, -17.1%] vs. -21.7% [-22.9%, -19.9%], P = 0.001). There were no significant differences in total wall (RVGLStotal, -19.3% [-23.9%, -18.4%] vs. -24.3% [-26.0%, -22.6%], P = 0.060) and free wall (RVGLSfw, -22.7% [-27.2%, -18.6%] vs. -28.8% [-30.4%, -24.1%], P = 0.066) right ventricle GLS (RVGLS). CONCLUSION: Patients with severe COVID-19 had lower LVEF and LVGLS. RVGLS was not different between patients with severe and nonsevere COVID-19.

Development of diagnostic molecular markers for marker-assisted breeding against bacterial wilt in tomato.

Bacterial wilt, caused by the Ralstonia pseudosolanacearum species complex, is an important vascular disease that limits tomato production in tropical and subtropical regions. Two major quantitative trait loci (QTL) of bacterial wilt resistance on chromosome 6 (Bwr-6) and 12 (Bwr-12) were previously identified in Solanum lycopersicum 'Hawaii 7996'; however, marker-assisted breeding for bacterial wilt resistance is not well established. To dissect the QTL, six cleaved amplified polymorphic sites (CAPS) and derived CAPS (dCAPS) markers within the Bwr-6 region and one dCAPS marker near Bwr-12 were developed, and resistance levels in 117 tomato cultivars were evaluated. Two markers, RsR6-5 on chromosome 6 and RsR12-1 on chromosome 12, were selected based on the genotypic and phenotypic analysis. The combination of RsR6-5 and RsR12-1 effectively distinguishes resistant and susceptible cultivars. Furthermore, the efficiency of the two markers was validated in the F3 generation derived from the F2 population between E6203 (susceptible) and Hawaii 7998 (resistant). Resistant alleles at both loci led to the resistance to bacterial wilt. These markers will facilitate marker-assisted breeding of tomato resistant to bacterial wilt.

Three-dimensional printing can provide customizable probes for sensing and monitoring in cryobiology applications.

Cryopreservation has been recognized as a powerful tool for long-term preservation of genetic resources. However, the outcomes of cryopreservation by different user groups often vary due to inconsistency in procedures and freezing equipment. Herein, we report on the feasibility of providing customizable sensing probes with three-dimensional (3-D) printing to monitor cryopreservation phenomena. The objectives were to: 1) introduce 3-D printing as a fabrication method for developing customizable probes to be used in cryogenic applications; 2) design and fabricate an example of a 3-D printed sensing probe and multiplexer capable of detecting phase-change phenomena based on quantitative data regarding sample electrical resistance and temperature, and 3) demonstrate the sensing platform in cryopreservation conditions and in combination with a custom-made 3-D printed freezing device. The sensing probe developed was designed to fit within standard 0.5-ml French straws. Phase-transition phenomena were detected by analyzing electrical resistance changes. The quantitative data from this device in conjugation with a 3-D printed freezer rack provided cryopreservation capability with high reproducibility and offered an alternative to expensive programmable freezers. The use of 3-D printing provided flexibility to develop new sensing probes or modify existing designs based on specific needs. After initial prototyping, fabrication, and testing of 3-D printed sensing probes, particularly useful designs can lead to the reduction of variation in performing standardized cryopreservation protocols.

A microfluidic device for motility and osmolality analysis of zebrafish sperm.

A microfluidic chip is described that facilitates research and quality control analysis of zebrafish sperm which, due to its miniscule (i.e., 2-5 µl) sample volume and short duration of motility (i.e., <1 min), present a challenge for traditional manual assessment methods. A micromixer molded in polydimethylsiloxane (PDMS) bonded to a glass substrate was used to activate sperm samples by mixing with water, initiated by the user depressing a transfer pipette connected to the chip. Sample flow in the microfluidic viewing chamber was able to be halted within 1 s, allowing for rapid analysis of the sample using established computer-assisted sperm analysis (CASA) methods. Zebrafish sperm cell activation was consistent with manual hand mixing and yielded higher values of motility at earlier time points, as well as more subtle time-dependent trends in motility, than those processed by hand. Sperm activation curves, which indicate sample quality by evaluating percentage and duration of motility at various solution osmolalities, were generated with on-chip microfabricated gold floor electrodes interrogated by impedance spectroscopy. The magnitude of admittance was linearly proportional to osmolality and was not affected by the presence of sperm cells in the vicinity of the electrodes. This device represents a pivotal step in streamlining methods for consistent, rapid assessment of sperm quality for aquatic species. The capability to rapidly activate sperm and consistently measure motility with CASA using the microfluidic device described herein will help improve the reproducibility of studies on sperm and assist development of germplasm repositories.

Hollow microcarriers for large-scale expansion of anchorage-dependent cells in a stirred bioreactor.

With recent advances in biotechnology, mammalian cells are used in biopharmaceutical industries to produce valuable protein therapeutics and investigated as effective therapeutic agents to permanently degenerative diseases in cell based therapy. In these exciting and actively expanding fields, a reliable, efficient, and affordable platform to culture mammalian cells on a large scale is one of the most vital necessities. To produce and maintain a very large population of anchorage-dependent cells, a microcarrier-based stirred tank bioreactor is commonly used. In this approach, the cells are exposed to harmful hydrodynamic shear stress in the bioreactor and the mass transfer rates of nutrients and gases in the bioreactor are often kept below an optimal level to prevent cellular damages from the shear stress. In this paper, a hollow microcarrier (HMC) is presented as a novel solution to protect cells from shear stress in stirred bioreactors, while ensuring sufficient and uniform mass transfer rate of gases and nutrients. HMC is a hollow microsphere and cells are cultured on its inner surface to be protected, while openings on the HMC provide sufficient exchange of media inside the HMC. As a proof of concept, we demonstrated the expansion of fibroblasts, NIH/3T3 and the expansion and cardiac differentiation of human induced pluripotent stem cells, along with detailed numerical analysis. We believe that the developed HMC can be a practical solution to enable large-scale expansion of shear-sensitive anchorage-dependent cells in an industrial scale with stirred bioreactors.

Robust Directional Angle Estimation of Underwater Acoustic Sources Using a Marine Vehicle.

Acoustic source localization is used in many underwater applications. Acquiring an accurate directional angle for an acoustic source is crucial for source localization. To achieve this purpose, this paper presents a method for directional angle estimation of underwater acoustic sources using a marine vehicle. It is assumed that the vehicle is equipped with two hydrophones and that the acoustic source transmits a specific signal repeatedly. The proposed method provides a probabilistic model for time delay estimation. The probability is recursively updated by prediction and update steps. The prediction step performs a probability transition using the angular displacement of the marine vehicle. The predicted probability is updated using a generalized cross correlation function with a verification process using entropy measurement. The proposed method can provide a reliable and accurate estimation of the directional angles of underwater acoustic sources. Experimental results demonstrate good performance of the proposed probabilistic directional angle estimation method in both an inland water environment and a harbor environment.

Variation in intraocular pressure caused by repetitive positional changes during laparoscopic colorectal surgery: a prospective, randomized, controlled study comparing propofol and desflurane anesthesia.

Laparoscopic surgery is often prolonged and requires positional changes to facilitate surgical access. Previous studies reported intraocular pressure (IOP) changes in one fixed position. This study investigated the effect of desflurane and propofol anesthesia on IOP during repeated positional changes. A total of 46 patients undergoing laparoscopic colorectal surgery were randomized into desflurane or propofol groups. IOP was measured using a handheld tonometer at seven time points: before induction (baseline), after endotracheal intubation, after pneumoperitoneum, after the first Trendelenburg and right tilt position, after reverse Trendelenburg and right tilt position, after the second Trendelenburg and right tilt position and before endotracheal extubation. Trendelenburg positioning raised IOP in both groups. In the desflurane group, IOP at the first Trendelenburg position was 6.27 and 8.48 mmHg higher than baseline IOP in left and right eye, respectively; IOP at the second Trendelenburg position was 7 and 9.44 mmHg higher than baseline in left and right eye, respectively. In the propofol group, IOP at the first Trendelenburg position was 2.04 and 4.04 mmHg higher than baseline in left and right eyes, respectively. It was 3.04 and 4.87 mmHg higher than baseline in left and right eye, respectively, at the second Trendelenburg position. In the desflurane group, 56.52% patients exhibited high IOP (≥ 25 mmHg) compared with 13.04% in the propofol group at the second Trendelenburg position in the right eyes (P = 0.005). There was a positive correlation between IOP and peak inspiratory pressure (P < 0.001). Propofol anesthesia mitigated wide variations in IOP caused by repetitive positional changes during laparoscopic colorectal surgery.

Probability-Based Recognition Framework for Underwater Landmarks Using Sonar Images †.

This paper proposes a probability-based framework for recognizing underwater landmarks using sonar images. Current recognition methods use a single image, which does not provide reliable results because of weaknesses of the sonar image such as unstable acoustic source, many speckle noises, low resolution images, single channel image, and so on. However, using consecutive sonar images, if the status-i.e., the existence and identity (or name)-of an object is continuously evaluated by a stochastic method, the result of the recognition method is available for calculating the uncertainty, and it is more suitable for various applications. Our proposed framework consists of three steps: (1) candidate selection, (2) continuity evaluation, and (3) Bayesian feature estimation. Two probability methods-particle filtering and Bayesian feature estimation-are used to repeatedly estimate the continuity and feature of objects in consecutive images. Thus, the status of the object is repeatedly predicted and updated by a stochastic method. Furthermore, we develop an artificial landmark to increase detectability by an imaging sonar, which we apply to the characteristics of acoustic waves, such as instability and reflection depending on the roughness of the reflector surface. The proposed method is verified by conducting basin experiments, and the results are presented.

The Association of -429T>C and -374T>A Polymorphisms in the RAGE Gene with Polycystic Ovary Syndrome.

Polycystic ovary syndrome (PCOS) is a complex disorder characterized by hyperandrogenism and insulin resistance. In addition, a number of females with PCOS have ovaries with multiple cysts, an irregular or no menstrual cycle, and an imbalance of female hormones compared to those of normal controls. A variety of genetic factors have been involved in the pathogenesis of PCOS. Among these genetic factors, the receptor for advanced glycation end products (RAGE) that is associated with diabetes and involved in the complications of PCOS, was selected. We aimed to assess the relationship between -429T>C and -374T>A single nucleotide polymorphisms (SNPs) of RAGE gene with the susceptibility to PCOS.128 controls and 265 PCOS patients were used for -374T>A polymorphism and 141 controls and 290 PCOS patients were used for -429T>C polymorphism, respectively. Genotyping of two polymorphisms were analyzed by polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) assay and statistical analysis was performed. P values for both alleles were higher than 0.05. Frequencies of genotype and allele of two polymorphisms in RAGE gene showed no significant differences between controls and PCOS patients. The initial study on the correlation between RAGE gene and PCOS indicates that the two polymorphisms of RAGE are not associated with the pathogenesis of PCOS. However, further studies regarding the association between RAGE gene and PCOS patients in different ethnic groups are required.

Experience from a single-center study on multimodal medication therapy for patients with complex regional pain syndrome.

BACKGROUND: Despite the application of various therapeutic methods, pain caused by complex regional pain syndrome (CRPS) is not sufficiently managed and often progresses to a chronic stage. For the systematic and effective treatment of CRPS, we developed an algorithm for multimodal medication therapy based on the established pathophysiology of CRPS to control CRPS-related pain. OBJECTIVE: In this study, we present the outcomes of our novel algorithm for multimodal medication therapy for patients with CRPS, consisting of three major components: multimodal oral medication, intravenous ketamine, and intravenous lidocaine therapy. METHODS: We retrospectively investigated patients with CRPS who received multimodal therapy. Pain severity scores were evaluated using a numerical rating scale at four time points (P1, pain at initial consultation; P2, pain after oral medication; P3, pain after ketamine treatment; and P4, pain after lidocaine treatment). The effect of the multimodal medication therapy algorithm on pain management was evaluated at each time point. RESULTS: In patients with CRPS, multimodal oral medication, intravenous ketamine, and intravenous lidocaine therapies led to significantly improved pain control (p< 0.05). Additionally, the combination of these three therapies (through the multimodal medication therapy algorithm) resulted in significant pain relief in patients with CRPS (p< 0.05). CONCLUSIONS: Our multimodal medication therapy algorithm effectively controlled pain in patients with CRPS. However, further prospective studies with large sample sizes and randomized controlled trials are needed for more accurate generalization.

Visualization system to identify structurally vulnerable links in OHT railway network in semiconductor FAB using betweenness centrality.

In semiconductor fabrication (FAB), wafers are placed into carriers known as Front Opening Unified Pods (FOUPs), transported by the Overhead Hoist Transport (OHT). The OHT, a type of Automated Guided Vehicle (AGV), moves along a fixed railway network in the FAB. The routes of OHTs on the railway network are typically determined by a Single Source Shortest Path (SSSP) algorithm such as Dijkstra's. However, the presence of hundreds of operating OHTs often leads to path interruptions, causing congestion or deadlocks that ultimately diminish the overall productivity of the FAB. This research focused on identifying structurally vulnerable links within the OHT railway network in semiconductor FAB and developing a visualization system for enhanced on-site decision-making. We employed betweenness centrality as a quantitative index to evaluate the structural vulnerability of the OHT railway network. Also, to accommodate the unique hierarchical node-port structure of this network, we modified the traditional Brandes algorithm, a widely-used method for calculating betweenness centrality. Our modification of the Brandes algorithm integrated node-port characteristics without increasing computation time while incorporating parallelization to reduce computation time further and improve usability. Ultimately, we developed an end-to-end web-based visualization system that enables users to perform betweenness centrality calculations on specific OHT railway layouts using our algorithm and view the results through a web interface. We validated our approach by comparing our results with historically vulnerable links provided by Samsung Electronics. The study had two main outcomes: the development of a new betweenness centrality calculation algorithm considering the node-port structure and the creation of a visualization system. The study demonstrated that the node-port structure betweenness centrality effectively identified vulnerable links in the OHT railway network. Presenting these findings through a visualization system greatly enhanced their practical applicability and relevance.

Video domain adaptation for semantic segmentation using perceptual consistency matching.

Unsupervised domain adaptation (UDA) aims to transfer knowledge in previous and related labeled datasets (sources) to a new unlabeled dataset (target). Despite the impressive performance, existing approaches have largely focused on image-based UDA only, and video-based UDA has been relatively understudied and received less attention due to the difficulty of adapting diverse modal video features and modeling temporal associations efficiently. To address this, existing studies use optical flow to capture motion cues between in-domain consecutive frames, but is limited by heavy compute requirements and modeling flow patterns across diverse domains is equally challenging. In this work, we propose an adversarial domain adaptation approach for video semantic segmentation that aims to align temporally associated pixels in successive source and target domain frames without relying on optical flow. Specifically, we introduce a Perceptual Consistency Matching (PCM) strategy that leverages perceptual similarity to identify pixels with high correlation across consecutive frames, and infer that such pixels should correspond to the same class. Therefore, we can enhance prediction accuracy for video-UDA by enforcing consistency not only between in-domain frames, but across domains using PCM objectives during model training. Extensive experiments on public datasets show the benefit of our approach over existing state-of-the-art UDA methods. Our approach not only addresses a crucial task in video domain adaptation but also offers notable improvements in performance with faster inference times.

Evaluation of clinical factors predicting dysphagia in patients with traumatic and non-traumatic cervical spinal cord injury: a retrospective study.

Introduction: Dysphagia is a common complication in patients with cervical spinal cord injury (C-SCI) and can cause various pulmonary complications, such as aspiration pneumonia and mechanical airway obstruction increasing mortality and morbidity. This study evaluated the clinical factors that predict dysphagia in patients with traumatic and non-traumatic C-SCI. Methods: Ninety-eight patients with C-SCI were retrospectively enrolled in this study and were divided into those with and without dysphagia. Clinical factors such as age, sex, tracheostomy, spinal cord independence measure, pulmonary function test (PFT) including forced vital capacity (FVC), forced expiratory volume in 1 s (FEV1) and FVC/FEV1, American Spinal Cord Injury Association score, Berg Balance Scale, and surgical approach were investigated retrospectively. Results: Multivariate logistic regression analysis revealed that FVC and the presence of tracheostomy were significantly correlated with dysphagia in patients with C-SCI (p < 0.05). FVC and the presence of tracheostomy are useful tools for detecting dysphagia in patients with C-SCI. Conclusion: Considering the results of our study, early PFTs, especially FVC, in patients with C-SCI and early initiation of dysphagia management and treatment in patients with C-SCI and tracheostomy will be advantageous in lowering the mortality and morbidity due to pulmonary aspiration in these patients.