Search

Copyright of Contemporary Accounting Research is the property of Canadian Academic Accounting Association and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Visible-infrared person re-identification aims to match the identity of the same person from different modalities. The main challenge is the modality difference between visible and infrared images. Most existing methods mainly use generative adversarial networks to generate compensatory images of the corresponding modality to reduce the modality difference, or design diverse two-stream networks to learn global feature representations and extract globally shared features. However, due to the substantial difference between visible and infrared modalities, the created pseudo-modalities often struggle to effectively bridge the gap between modalities and tend to introduce noise. The extracted modality-shared features typically exhibit weak discriminative capability, inevitably leading to the loss of critical discriminative features related to person identity and a lack of robustness to noisy images. To tackle these challenges, we introduce a modality synergy alignment learning network. This network incorporates a novel data augmentation technique known as SliceMix, which mixes random sections of cross-modality images to synthesize a new sample that exhibits both discriminative to identity and robust to noise, thereby facilitating the learning of modality-invariant feature representations. By adjusting the mixing ratio, mixed modalities can be generated flexibly to minimize the impact of modality imbalance. Additionally, a modality alignment module is introduced to ensure similarity within the modality class and accentuate the differences between modalities. Moreover, we propose a data augmentation method called random channel grayscale, which enhances the network’s robustness to color changes and expands data diversity. Comprehensive experiments on mainstream datasets, including SYSU-MM01 and RegDB, demonstrated that our method significantly improves the performance of cross-modality retrieval. [ABSTRACT FROM AUTHOR]

This paper presents an enhanced approach for classifying Bengali songs into different genres by leveraging feature importance analysis and deep learning techniques. The research addresses the challenge of limited data points in the Bengali Song Dataset by employing strategies, including audio segmentation and feature importance analysis, to enhance model performance. Multiple machine learning and deep learning architectures are evaluated to identify the most effective models for Bengali song classification. Additionally, this research conducts feature importance analysis to identify significant audio features contributing to classification accuracy. The best-performing deep learning model achieves an impressive validation accuracy of 94.17%, showcasing the project efficacy of the proposed methodology. Our findings highlight the effectiveness of our proposed methodology, demonstrating significant improvements in classification accuracy and contributing to advancements in Bengali music classification research. [ABSTRACT FROM AUTHOR]

Lumpy skin disease virus (LSDV) is an extremely infectious, viral, and chronic skin disease that is caused by the Capripox virus. This viral disease is predominantly found in cows. Mosquitoes and ticks are the primary transmitters for the spread of this virus. Recently, LSDV has been rapidly spreading all over the world, especially in several areas of Pakistan, India, and Iran. Thousands of cows have died due to this infectious virus in Pakistan and early detection of LSDV is needed to avoid further loss. The prediction and classification of LSDV are hindered by the lack of publicly available datasets. Despite a few studies using LSDV datasets, such datasets are often small, which may lead to model overfitting. In this regard, we collect the dataset from several online sources, as well as, collecting images from veterinary farms in different areas of Pakistan. Deep learning has been widely used in the medical field for disease detection and classification. Therefore, this study leverages DenseNet deep learning models for LSDV detection and classification. Experiments are performed using VGG-16, ResNet-50, MobileNet-V2, custom-designed convolutional neural network, and Inception-V3. The DenseNet architecture presents a Convolutional Block Attention Module (CBAM) and Spatial Attention (SA) for the prediction and classification of LSD. Results demonstrate that a 99.11% accuracy can be obtained on the augmented dataset while a 94.23% accuracy can be achieved with the original dataset for chicken pox, monkey pox, and LSDV. Comparison with state-of-the-art studies corroborates the superior performance of the proposed model. [ABSTRACT FROM AUTHOR]

Spasticity management should be provided within the context of a comprehensive person-centered rehabilitation program. Furthermore, active goal setting for specific spasticity interventions is also important, with a well-established "more is better" approach. It is critical to consider adjunctive therapy and multimodal approaches if patients are not attaining their treatment goals. Often used interchangeably, there may be confusion between the terms adjunctive and multimodal therapy. Yet it is imperative to understand the differences between these approaches to achieve treatment goals in spasticity management. Addition of a secondary pharmacologic or non-pharmacologic treatment to optimize the efficacy of the initial modality, such as adding electrical stimulation or casting to BoNT-A, is considered an adjunctive therapy. Adjunctive therapy is time-specific and requires the added therapy be initiated within a specific period to enhance the primary treatment; usually within 2 weeks. Multimodal therapy is an integrated, patient-centric program of pharmacologic and nonpharmacologic strategies utilized in a concurrent/integrated or sequential manner to enhance the overall treatment effect across a variety of spasticityassociated impairments (e.g., neural and non-neural components). Moreover, within a multimodal approach, adjunctive therapy can be used to help enhance the treatment effect of one specific modality. The objectives of this paper are to clarify the differences between adjunctive and multimodal therapies, provide a brief evidence-based review of such approaches, and highlight clinical insights on selecting multimodal and adjunctive therapies in spasticity management. [ABSTRACT FROM AUTHOR]

Liver cancer is one of the leading causes of cancer death in the world, and early diagnosis is important. However, the similarity in shape, texture, and intensity values between the liver, tumors, and other neighboring organs such as the heart, spleen, stomach, and kidneys often complicates visual differentiation. Manual identification of tumors in the liver is time‐consuming, intricate, and susceptible to errors with potential repercussions for patient care. While machine learning–based approaches have emerged for liver organ recognition and segmenting the tumor, they continue to face challenges related to recognition accuracy and the inability to distinguish tumors of varied sizes. To solve the problems, a multiattention network made up of cascaded ResUNet and U‐Net with attention mechanisms was proposed in this study. We investigated liver tumor segmentation with various configurations of U‐Net, ResUNet, U‐Net with attention mechanisms, and ResUNet with attention mechanisms on augmented and nonaugmented data. We used the 3Dircadb dataset for training and validation purposes, and the proposed method was evaluated on dice score, intersection of union (IoU), recall, and precision. The performance metrics achieved with this method on the dataset are as follows: approximately 0.89 for the dice coefficient, 0.90 for IoU, 0.93 for recall, and 0.96 for precision in the case of liver segmentation without data augmentation and 0.92, 0.90, 0.92, and 0.94, respectively, for dice score, IoU, recall, and precision with data augmentation. For tumor segmentation, the metrics include 0.70 for dice coefficient, 0.61 for IoU, 0.91 for recall, and 0.94 for precision when the data were augmented but 0.83 for dice score, 0.78 for IoU, and 0.89 and 0.90, respectively, for recall and precision. [ABSTRACT FROM AUTHOR]

Significant progress has been made recently with the contribution of technological advances in studies on brain cancer. Regarding this, identifying and correctly classifying tumors is a crucial task in the field of medical imaging. The disease-related tumor classification problem, on which deep learning technologies have also become a focus, is very important in the diagnosis and treatment of the disease. The use of deep learning models has shown promising results in recent years. However, the sparsity of ground truth data in medical imaging or inconsistent data sources poses a significant challenge for training these models. The utilization of StyleGANv2-ADA is proposed in this paper for augmenting brain MRI slices to enhance the performance of deep learning models. Specifically, augmentation is applied solely to the training data to prevent any potential leakage. The StyleGanv2-ADA model is trained with the Gazi Brains 2020, BRaTS 2021, and Br35h datasets using the researchers' default settings. The effectiveness of the proposed method is demonstrated on datasets for brain tumor classification, resulting in a notable improvement in the overall accuracy of the model for brain tumor classification on all the Gazi Brains 2020, BraTS 2021, and Br35h datasets. Importantly, the utilization of StyleGANv2-ADA on the Gazi Brains 2020 Dataset represents a novel experiment in the literature. The results show that the augmentation with StyleGAN can help overcome the challenges of working with medical data and the sparsity of ground truth data. Data augmentation employing the StyleGANv2-ADA GAN model yielded the highest overall accuracy for brain tumor classification on the BraTS 2021 and Gazi Brains 2020 datasets, together with the BR35H dataset, achieving 75.18%, 99.36%, and 98.99% on the EfficientNetV2S models, respectively. This study emphasizes the potency of GANs for augmenting medical imaging datasets, particularly in brain tumor classification, showcasing a notable increase in overall accuracy through the integration of synthetic GAN data on the used datasets. [ABSTRACT FROM AUTHOR]

Copyright of Arthroskopie is the property of Springer Nature and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Background: Despite high complication rates, patients persistently present for single-stage augmentation mastopexy. In empty, deflated breasts, we perform one-stage augmentation mastopexy; however, in heavy ptotic breasts, our preference is to stage the procedure with mastopexy and fat graft first. With volume from fat grafting focussing on the upper pole and cleavage areas, many of our patients avoid implants altogether. This reduces subsequent risks of waterfall deformity, implant displacement, rupture and a lifetime of implant exchanges. Objectives: We aim to describe our findings and technique for reducing progression to the second stage of a two-stage augmentation mastopexy with the appropriate use of moderate to high volume of fat grafting at the primary operation. Methods: This is a retrospective review of all patients who presented to the senior author (KT) requesting breast implants and requiring mastopexy, from January 2018 to December 2022. Results: Over the five-year period, 137 patients were identified. Seventy-one (51.8%) underwent single-stage augmentation mastopexy, 55 (40.1%) underwent mastopexy with fat grafting and 11 (8.0%) underwent mastopexy with no fat grafting. Our key finding in this study is that 52 of 66 (78.8%) of planned staged patients, who underwent mastopexy with or without fat grafting, were happy with the volume attained and no longer wished to undergo further implant augmentation. Conclusion: In selected patients, appropriate volume and position of fat grafting at the time of primary mastopexy can significantly obviate the need for a second stage implant (alloplastic) augmentation. Level of Evidence V: This journal requires that authors assign a level of evidence to each article. For a full description of these Evidence-Based Medicine ratings, please refer to the Table of Contents or the online Instructions to Authors www.springer.com/00266. [ABSTRACT FROM AUTHOR]

Data augmentation is widely recognized as an effective means of bolstering model robustness. However, when applied to monocular 3D object detection, non-geometric image augmentation neglects the critical link between the image and physical space, resulting in the semantic collapse of the extended scene. To address this issue, we propose two geometric-level data augmentation operators named Geometric-Copy-Paste (Geo-CP) and Geometric-Crop-Shrink (Geo-CS). Both operators introduce geometric consistency based on the principle of perspective projection, complementing the options available for data augmentation in monocular 3D. Specifically, Geo-CP replicates local patches by reordering object depths to mitigate perspective occlusion conflicts, and Geo-CS re-crops local patches for simultaneous scaling of distance and scale to unify appearance and annotation. These operations ameliorate the problem of class imbalance in the monocular paradigm by increasing the quantity and distribution of geometrically consistent samples. Experiments demonstrate that our geometric-level augmentation operators effectively improve robustness and performance in the KITTI and Waymo monocular 3D detection benchmarks. [ABSTRACT FROM AUTHOR]

Introduction: Restless legs syndrome (RLS) is a sensory-motor sleep disorder that affects up to 13% of adults in the Western world and 2–4% of children. It impairs night sleep with an impact on daily performances and life quality. Thus, moderate-to-severe RLS requires pharmacological treatment. Areas covered: In the present review, which is based on PubMed searches with no time limits, the authors discuss the recommended pharmacotherapy for RLS in addition to other emerging treatment options. The authors provide coverage to the current recommendations for both adults and pediatric patients with RLS. Expert opinion: Current evidence suggests removing all causes of secondary RLS, including iron deficiency, chronic renal failure, drugs, and treating other sleep disorders that may worsen symptoms. Also, intermittent RLS should be addressed with behavioral measures and on-demand therapy. For chronic persistent RLS, α2δ calcium channel ligands are a first-line pharmacological approach, whereas dopamine agonists are associated with increased risk and should be spared. When RLS is refractory to first-line treatment, polytherapy, or opioid monotherapy should be considered. Nonetheless, some patients may not reach sustained symptom relief. Further research is needed to better understand the pathophysiology of RLS and to develop newer more effective drugs. [ABSTRACT FROM AUTHOR]

We estimate relative hazards and absolute risks (or cumulative incidence or crude risk) under cause-specific proportional hazards models for competing risks from double nested case-control (DNCC) data. In the DNCC design, controls are time-matched not only to cases from the cause of primary interest, but also to cases from competing risks (the phase-two sample). Complete covariate data are available in the phase-two sample, but other cohort members only have information on survival outcomes and some covariates. Design-weighted estimators use inverse sampling probabilities computed from Samuelsen-type calculations for DNCC. To take advantage of additional information available on all cohort members, we augment the estimating equations with a term that is unbiased for zero but improves the efficiency of estimates from the cause-specific proportional hazards model. We establish the asymptotic properties of the proposed estimators, including the estimator of absolute risk, and derive consistent variance estimators. We show that augmented design-weighted estimators are more efficient than design-weighted estimators. Through simulations, we show that the proposed asymptotic methods yield nominal operating characteristics in practical sample sizes. We illustrate the methods using prostate cancer mortality data from the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial Study of the National Cancer Institute. [ABSTRACT FROM AUTHOR]

The article considers the available methods of image recognition of continuous wavelet spectra of noisy signals with linear and nonlinear frequency modulations using convolutional neural networks. A procedure is proposed for preparing spectral images for their processing in a neural network, ensuring a sufficient probability of recognizing a given type of signal out of twenty possible ones. The methodology for solving the problem finds an image preparation algorithm that provides image augmentation by changing continuous wavelets, which ensures the identification of signals under conditions of limited resonance frequency and bandwidth. The algorithm changes the frequency of the continuous spectrum by processing the phase grating signal with different continuous wavelets after adding non-stationary noise. Signals with linear and nonlinear modulation prepared in this way, as well as signal spectra of other regular forms, are used as input data of the convolutional neural network. The procedure of dividing images of wavelet spectra into classes is performed by checking the homogeneity of the class based on the Shannon entropy value. The minimum entropy value indicates the homogeneity of the subset and the absence of "impurities" from images of other classes. The developed model of a neural network with augmentation by continuous wavelet spectra under the conditions of a limited data set has an accuracy of up to 97.95%. [ABSTRACT FROM AUTHOR]

Background: Machine learning models can provide quick and reliable assessments in place of medical practitioners. With over 50 million adults in the United States suffering from osteoarthritis, there is a need for models capable of interpreting musculoskeletal ultrasound images. However, machine learning requires lots of data, which poses significant challenges in medical imaging. Therefore, we explore two strategies for enriching a musculoskeletal ultrasound dataset independent of these limitations: traditional augmentation and diffusion-based image synthesis. Methods: First, we generate augmented and synthetic images to enrich our dataset. Then, we compare the images qualitatively and quantitatively, and evaluate their effectiveness in training a deep learning model for detecting thickened synovium and knee joint recess distension. Results: Our results suggest that synthetic images exhibit some anatomical fidelity, diversity, and help a model learn representations consistent with human opinion. In contrast, augmented images may impede model generalizability. Finally, a model trained on synthetically enriched data outperforms models trained on un-enriched and augmented datasets. Conclusions: We demonstrate that diffusion-based image synthesis is preferable to traditional augmentation. Our study underscores the importance of leveraging dataset enrichment strategies to address data scarcity in medical imaging and paves the way for the development of more advanced diagnostic tools. [ABSTRACT FROM AUTHOR]

Electrocardiograms (ECGs) provide essential data for diagnosing arrhythmias, which can potentially cause serious health complications. Early detection through continuous monitoring is crucial for timely intervention. The Massachusetts Institute of Technology-Beth Israel Hospital (MIT-BIH) arrhythmia dataset employed for arrhythmia analysis research comprises imbalanced data. It is necessary to create a robust model independent of data imbalances to classify arrhythmias accurately. To mitigate the pronounced class imbalance in the MIT-BIH arrhythmia dataset, this study employs advanced augmentation techniques, specifically variational autoencoder (VAE) and conditional diffusion, to augment the dataset. Furthermore, accurately segmenting the continuous heartbeat dataset into individual heartbeats is crucial for confidently detecting arrhythmias. This research compared a model that employed annotation-based segmentation, utilizing R-peak labels, and a model that utilized an automated segmentation method based on a deep learning model to segment heartbeats. In our experiments, the proposed model, utilizing MobileNetV2 along with annotation-based segmentation and conditional diffusion augmentation to address minority class, demonstrated a notable 1.23% improvement in the F1 score and 1.73% in the precision, compared to the model classifying arrhythmia classes with the original imbalanced dataset. This research presents a model that accurately classifies a wide range of arrhythmias, including minority classes, moving beyond the previously limited arrhythmia classification models. It can serve as a basis for better data utilization and model performance improvement in arrhythmia diagnosis and medical service research. These achievements enhance the applicability in the medical field and contribute to improving the quality of healthcare services by providing more sophisticated and reliable diagnostic tools. [ABSTRACT FROM AUTHOR]

The paper considers the existing methods of image recognition of continuous wavelet spectra of noisy signals with linear and nonlinear frequency modulation using convolutional neural networks. A procedure for preparing spectral images for processing in a neural network is proposed, which ensures a sufficient probability of recognizing a given type of signal out of twenty possible ones. The methodology for solving the problem consists of finding an image preparation algorithm that provides image augmentation by the method of changing continuous wavelets, which ensures the identification of signals under conditions of limited resonance frequency and bandwidth. The algorithm involves changing the frequency of the continuous spectrum by processing the phase grating signal with different continuous wavelets after the additive addition of non-stationary noise. Signals with linear and nonlinear modulation prepared in this way, as well as signal spectra of other regular forms, are used as input data of the convolutional neural network. The procedure of dividing wavelet spectrum images into classes is performed by checking the homogeneity of the class based on the Shannon entropy value. The minimum entropy value indicates the homogeneity of the subset and the absence of “impurities” from images of other classes. The developed model of a neural network with augmentation by continuous wavelet spectra in the conditions of a limited data set has an accuracy of up to 97.95%. [ABSTRACT FROM AUTHOR]

Global Navigation Satellite Systems (GNSS) has become an essential component in modern times for positioning, navigation, and timing. A fast-growing economic region in Malaysia required a GNSS-based augmentation positioning and navigation service. To improve navigation solution, several augmentation techniques exist, such as a differential Global Positioning System (DGPS) assisted by reference station, or augmentation service provided by commercial communication satellite. The DGPS correction is applied through a real-time communication medium and can be received at the user side by several communication methods including internet-based, and this method is favourable for land and near-coast areas. This case study aims to investigate the reliability of internet-based DGPS and SBAS along the Peninsular Malaysia. A test was conducted and data were collected over 15 hours at a rate of 1 Hz from the available GNSS satellite using a geodetic-grade receiver mounted on a moving vehicle. The obtained results showed that DGPS and SBAS perform better than the navigation solution with an accuracy of 1.536 m and 0.955 m respectively, compared to the navigation solution with an accuracy of 3.159 m. The limitations of both augmentation techniques were also analysed and discussed in this study. [ABSTRACT FROM AUTHOR]

Dermatology is the most appropriate field to utilize pattern recognition-based automated techniques for objective, accurate, and rapid diagnosis because diagnosis mainly relies on visual examinations of skin lesions. Recent approaches utilizing deep learning techniques have shown remarkable results in this field. However, they necessitate a substantial quantity of images and the availability of dermoscopy images is often limited. Also, even if enough images are available, their labeling requires expert knowledge and is time-consuming. To overcome these issues, an efficient augmentation approach is needed to expand training datasets from input images. Therefore, in this work, a generative adversarial network has been developed using a new hybrid loss function constructed with traditional loss functions to enhance the generation power of the architecture. Also, the effect of the proposed approach and different generative network-based augmentations, which have been used with dermoscopy images in the literature, on the classification of skin lesions has been investigated. Therefore, the main contributions of this work are: (i) introducing a new generative model for the augmentation of dermoscopy images; (ii) presenting the effect of the proposed model on the classification of the images; (iii) comparative evaluations of the effectiveness of different generative network-based augmentations in the classification of seven forms of skin lesions. The classification accuracy when the proposed augmentation is used is 93.12%, which is higher than its counterparts. Experimental results indicate the significance of augmentation techniques in the classification of skin lesions and the efficiency of the proposed structure in improving the classification accuracy. [ABSTRACT FROM AUTHOR]

This article seeks to reflect on the opportunities that mobile smartphones (MSPs) present as ICT integration tools in teaching geography. The more extensive study, underpinned by the Professional Development Framework for Digital Learning (PDFDL) in ICT integration, employed a qualitative research approach. Lensed by the Professional Development Framework for Digital Learning (PDFDL), the article used the qualitative approach to garner insights from the participants regarding using MSPs as tools to integrate ICT in geography teaching. Data collection tools included interviews, observations, and document reviews. Researchers sampled (n = 4) schools, interviewed and observed (n = 13) teachers, and interviewed (n = 10) learners and (n = 8) parents in the province of KwaZulu-Natal. Furthermore, they used a purposive sampling technique to access the participants, basing the research on the premise that MSPs promote virtual reality for an array of learners. As the findings revealed, although some participants viewed the use of MSPs as a distractor in the learning space, teachers felt compelled to heed the call to modify their teaching pedagogies, such that they integrated mobile phones fruitfully in their teaching. The findings further revealed that such a paradigm shift would benefit homeschooling and facilitate a dual teaching mode at learning institutions. Curriculum planners are responsible for helping teachers accept that uncertainty is the only certainty about the future, considering the volatility, uncertainty, complexity, and augmentation (VUCA) challenges brought on by the COVID-19 pandemic. Extended lockdown periods accelerated the use of MSPs in teaching, requiring every stakeholder in the educational space to become a life-long learner by using a range of technologies and platforms. [ABSTRACT FROM AUTHOR]

The aim of this paper is to investigate the impact of selected data augmentation techniques on the learning performance of neural networks for dynamic signature verification. The paper investigates selected data augmentation techniques in deep learning for verification purpose of dynamic signature. Two neural networks were used as classifiers: MLP and LSTM-FCN. Investigation of five selected augmentation methods and experiments were performed on the open source signature database SVC2004. The authors tested both classifiers without augmentation and then with data augmentation for three extensions of the learning set and three sizes of the user database. They presented the results of the experiments in tabular form for each augmentation method. The results were compared with the existing dynamic signature verification methods and given in the paper. [ABSTRACT FROM AUTHOR]

Background: The ulnar collateral ligament (UCL) is frequently injured in overhead throwing athletes, leading to elbow instability, pain and decreased performance, and requiring surgical intervention. Augmenting with suture tape provides a novel approach to UCL repair, offering stability while preserving native anatomy, proprioception, and minimizing bone loss, with the added benefit of an expected faster rehabilitation and return to sport (RTS) compared to traditional UCL reconstruction. The purpose of this study is to assess the variability of the current publicly available rehabilitation protocols for elbow UCL repair with suture tape augmentation. Methods: A systematic review of Google and PubMed was performed to find rehabilitation protocols for UCL repair with suture tape augmentation. Two hundred nine orthopedic surgery residency programs were identified using the Fellowship and Residency Electronic Interactive Database Access. The programs were searched on Google, a broad nonspecific Google search, and a systematic literature search of journal databases for UCL repair with suture tape augmentation rehabilitation protocols was conducted. Results: Thirteen rehabilitation protocols met the inclusion criteria for analysis. Of these, 12 protocols outlined the use of an articulating brace at varying range of motion settings for the first 4-6 weeks following surgery. Return to throwing was included in all 13 protocols and began between 10 and 12 weeks following surgery. RTS was included in 11 protocols and expected athletes competing as early as 16 weeks with a mean recommended RTS at 21.5 weeks. Conclusion: Rehabilitation protocols for UCL repair with suture tape augmentation were often structured around a 5-phase program with RTS approaching 20 weeks. They utilized immobilization and range of motion restriction as well as strengthening and gradual RTS procedures. Overall, the included protocols had mild variability with initiation of throwing and RTS 12 to 24 weeks faster than traditional UCL reconstruction.

Pathology of the long head of the biceps tendon is commonly associated with massive rotator cuff tears (MRCTs), which account for roughly one third of all rotator cuff tears. Treatment options for this condition include tenotomy, tenodesis, augmentation, and the use of the tendon as a graft for partial superior capsule reconstruction. Augmentation and superior capsular reconstruction are evolving techniques in the management of MRCTs. However, similar to the lack of consensus on the treatment of MRCTs, there are no clear guidelines for the management of concurrent biceps tendon pathology.

Abstract Lumpy skin disease virus (LSDV) is an extremely infectious, viral, and chronic skin disease that is caused by the Capripox virus. This viral disease is predominantly found in cows. Mosquitoes and ticks are the primary transmitters for the spread of this virus. Recently, LSDV has been rapidly spreading all over the world, especially in several areas of Pakistan, India, and Iran. Thousands of cows have died due to this infectious virus in Pakistan and early detection of LSDV is needed to avoid further loss. The prediction and classification of LSDV are hindered by the lack of publicly available datasets. Despite a few studies using LSDV datasets, such datasets are often small, which may lead to model overfitting. In this regard, we collect the dataset from several online sources, as well as, collecting images from veterinary farms in different areas of Pakistan. Deep learning has been widely used in the medical field for disease detection and classification. Therefore, this study leverages DenseNet deep learning models for LSDV detection and classification. Experiments are performed using VGG-16, ResNet-50, MobileNet-V2, custom-designed convolutional neural network, and Inception-V3. The DenseNet architecture presents a Convolutional Block Attention Module (CBAM) and Spatial Attention (SA) for the prediction and classification of LSD. Results demonstrate that a 99.11% accuracy can be obtained on the augmented dataset while a 94.23% accuracy can be achieved with the original dataset for chicken pox, monkey pox, and LSDV. Comparison with state-of-the-art studies corroborates the superior performance of the proposed model.

Abstract Insomnia was diagnosed by analyzing sleep stages obtained during polysomnography (PSG) recording. The state-of-the-art insomnia detection models that used physiological signals in PSG were successful in classification. However, the sleep stages of unbalanced data in small-time intervals were fed for classification in previous studies. This can be avoided by analyzing the insomnia detection structure in different frequency bands with artificially generated data from the existing one at the preprocessing and post-processing stages. Hence, the paper proposes a double-layered augmentation model using Modified Conventional Signal Augmentation (MCSA) and a Conditional Tabular Generative Adversarial Network (CTGAN) to generate synthetic signals from raw EEG and synthetic data from extracted features, respectively, in creating training data. The presented work is independent of sleep stage scoring and provides double-layered data protection with the utility of augmentation methods. It is ideally suited for real-time detection using a single-channel EEG provides better mobility and comfort while recording. The work analyzes each augmentation layer’s performance individually, and better accuracy was observed when merging both. It also evaluates the augmentation performance in various frequency bands, which are decomposed using discrete wavelet transform, and observed that the alpha band contributes more to detection. The classification is performed using Decision Tree (DT), Ensembled Bagged Decision Tree (EBDT), Gradient Boosting (GB), Random Forest (RF), and Stacking classifier (SC), attaining the highest classification accuracy of 94% using RF with a greater Area Under Curve (AUC) value of 0.97 compared to the existing works and is best suited for small datasets.

Abstract Significant progress has been made recently with the contribution of technological advances in studies on brain cancer. Regarding this, identifying and correctly classifying tumors is a crucial task in the field of medical imaging. The disease-related tumor classification problem, on which deep learning technologies have also become a focus, is very important in the diagnosis and treatment of the disease. The use of deep learning models has shown promising results in recent years. However, the sparsity of ground truth data in medical imaging or inconsistent data sources poses a significant challenge for training these models. The utilization of StyleGANv2-ADA is proposed in this paper for augmenting brain MRI slices to enhance the performance of deep learning models. Specifically, augmentation is applied solely to the training data to prevent any potential leakage. The StyleGanv2-ADA model is trained with the Gazi Brains 2020, BRaTS 2021, and Br35h datasets using the researchers’ default settings. The effectiveness of the proposed method is demonstrated on datasets for brain tumor classification, resulting in a notable improvement in the overall accuracy of the model for brain tumor classification on all the Gazi Brains 2020, BraTS 2021, and Br35h datasets. Importantly, the utilization of StyleGANv2-ADA on the Gazi Brains 2020 Dataset represents a novel experiment in the literature. The results show that the augmentation with StyleGAN can help overcome the challenges of working with medical data and the sparsity of ground truth data. Data augmentation employing the StyleGANv2-ADA GAN model yielded the highest overall accuracy for brain tumor classification on the BraTS 2021 and Gazi Brains 2020 datasets, together with the BR35H dataset, achieving 75.18%, 99.36%, and 98.99% on the EfficientNetV2S models, respectively. This study emphasizes the potency of GANs for augmenting medical imaging datasets, particularly in brain tumor classification, showcasing a notable increase in overall accuracy through the integration of synthetic GAN data on the used datasets.

Introduction: Ankle sprains remain one of the most commonly reported injuries in athletes. These injuries occur as a result of forced plantar flexion and inversion causing failure of the anterior talofibular ligament followed by the calcaneofibular ligament. Acute ankle sprains are generally treated conservatively. Nevertheless, 20%–40% of patients with an acute ankle sprain might develop chronic ankle instability (CAI). Surgical treatment is advocated for active young patients who develop CAI. MATERIALS AND METHODS: This study is a case series of 10 cases of CAI treated by the modified open Brostrom-Gould procedure from 2014 to 2021. The AOFAS ankle-hindfoot score and visual analog scores were used to evaluate the patient preoperatively and at regular intervals postoperatively. Results: All the patients had good-to-excellent results. There was a significant improvement in the mean AOFAS ankle-hindfoot scores from a preoperative 59.9 to a postoperative value of 91.3. The mean visual analog scores significantly dropped from a preoperative value of 5.9 to a postoperative value of 1.2. The minimum follow-up duration is around 1 year, while the maximum is around 8 years. Conclusion: The open modified Brostrom-Gould lateral ligament repair provides excellent results for chronic lateral ankle instability in active young individuals.

Purpose: The aim of this systematic review is to review the functional, clinical and radiological outcomes of patients undergoing large to massive rotator cuff repair with long head of biceps tendon (LHBT) autograft, as well as compare these to standard arthroscopic cuff repair. Methods: A review of the online Medline database was conducted on 20 October 2022 according to PRISMA guidelines and registered prospectively on the PROSPERO database. Clinical studies assessing patients with large to massive rotator cuff tears undergoing LHBT autograft repair were included. All studies reported on functional outcomes, range of movement (ROM) and radiological re-tear rates. The Methodological Index for Non-Randomised Studies (MINORS) tool was used to appraise all studies. Results: The search strategy identified ten studies for inclusion including a total of 594 patients. Five studies were comparable (346 patients), assessing LHBT autograft repair against arthroscopic rotator cuff repair without autograft. A variety of techniques of LHBT autograft were used across all studies, including bridging and augmentation styles. Radiographic comparison showed lower re-tear rates in the LHBT autograft group with two studies demonstrating statistically significant results. Pain scores, functional outcomes and ROM were significantly improved post-operatively in all studies for LHBT autograft patients, with no significant difference when compared to standard arthroscopic repair. Conclusion: LHBT autograft looks to significantly improve functional scoring and range of motion in patients with large to massive rotator cuff tears. When compared to standard arthroscopic cuff repair, LHBT autograft appears to significantly reduce the re-tear rate. Further randomised studies are needed to assess the efficacy of this technique. [ABSTRACT FROM AUTHOR]

Background: Machine learning models can provide quick and reliable assessments in place of medical practitioners. With over 50 million adults in the United States suffering from osteoarthritis, there is a need for models capable of interpreting musculoskeletal ultrasound images. However, machine learning requires lots of data, which poses significant challenges in medical imaging. Therefore, we explore two strategies for enriching a musculoskeletal ultrasound dataset independent of these limitations: traditional augmentation and diffusion-based image synthesis. Methods: First, we generate augmented and synthetic images to enrich our dataset. Then, we compare the images qualitatively and quantitatively, and evaluate their effectiveness in training a deep learning model for detecting thickened synovium and knee joint recess distension. Results: Our results suggest that synthetic images exhibit some anatomical fidelity, diversity, and help a model learn representations consistent with human opinion. In contrast, augmented images may impede model generalizability. Finally, a model trained on synthetically enriched data outperforms models trained on un-enriched and augmented datasets. Conclusions: We demonstrate that diffusion-based image synthesis is preferable to traditional augmentation. Our study underscores the importance of leveraging dataset enrichment strategies to address data scarcity in medical imaging and paves the way for the development of more advanced diagnostic tools.

Abstract Background High rates of labour augmentation with oxytocin have been found in some low- and lower-middle-income countries, causing potential perinatal harm. It is critical to understand the reasons for this overuse. Aim was to explore factors that shape practices around using oxytocin for labour augmentation in a high-volume labour ward in Dar es Salaam, Tanzania. Methods Mixed-methods data collection was conducted from March 2021 to February 2022, including structured observations of 234 births, 220 h of unstructured labour ward observations and 13 individual in-depth interviews with birth attendants. Thematic network analysis and descriptive statistics were used to analyse data. We used a time-lens to understand practices of oxytocin for labour augmentation in time-pressured labour wards. Results Birth attendants constantly had to prioritise certain care practices over others in response to time pressure. This led to overuse of oxytocin for augmentation to ensure faster labour progression and decongestion of the, often overburdened, ward. Simultaneously, birth attendants had little time to monitor foetal and maternal condition. Surprisingly, while oxytocin was used in 146 out of 234 (62.4%) structured labour observations, only 9/234 (4.2%) women had active labour lasting more than 12 h. Correspondingly, 21/48 (43.8%) women who were augmented with oxytocin in the first stage of labour had uncomplicated labour progression at the start of augmentation. While the partograph was often not used for decision-making, timing of starting oxytocin often correlated with natural cycles of ward-rounds and shift-turnovers instead of individual women’s labour progression. This resulted in co-existence of ‘too early’ and ‘too late’ use of oxytocin. Liberal use of oxytocin for labour augmentation was facilitated by an underlying fear of prolonged labour and low alertness of oxytocin-related risks. Conclusions Time scarcity in the labour ward often made birth attendants deviate from clinical guidelines for labour augmentation with oxytocin. Efforts to navigate time pressure resulted in too many women with uncomplicated labour progression receiving oxytocin with little monitoring of labour. Fear of prolonged labour and low alertness to oxytocin-mediated risks were crucial drivers. These findings call for research into safety and benefits of oxytocin in low-resource settings and interventions to address congestion in labour wards to prevent using oxytocin as a time-management tool.

Fractures of the proximal humerus account for 4%-8% of injuries to the appendicular skeleton. Most are stable, minimally displaced osteoporotic fractures in the elderly, and are the result of low-energy falls. A large majority of these patients regain adequate shoulder function without operative intervention. Surgery is considered in approximately 20% of patients because they require improved shoulder function for their activities of daily living or because of the significant deformity of their fracture and the need to restore functional alignment, length, and rotation in active, higher demand individuals. However, fixation of these fractures can pose a challenge due to poor bone quality and displacing forces of the rotator cuff. This is especially true in 3-part and 4-part fractures. These factors lead to the high failure rates seen with early attempts at osteosynthesis. In the last 2 decades, locking plate technology has been an innovation in treating these complex fractures. Despite the improvements in torsional strength and rigidity, outcome studies on locking plate technology demonstrate equivocal results with complication rates as high as 20%-30% and a revision rate of 10%. Specifically, these complications include avascular necrosis, varus collapse, intra-articular screw penetration, and postoperative stiffness. Varus collapse occurs when the weak osteoporotic bone fails around the implant. In turn, fibular strut endosteal augmentation was introduced to provide additional support and decrease implant failure rates in displaced fractures with varus coronal malalignment and significant metaphyseal bone loss. Although clinically successful and biomechanically superior to plate-only constructs, a few concerns remain. In turn, we introduce a novel technique of creating individual cancellous femoral head allograft struts or “French fries” that provides structural support for the humeral head but does not have the potential problems of a cortical fibular strut.

Human-AI collaboration is increasingly considered an alternative to fears of automation and large-scale job loss that are typically attributed to AI. However, the current scholarly understanding of this type of collaboration falls victim to a one-sided, rather optimistic view. More specifically, human-AI collaboration is commonly associated with augmenting or enhancing work. In this scenario, I use insights from qualitative research on the use of AI in practice to develop alternative perspectives around unexpected and unintended consequences of human-AI collaboration for worker safety. As such, I argue for future research to study human-AI collaboration in practice, to carefully unpack the ripple effects of changing or “enhanced” work practices, and to consider AI systems as being embedded in wider systems of cognition including, but not limited to, the mind, the body, and technologies.

In the last 30-40 years, there has been a trend in the incidence of cholelithiasis, which, in turn, is often complicated by the development of papillostenosis of various degrees requiring surgical correction. Currently, it is widely used in the diagnosis and treatment of papillostenosis: retrograde choledochopancreatography and endoscopic papillosphincterotomy, but these types of surgical correction of the pathology of large duodenal papilla have a number of disadvantages. In this connection, antegrade choledochoscopy with subsequent augmentation is an option for the diagnosis and correction of papillostenosis narrowing.Aim: of the study is to study domestic and foreign literature on methods of diagnosis and treatment of papillostenosis. To conduct a comparative analysis of the instrumental methods presented in practice, to evaluate the effectiveness of endoscopic antegrade choledochoscopy in papillostenosis.Materials and methods. 31 case histories of a patient who was hospitalized in the Department of Surgery of the Hospital No. 18 from 2013 to 2022 with papilla stenosis complicated by mechanical jaundice, who underwent followed by papillaa ugmentation, were analyzed. A narrowing of the lumen of the papilla of less than 2 mm was considered an indication for antegrade choledochoscopy with buging. The postoperative period in 100 % of patients proceeded without peculiarities. The number of bed-days of hospital stay is 4-6 days, discharged with improvement.Of 100 % of patients - 3.2 % (1 patient) – had a history of papilla restenosis after undergoing laparoscopic cholecystectomy, antegrade choledoscopy and augmentation (4.5 years after the first intervention), he underwent retrograde choledochopancreatography and endoscopic papillosphincterotomy.Conclusion. As presented for 10 years of experience in the use of antegrade choledochoscopy with the bougie of papilla in laparoscopic cholecystectomy, this technique for the diagnosis and treatment of papillostenosis has a large 5-year effectiveness (96.8 %), sensitivity, both due to the optical component of diagnosis and due to the objective component – bougie. In this connection, this method can be used in the practice of a surgeon.

Anxiety and depressive disorders are common in restless legs syndrome (RLS); in some cases, the disease is accompanied by cognitive impairment and a deterioration in quality of life. The most important treatment method for RLS is the use of dopaminergic medications. In some cases, the use of these drugs leads to a complication - the phenomenon of augmentation, which manifests itself in an increase in RLS symptoms as the dose of medication increases.Objective: to determine the clinical and neurophysiological characteristics of RLS patients with the augmentation phenomenon.Material and methods. 40 patients with RLS were examined: 20 with augmentation (main group, MG; 5 men and 15 women) and 20 without augmentation (comparison group, CG; 7 men and 13 women). The median age of the patients in the MG was 63.5 [56; 71] years, and 62.0 [43.5; 71.5] years in the CG. Clinical assessment was performed using the RLS Severity Rating Scale (RS), Montreal Cognitive Assessment Scale (MoCA), Beck Anxiety Inventory, Beck Depression Inventory, Quality of Life Scale (SF-36), Insomnia Severity Index (ISI), Trail Making Test (Part A), Trail Making Test (part B), phonemic and semantic speech activity tests. All patients underwent suggested immobilization test (SIT) to assess the urge to move on a 10-point numerical rating scale (NRS) and periodic limb movements (PLM) while awake, as well as a polysomnographic examination.Results. Augmentation significantly more frequently resulted in an expansion of the area of distribution of the urge to move and other unpleasant sensations over the entire surface of the legs (p=0.01), painful discomfort in the legs (p=0.001), early onset of symptoms (04:00—18:00; p=0.04), shortening of the latency period for the onset of symptoms (p=0.001), twitching in the legs while awake (p=0.04), taking higher doses of dopaminergic medications (p=0.004). In augmentation, the MoCA score is lower (p=0.01), such patients use fewer words in the semantic speech activity test (p=0.049), have a higher score on the RS (p=0.001) and ISI (p=0.02), a greater number of PLMs while awake according to SIT (p=0.01) compared to CG. No significant differences were found between groups in terms of age, gender, ferritin level, total score on the Beck Anxiety and Depression Inventory, SF-36 Quality of Life Scale, Trail Making Test (Part A), Trail Making Test (part B), phonemic speech activity test, polysomnography indicators (including motor activity during sleep).Conclusion. From a clinical and neurophysiological point of view, the phenomenon of augmentation is not simply a manifestation of a more severe course of RLS but has features that reflect the pathogenesis of this disorder. During augmentation, patients tend to describe the sensations as painful and their involuntary motor activity increases. This reflects changes in the activity of the diencephalospinal tract due to excessive dopaminergic stimulation.

Rotator cuff tears (RCT) and instability are the most common surgically treated shoulder pathologies. The concept of augmentation using the long head of the biceps tendon (LHBT) autograft was created to improve the results of surgical treatment of these pathologies, especially in cases of chronic and massive injuries. The popularity of using the LHBT for augmentation is evidenced by the significant number of publications on this topic published in the last 3 years; however, only one systematic review has been published regarding only LHBT augmentation for massive RCTs. Several studies comparing partial repair with partial repair and additional LHBT augmentation for RCT showed superior clinical outcomes and lower re-tear rates when LHBT augmentation was performed. There is a rising popularity of using LHBT as an autograft to perform superior capsule reconstruction (SCR) in case of irreparable rotator cuff tears. In recent years, shoulder stabilization by arthroscopic Bankart repair with biceps augmentation has been promoted with very promising short-term results. The evidence provided by studies appears to be sufficient to recommend the use of LHBT for augmentation whenever necessary; however, larger studies with long-term follow-up are needed.

Leaf blight spot disease, caused by bacteria and fungi, poses a threat to plant health, leading to leaf discoloration and diminished agricultural yield. In response, we present a MobileNetV3 based classifier designed for the Jasmine plant, leveraging lightweight Convolutional Neural Networks (CNNs) to accurately identify disease stages. The model integrates depth wise convolution layers and max pool layers for enhanced feature extraction, focusing on crucial low level features indicative of the disease. Through preprocessing techniques, including data augmentation with Conditional GAN and Particle Swarm Optimization for feature selection, the classifier achieves robust performance. Evaluation on curated datasets demonstrates an outstanding 97% training accuracy, highlighting its efficacy. Real world testing with diverse conditions, such as extreme camera angles and varied lighting, attests to the model's resilience, yielding test accuracies between 94% and 96%. The dataset's tailored design for CNN based classification ensures result reliability. Importantly, the model's lightweight classification, marked by fast computation time and reduced size, positions it as an efficient solution for real time applications. This comprehensive approach underscores the proposed classifier's significance in addressing leaf blight spot disease challenges in commercial crops.

Background: High rates of labour augmentation with oxytocin have been found in some low- and lower-middle-income countries, causing potential perinatal harm. It is critical to understand the reasons for this overuse. Aim was to explore factors that shape practices around using oxytocin for labour augmentation in a high-volume labour ward in Dar es Salaam, Tanzania. Methods: Mixed-methods data collection was conducted from March 2021 to February 2022, including structured observations of 234 births, 220 h of unstructured labour ward observations and 13 individual in-depth interviews with birth attendants. Thematic network analysis and descriptive statistics were used to analyse data. We used a time-lens to understand practices of oxytocin for labour augmentation in time-pressured labour wards. Results: Birth attendants constantly had to prioritise certain care practices over others in response to time pressure. This led to overuse of oxytocin for augmentation to ensure faster labour progression and decongestion of the, often overburdened, ward. Simultaneously, birth attendants had little time to monitor foetal and maternal condition. Surprisingly, while oxytocin was used in 146 out of 234 (62.4%) structured labour observations, only 9/234 (4.2%) women had active labour lasting more than 12 h. Correspondingly, 21/48 (43.8%) women who were augmented with oxytocin in the first stage of labour had uncomplicated labour progression at the start of augmentation. While the partograph was often not used for decision-making, timing of starting oxytocin often correlated with natural cycles of ward-rounds and shift-turnovers instead of individual women's labour progression. This resulted in co-existence of 'too early' and 'too late' use of oxytocin. Liberal use of oxytocin for labour augmentation was facilitated by an underlying fear of prolonged labour and low alertness of oxytocin-related risks. Conclusions: Time scarcity in the labour ward often made birth attendants deviate from clinical guidelines for labour augmentation with oxytocin. Efforts to navigate time pressure resulted in too many women with uncomplicated labour progression receiving oxytocin with little monitoring of labour. Fear of prolonged labour and low alertness to oxytocin-mediated risks were crucial drivers. These findings call for research into safety and benefits of oxytocin in low-resource settings and interventions to address congestion in labour wards to prevent using oxytocin as a time-management tool. [ABSTRACT FROM AUTHOR]

In the context of autonomous driving, the augmentation of existing data through simulations provides an elegant solution to the challenge of capturing the full range of adverse weather conditions in training datasets. However, existing physics-based augmentation models typically rely on single scattering approximations to predict light propagation under unfavorable conditions, such as fog. This can prevent the reproduction of important signal characteristics encountered in a real-world environment. Consequently, in this work, Monte Carlo simulations are employed to assess the relevance of multiple-scattered light to the detected LiDAR signal in different types of fog, with scattering phase functions calculated from Mie theory considering real particle size distributions. Bidirectional path tracing is used within the self-developed GPU-accelerated Monte Carlo software to compensate for the unfavorable photon statistics associated with the limited detection aperture of the LiDAR geometry. To validate the Monte Carlo software, an analytical solution of the radiative transfer equation for the time-resolved radiance in terms of scattering orders is derived, thereby providing an explicit representation of the double-scattered contributions. The results of the simulations demonstrate that the shape of the detected signal can be significantly impacted by multiple-scattered light, depending on LiDAR geometry and visibility. In particular, double-scattered light can dominate the overall signal at low visibilities. This indicates that considering higher scattering orders is essential for improving AI-based perception models. [ABSTRACT FROM AUTHOR]

The main objective of this paper is to present a repository of image databases whose features are augmented with embedded vector field (VF) features. The repository is designed to provide the user with image databases that enhance machine learning (ML) classification. Also, six VFs are provided, and the user can embed them into her/his own image database with the help of software named ELPAC. Three of the VFs generate real-shaped singular points (SPs): springing, sinking, and saddle. The other three VFs generate seven kinds of SPs, which include the real-shaped SPs and four complex-shaped SPs: repelling and attracting (out and in) spirals and clockwise and counterclockwise orbits (centers). Using the repository, this work defines the locations of the SPs according to the image objects and the mappings between the SPs' shapes if separate VFs are embedded into the same image. Next, this paper produces recommendations for the user on how to select the most appropriate VF to be embedded in an image database so that the augmented SP shapes enhance ML classification. Examples of images with embedded VFs are shown in the text to illustrate, support, and validate the theoretical conclusions. Thus, the contributions of this paper are the derivation of the SP locations in an image; mappings between the SPs of different VFs; and the definition of an imprint of an image and an image database in a VF. The advantage of classifying an image database with an embedded VF is that the new database enhances and improves the ML classification statistics, which motivates the design of the repository so that it contains image features augmented with VF features. [ABSTRACT FROM AUTHOR]

Background: Legumes, such as lentils, field peas, Faba beans and chickpeas, are high in vitamins, fiber, important minerals and protein and can help avoid obesity and cardiovascular illnesses. They also contribute to ecosystem services, such as nitrogen fixation and resilience to environmental stresses. Despite a 60% increase in global pulse production from 2000 to 2021, a demand-supply gap, especially in South Asia, raises concerns about nutritional access. Since illnesses are currently an issue to the food security of faba beans, machine learning is required for efficient disease identification. Methods: This research employs Convolutional Neural Networks (CNNs) for robust Faba bean leaf disease identification. The CNN model is trained with diverse images representing specific diseases. The study focuses on diseases like Chocolate Spot, Faba Bean Gall, Rust and Healthy leaves. Image processing involves resizing, grayscale conversion and labeling. The CNN architecture includes eight convolutional layers, four max-pooling layers and three dropout layers. The model is trained using 80% of the dataset, validated using 20% and tested for accuracy. Result: The CNN model achieves an accuracy of 99.37% during training and 89.69% during validation after 75 epochs. Confusion matrix and classification report illustrate the model’s performance. It shows high precision, recall and F1 scores for each class, indicating balanced performance. Chocolate Spot and Rust exhibit the highest precision and F1 scores. The overall accuracy is 91%, comparable to other studies on Faba bean disease detection. The study presents a CNN-based disease identification system for Faba beans, demonstrating high accuracy and balanced performance across different diseases. The model’s effectiveness is comparable to other advanced techniques. The research highlights the potential of machine learning in optimizing disease management for Faba beans. Future work could explore a broader range of diseases and incorporate hybrid machine learning algorithms for further improvement. [ABSTRACT FROM AUTHOR]

The classification of a pre-processed fMRI dataset using functional connectivity (FC)-based features is considered a challenging task because of the set of high-dimensional FC features and the small dataset size. To tackle this specific set of FC high-dimensional features and a small-sized dataset, we propose here a conditional Generative Adversarial Network (cGAN)-based dataset augmenter to first train the cGAN on computed connectivity features of NYU dataset and use the trained cGAN to generate synthetic connectivity features per category. After obtaining a sufficient number of connectivity features per category, a Multi-Head attention mechanism is used as a head for the classification. We name our proposed approach "ASD-GANNet", which is end-to-end and does not require hand-crafted features, as the Multi-Head attention mechanism focuses on the features that are more relevant. Moreover, we compare our results with the six available state-of-the-art techniques from the literature. Our proposed approach results using the "NYU" site as a training set for generating a cGAN-based synthetic dataset are promising. We achieve an overall 10-fold cross-validation-based accuracy of 82%, sensitivity of 82%, and specificity of 81%, outperforming available state-of-the art approaches. A sitewise comparison of our proposed approach also outperforms the available state-of-the-art, as out of the 17 sites, our proposed approach has better results in the 10 sites. [ABSTRACT FROM AUTHOR]

Rationale: The efficacy and safety of antidepressant augmentation therapy with aripiprazole (AATA) has been established; however, the ongoing effects of continuing aripiprazole after remission remain unclear because no studies have examined this issue. Objectives: We aimed to explore the effect of AATA discontinuation on the major depressive disorder (MDD) recurrence risk in patients with remitted MDD after AATA. Methods: This 24-week, multicenter, placebo-controlled, double-blind, randomized trial evaluated recurrence risk in patients with MDD who achieved remission with AATA. Differences in MDD recurrence, as defined by the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, between the two groups were compared using survival analysis. The differences in depressive symptom severity and social functioning between the two groups were compared using a mixed model with repeated measures. Extrapyramidal symptoms and akathisia were also assessed. Results: Twenty-three participants were randomized and treated. Two patients in each group experienced recurrence during the study. Kaplan–Meier analysis with Log-rank comparison showed no difference in recurrence between groups (p = 0.642). No significant difference in interactions between group and period was observed in the 17-item Hamilton depression rating scale (p = 0.492) or the Social and Occupational Functioning Assessment Scale (p = 0.638). No patients developed extrapyramidal symptoms or akathisia. Conclusions: Definitive conclusions could not be drawn owing to the small sample size. This study represents a starting point for investigating the safety of aripiprazole discontinuation on recurrence in patients with MDD who have achieved remission with AATA. Future studies with appropriate sample sizes calculated based on this study are needed. [ABSTRACT FROM AUTHOR]

Neuroimaging has emerged over the last few decades as a crucial tool in diagnosing Alzheimer's disease (AD). Mild cognitive impairment (MCI) is a condition that falls between the spectrum of normal cognitive function and AD. However, previous studies have mainly used handcrafted features to classify MCI, AD, and normal control (NC) individuals. This paper focuses on using gray matter (GM) scans obtained through magnetic resonance imaging (MRI) for the diagnosis of individuals with MCI, AD, and NC. To improve classification performance, we developed two transfer learning strategies with data augmentation (i.e. shear range, rotation, zoom range, channel shift). The first approach is a deep Siamese network (DSN), and the second approach involves using a cross-domain strategy with customized VGG-16. We performed experiments on the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset to evaluate the performance of our proposed models. Our experimental results demonstrate superior performance in classifying the three binary classification tasks: NC vs. AD, NC vs. MCI, and MCI vs. AD. Specifically, we achieved a classification accuracy of 97.68%, 94.25%, and 92.18% for the three cases, respectively. Our study proposes two transfer learning strategies with data augmentation to accurately diagnose MCI, AD, and normal control individuals using GM scans. Our findings provide promising results for future research and clinical applications in the early detection and diagnosis of AD. [ABSTRACT FROM AUTHOR]

Malaria is a significant global health issue, responsible for the highest rates of morbidity and mortality globally. This paper introduces a very effective and precise convolutional neural network (CNN) method that employs advanced deep learning techniques to automate the detection of malaria in images of red blood cells (RBC). Furthermore, we present an emerging and efficient deep learning method for differentiating between cells infected with malaria and those that are not infected. To thoroughly evaluate the efficiency of our approach, we do a meticulous assessment that involves comparing different deep learning models, such as ResNet-50, MobileNet-v2, and Inception-v3, within the domain of malaria detection. Additionally, we conduct a thorough comparison of our proposed approach with current automated methods for malaria identification. An examination of the most current techniques reveals differences in performance metrics, such as accuracy, specificity, sensitivity, and F1 score, for diagnosing malaria. Moreover, compared to existing models for malaria detection, our method is the most successful, achieving an accurate score of 1.00 in all statistical matrices, confirming its promise as a highly efficient tool for automating malaria detection. [ABSTRACT FROM AUTHOR]

Training complex deep neural networks can result in overfitting when the networks are trained from random weight initialization on small datasets. Augmentation helps to reduce the negative effects of overfitting. The findings in computer vision and audio recognition research reveals that the performance of machine learning classifiers is significantly improved when augmentation is used. In the context of ecology, researchers conduct field surveys whereby microphones are placed in some location and audio data is recorded over a period of time. There is however no guarantee that the particular species of interest in the field survey will vocalize frequently near the microphone. Thus, the amount of data captured for the species of interest might be limited, and it may then be the source of overfitting. The main contribution of this paper is in performing experiments with time and frequency masking, and noise addition augmentation techniques in training a visual convolutional neural networks (CNN) repurposed for pattern recognition in acoustic spectrograms. These techniques increased the audio examples for the pin-tailed whydah and the Cape robin-chat to create a robust audio vocalization classifiers. To evaluate the performance of the augmentation techniques we conducted a comparison between experiments run with and without augmentation. We chose to use CNN as our classifier given that they are state-of-the-art in audio recognition tasks and they have revealed good performance. In the used augmentation techniques; time masking achieved 90.2% as the highest testing accuracy while pink noise is the most successful best classifier. [ABSTRACT FROM AUTHOR]

Wheat leaf disease prevention and treatment requires an accurate and rapid classification of wheat leaf diseases and their extent. Using healthy wheat, leaf rust, crown, root rot, and wheat loose smut as research objects, this study proposes a deep learning-based technique for classifying wheat leaf diseases. A collaborative generative adversarial network is used as an image imputation in the proposed methodology, allowing a generator and discriminator network to properly estimate the missing data in the dataset using the residual method. It is used to improve feature extraction in wheat leaf images. The major contribution of this study is to use a pre-trained deep learning convolutional neural network architecture as a foundation to improve and construct an automated tool for wheat leaf disease image categorization. To classify wheat leaf diseases, a modification to the Residual Network with 50 layers (ResNet50) is being suggested. The ′Conv′, ′Batch Normaliz′, and ′Activation Leaky Relu′ layers were added as part of this modification. These layers are inserted into the ResNet50 architecture for accurate feature extraction and discrimination. Extensive tests are carried out to evaluate the proposed model's performance on photos from a large wheat disease classification dataset. The suggested approach outperforms ResNet50, InceptionV3, and DenseNet, according to the experimental findings. The suggested method achieves the greatest identification accuracy of 98.44%. These discoveries might aid in the accurate detection and categorization of wheat leaf diseases. [ABSTRACT FROM AUTHOR]

Purpose: Management of massive rotator cuff tears (MRCTs) remains debated, and various arthroscopic and open techniques have been described for their management. Nevertheless, the optimal strategy remains unclear. The present study evaluated the clinical results in patients managed arthroscopically for MRCTs augmented with the long head biceps tendon (LHBT) at a minimum 1‐year follow‐up, considering different type of tears, demographic data and number of torn tendons. Methods: Patients treated in a secondary referral centre from January 2021 to April 2022 were enroled prospectively. Inclusion criteria were pain, inability to fully elevate the affected shoulder, irreparable tears and active and motivated patients. All patients were managed within 2 months from diagnosis in a single centre by the same surgeons. Preoperative shoulder radiographs and magnetic resonance imaging (MRI) were collected, and clinical assessment was also performed using the Numerical analogue scale (NAS), Constant score (CS) American Shoulder and Elbow Surgeons Shoulder Score (ASES). Tissue retraction and tendon fatty infiltration were evaluated using Patte and Fuchs scale, respectively. Clinical assessment was performed using the same scales at 3–6 months and 1‐year follow‐up. Results: A total of 55 patients (31 female and 24 male) with a mean age of 60 ± 7.1 years were enroled for a mean follow‐up of 18.2 ± 4.3 months. The mean preoperative NAS was 7.8 ± 0.6, CS was 20.5 ± 7.6 and ASES was 22.6 ± 9.2, increasing, respectively, to 0.3 ± 0.6, 91.5 ± 6.9 and 94.2 ± 6.7. No adverse side effects (infection, rejection, allergy) were reported during the study period. All patients were evaluated after surgery at 3 and 6 months and 1 year with statistically significant improvement for each score at the first and last follow‐up (p < 0.05). Conclusions: The use of LHBT augmentation in patients with MRCTs in appropriately selected patients is safe and effective and can lead to pain relief and acceptable clinical outcomes. Furthermore, its use carries low donor site morbidity and is cost effective. Comparative studies, including randomised controlled trials, with other proposed techniques are needed to confirm these findings. Level of Evidence: Level IV. [ABSTRACT FROM AUTHOR]

This article reviews the literature that examines the potential, limitations, and consequences of robots and artificial intelligence (AI) in automation and augmentation across various disciplines. It presents key observations and suggestions from the literature review. Firstly, displacement effects from task automation continue to persist. However, one should not assume an unequivocally increasing efficacy of technology in automation or augmentation, especially given the declining productivity growth in high-income countries and some large emerging economies in recent decades. Jobs less likely to be negatively impacted are those that require diverse tasks, physical dexterity, tacit knowledge, or flexibility, or are protected by professional or trade associations. Despite countervailing effects, without policy intervention, automation and augmentation could widen inequality between social groups, labor and capital, and firms. Secondly, AI's promise in task automation and labor augmentation is mixed. AI tools can cause harm, and dissatisfaction and disengagement often arise from their opaqueness, errors, disregard for critical contexts, lack of tacit knowledge, and lack of domain expertise, as well as their demand for extra labor time and resources. The inadequate autonomy to override AI-based assessments further frustrates users who have to use these AI tools at work. Finally, the article calls for sociological research to specify conditions and mechanisms that ameliorate adverse consequences and enhance labor augmentation by embedding the study of automation and augmentation in concrete social and political contexts at multiple levels. [ABSTRACT FROM AUTHOR]