Standard Metrics Research Articles

Bangladeshi crops leaf disease detection using YOLOv8

The agricultural sector in Bangladesh is a cornerstone of the nation's economy, with key crops such as rice, corn, wheat, potato, and tomato playing vital roles. However, these crops are highly vulnerable to various leaf diseases, which pose significant threats to crop yields and food security if not promptly addressed. Consequently, there is an urgent need for an automated system that can accurately identify and categorize leaf diseases, enabling early intervention and management. This study explores the efficacy of the latest state-of-the-art object detection model, YOLOv8 (You Only Look Once), in surpassing previous models for the automated detection and categorization of leaf diseases in these five major crops. By leveraging modern computer vision techniques, the goal is to enhance the efficiency of disease detection and management. A dataset comprising 19 classes, each with 150 images, totaling 2850 images, was meticulously curated and annotated for training and evaluation. The YOLOv8 framework, known for its capability to detect multiple objects simultaneously, was employed to train a deep neural network. The system's performance was evaluated using standard metrics such as mean Average Precision (mAP) and F1 score. The findings demonstrate that the YOLOv8 framework successfully identifies leaf diseases, achieving a high mAP of 98% and an F1 score of 97%. These results underscore the significant potential of this approach to enhance crop disease management, thereby improving food security and promoting agricultural sustainability in Bangladesh.

Reaching While Learning to Sit: Capturing the Kinematics of Co-Developing Skills at Home.

This study examined the co-development of infant reaching and postural control across the transition to arms-free sitting at home. We observed infants with typical likelihood (TL; n=24) and elevated likelihood (EL; n=20) for autism at four biweekly sessions spanning the transition to arms-free sitting (infant age=4.5-8 months at first session). At each session, infants sat on a pressure-sensitive mat with external support or independently, wore magneto-inertial sensors on both wrists, and reached for toys presented at midline. Analyses focused on characterizing and comparing control of sitting during reaching actions and standard kinematic metrics of reaching during Supported versus Independent Sitting. Although EL infants achieved arms-free sitting later than TL peers, there were no group differences on any measures. Across sessions, infants' control of the sitting posture during concurrent reaching movements improved in both contexts, though they were less stable as they reached when sitting independently compared to when sitting with support. A similar effect was apparent in the kinematics of reaches, with overall improvement over time, but evidence of poorer control in Independent relative to Supported Sitting. Taken together, these findings underscore the mutually influential and dynamic relations between emerging skills and well-established behaviors.

Modernizing Mainframe Monitoring: A Comparative Analysis of Observability Dashboarding Tools

Abstract: This article explores the critical role of observability dashboarding in mainframe environments, focusing on the integration and support of modern observability tools. It examines the functionalities and benefits of leading solutions such as Prometheus with Grafana, Dynatrace, Datadog, AppDynamics, IBM Instana, and Checkmk, providing insights into how organizations can enhance their mainframe monitoring capabilities. The article emphasizes the significance of a multi-tool observability strategy, highlighting its advantages in flexibility, redundancy, and comprehensive coverage. Key benefits of effective observability, including proactive issue detection, downtime reduction, improved capacity planning, and enhanced security monitoring, are thoroughly discussed. The article also outlines best practices for implementing observability in mainframes, covering unified dashboards, standardized metrics, automation, training, and continuous improvement. By analyzing these tools and strategies, the research offers valuable guidance for organizations seeking to streamline their observability approaches, ultimately enhancing customer experience and operational efficiency in mainframe-dependent environments. This comprehensive examination of mainframe observability techniques provides a roadmap for leveraging legacy systems alongside cutting-edge monitoring solutions, addressing the evolving needs of modern IT infrastructures.

Quantifying antimicrobial consumption in the Chilean salmon industry using standardized metrics and indicators: The case of florfenicol and oxytetracycline

The adoption of standardized metrics and indicators of antimicrobial use (AMU) in the food animal industry is essential for the success of programs aimed at promoting the responsible and judicious use of antimicrobials in this activity. The objective of this study was to introduce the use of standardized AMU metrics and indicators to quantify the use of florfenicol and oxytetracycline in the Chilean salmon industry, and in this way evaluate the feasibility of their use given the type of health and production information currently managed by the National Fisheries and Aquaculture Service (SERNAPESCA), the Chilean agency responsible for regulating aquaculture in Chile. The data available from SERNAPESCA allowed the construction and evaluation of the most data-demanding AMU metrics and indicators. Consequently, the use of florfenicol and oxytetracycline administered by oral and parenteral routes was quantified using the treatment incidence based on both animal defined daily dose (TIDDDvet) and animal used daily dose (TIUDDA). To that end, the study included information from 1320 closed production cycles from farms rearing Atlantic salmon, coho salmon and rainbow trout that were active between January 2017 and December 2021. By applying standardized AMU metrics and indicators, we were able to determine that the median of TIDDDvet for florfenicol was 75.1 (80 % range, 20.0–158.0) DDDvet per ton-year at risk for oral procedures and 0.36 (80 % range, 0.07–1.19) DDDvet per ton-year at risk for parenteral procedures. For oxytetracycline, the median TIDDDvet was 3.09 (80 % range, 0.74–42.8) and 0.47 (80 % range, 0.09–1.68) DDDvet per ton-year at risk for oral and parenteral procedures, respectively. The median TIUDDA for treatments with florfenicol was 45.6 (80 % range, 10.9–96.5) UDDA per ton-year at risk for oral treatments and 0.28 (80 % range, 0.05–0.80) UDDA per ton-year at risk for parenteral treatments. For oxytetracycline, the median TIUDDA was 2.63 (80 % range, 0.61–28.2) UDDA per ton-year at risk for oral treatments and 0.41 (80 % range, 0.08–1.29) UDDA per ton-year at risk for parenteral treatments. This study demonstrates that it is feasible to move from traditional AMU metrics and indicators to standardized ones in the Chilean salmon industry. This is possible because the competent authority requires salmon farms to report detailed health and production information at a high frequency. The use of standardized AMU metrics and indicators can help the authority to have a more comprehensive view of the antimicrobial use in the Chilean salmon industry.

Lite approaches for long-range multi-step water quality prediction

Forecasting accurate water quality is very important in aquaculture, environment monitoring, and many other applications. Many internal and external factors influence water quality. Therefore, water quality parameters exhibit complex time series characteristics. Consequently, long-range accurate prediction of water quality parameters suffers from poor propagation of information from past timepoints to further future timepoints. Moreover, to synchronise the prediction model with the changes in the time series characteristics, periodic retraining of the prediction model is required and such retraining is to be done on resource-restricted computation devices. In this work, we present a low-cost training approach to improve long-range multi-step water quality prediction. We train a short-range predictor to save training effort. Then, we strive to achieve and/or improve long-range prediction using multi-step iterative ensembling during inference. Experimental results on 9 water quality datasets demonstrate that the proposed method achieves significantly lower error than the existing state-of-the-art approaches. Our approach significantly outperforms the existing approaches in several standard metrics, even in the case of future timepoints at long distances.

GAN-SkipNet: A Solution for Data Imbalance in Cardiac Arrhythmia Detection Using Electrocardiogram Signals from a Benchmark Dataset

Electrocardiography (ECG) plays a pivotal role in monitoring cardiac health, yet the manual analysis of ECG signals is challenging due to the complex task of identifying and categorizing various waveforms and morphologies within the data. Additionally, ECG datasets often suffer from a significant class imbalance issue, which can lead to inaccuracies in detecting minority class samples. To address these challenges and enhance the effectiveness and efficiency of cardiac arrhythmia detection from imbalanced ECG datasets, this study proposes a novel approach. This research leverages the MIT-BIH arrhythmia dataset, encompassing a total of 109,446 ECG beats distributed across five classes following the Association for the Advancement of Medical Instrumentation (AAMI) standard. Given the dataset’s inherent class imbalance, a 1D generative adversarial network (GAN) model is introduced, incorporating the Bi-LSTM model to synthetically generate the two minority signal classes, which represent a mere 0.73% fusion (F) and 2.54% supraventricular (S) of the data. The generated signals are rigorously evaluated for similarity to real ECG data using three key metrics: mean squared error (MSE), structural similarity index (SSIM), and Pearson correlation coefficient (r). In addition to addressing data imbalance, the work presents three deep learning models tailored for ECG classification: SkipCNN (a convolutional neural network with skip connections), SkipCNN+LSTM, and SkipCNN+LSTM+Attention mechanisms. To further enhance efficiency and accuracy, the test dataset is rigorously assessed using an ensemble model, which consistently outperforms the individual models. The performance evaluation employs standard metrics such as precision, recall, and F1-score, along with their average, macro average, and weighted average counterparts. Notably, the SkipCNN+LSTM model emerges as the most promising, achieving remarkable precision, recall, and F1-scores of 99.3%, which were further elevated to an impressive 99.60% through ensemble techniques. Consequently, with this innovative combination of data balancing techniques, the GAN-SkipNet model not only resolves the challenges posed by imbalanced data but also provides a robust and reliable solution for cardiac arrhythmia detection. This model stands poised for clinical applications, offering the potential to be deployed in hospitals for real-time cardiac arrhythmia detection, thereby benefiting patients and healthcare practitioners alike.

A Rank Graduation Box for SAFE AI

The growth of Artificial Intelligence applications requires to develop risk management models that can balance opportunities with risks.We contribute to the development of Artificial Intelligence risk models proposing a Rank Graduation Box (RGB), a set of integrated statistical metrics that can measure the “Sustainability”, “Accuracy”, “Fairness” and “Explainability” of any Artificial Intelligence application.Our metrics are consistent with each other, as they are all derived from a common underlying statistical methodology: the Lorenz curve.The validity of the metrics is assessed by means of their practical application to both simulated and real data. The results from the comparison of alternative machine learning models to simulated data are aligned with the generating models, in general indicating linear regression models as the most accurate, regression tree models as the most fair, Random Forest models as the most robust; and leading to model explanations similar to the true ones from the generating model.The outcomes from the application of Random Forest models to real data show that the proposed RGB metrics are more interpretable and more consistent with the expectations, with respect to standard metrics such as AUC, RMSE and Shapley values. The evidence also shows that the RGB metrics are very general and can be applied to any machine learning method, regardless of the underlying data and model.

Neurological Surgery Resident ABNS Written Exam Scores Before and After Introduction of a Weekly Didactic Educational Intervention: A 12-Year Single-Institution Retrospective Study.

United States neurological surgery residency education has undergone substantive changes over the past 2 decades. Neurosurgical professional bodies have developed numerous initiatives providing standardized assessments and training opportunities for residency programs. However, there have been few studies using standardized measures to assess core components of educational programming in individual programs. We conducted a 12-year retrospective review of resident American Board of Neurological Surgery (ABNS) board scores using our institutional data from 2010 to 2021 to determine the effect of introducing a weekly didactic resident education hour (REH) on resident scores in the ABNS written in-training examination. ABNS scaled scores were analyzed before (2010-2016) and after (2017-2021) REH introduction. To account for a practice effect, we used a 2-factor linear regression model with an interaction term. We obtained ABNS scores from 43 residents representing 132 test attempts. The average ABNS scaled score significantly improved after the introduction of REH (319 vs 410, t = -3.44, P = .0008). Accounting for the practice effect revealed a significant interaction effect between the number of attempts taking the ABNS examination and whether formal didactics were taught, accounting for 46.2 points on the examination (t = 2.309, P = .023); however, REH alone did not have a significant effect on the scaled scores (t = -1.649, P = .102). ABNS written board scores represent a standardized metric by which educational initiatives within training programs may be assessed for efficacy. Further research is needed to identify educational approaches that are effective to meet the goal of demonstrated mastery of fundamental knowledge in neurosurgery across a diversity of neurological surgery residency programs.

A Linear time shrinking-SL(t)-ViT approach for brain tumor identification and categorization

Automated brain tumor detection and classification systems have gained popularity in recent years because traditional diagnosis procedures are time-consuming and costly in nature. Deep learning(DL) methods, specifically pre-trained convolutional neural networks (CNNs), have shown promising results in accurately and rapidly classifying brain tumors. However, the lack of diverse magnetic resonance imaging (MRI) datasets has hindered the ability of DL algorithms to generalize effectively. To address this issue, this paper proposes a DL model using generative adversarial networks (GANs) in conjunction with data augmentation strategies and a structural similarity loss function employed for generating annotated images. A novel DL model inspired from Vision Transformer, Shrinking Linear Time Vision Transformer (SL(t)-ViT) network model is proposed for disease data classification. The model underwent extensive evaluation across multiple datasets, employing standard performance metrics to assess its efficacy in brain tumor identification. The proposed model achieved remarkable testing accuracies of 0.995, 0.996, 0.9954, 0.998, and 0.997 for binary classification tasks across multiple datasets, and 0.986, 0.982, 0.985, and 0.993 for multi-class classification tasks. These results underscore the superior performance of our proposed model, showcasing its capability to outperform state-of-the-art techniques. Specifically, it demonstrated a substantial margin of improvement, ranging from 1-2% for binary classification and 9-10% for multi-class classification, solidifying its position as a leading approach in brain tumor identification. Results demonstrate the efficacy of the proposed model, outperforming other models. Overall, this study highlights the potential of SL(t)-ViT and GANs in improving the accuracy, resource consumption, and efficiency of brain tumor diagnosis.

Comparative Analysis of Supervised Machine Learning Algorithms for Diabetes Prediction

Diabetes Mellitus is a chronic disease with far-reaching consequences, necessitating innovative diagnostic approaches. Current methods have limitations, highlighting the need for advanced techniques. This research leveraged machine learning algorithms to predict diabetes using the Pima Indian Diabetes Dataset. Six algorithms were developed and compared: Logistic Regression, Support Vector Machine, K-Nearest Neighbor, Naive Bayes, Random Forest, and Decision Tree. Standard evaluation metrics assessed their performance. The Random Forest Classifier emerged as the top-performing algorithm, achieving an impressive 91% accuracy and demonstrating exceptional reliability. This study showcases machine learning's potential to transform diabetes diagnosis and management. The developed web application provides a user-friendly interface for predicting diabetes, facilitating timely intervention and improved health outcomes. The findings contribute significantly to the development of intelligent disease prediction systems, promoting early detection and enhanced patient care. By harnessing machine learning, this research pioneers a new frontier in diabetes diagnosis, paving the way for improved patient outcomes and enhanced healthcare services.

Developing an Index of National Research Capacity

Abstract We test the feasibility of incorporating broad social, political, and governance indicators with standard metrics as a way to enrich assessment of national research capacity. We factor analyze two sets of variables for 174 countries from 2012 to 2021, one being traditional measures associated with national science and technology capacity such spending, and a second being broader social, political and governance measures, such as academic freedom. As expected, two factors emerge, one for raw or “core” research capacity and the other indicating the wider governance context. Further analysis shows convergent validity within the two factors and divergent validity between them. The analysis also quantifies the contribution of each indicator to each factor. Nations rank differently for each factor and also when combined . Ranks vary as a function of the chosen aggregation method. As a test of the predictive validity of the capacity index, we find both factors to be associated with country level field weighted citation index. Policymakers and analysts may find useful feedback from this approach to quantifying national research strength. Peer Review https://www.webofscience.com/api/gateway/wos/peer-review/10.1162/qss_a_00325

Ensemble neural models for ICD code prediction using unstructured and structured healthcare data

Disease coding is the process of assigning one or more standardized diagnostic codes to clinical notes that are maintained by health practitioners (e.g. clinicians) to track patient condition. Such a coding process is often expensive and error-prone, as human medical coders primarily perform it. Automating diagnostic coding using Artificial Intelligence is seen as an essential solution in Hospital Information Management Systems and approaches built on Convolutional Neural Networks currently perform best. In this work, a neural model built on unstructured clinical text for enabling automatic diagnostic coding for given patient discharge summaries is proposed. We incorporate a structured self-attention mechanism designed to boost learning of label-specific vectors and the significant clinical text snippets associated with a certain label for this purpose. These vectors are then combined with a novel code description pipeline leveraging the descriptions provided for each standardized diagnostic code. The proposed model achieved best performance in terms of standard metrics over the MIMIC-III dataset, outperforming models based on Longformers and Knowledge graphs. Furthermore, to leverage structured clinical data to enhance the proposed model, and to enable improved diagnostic code prediction, model ensembling is considered. A neural model built on structured data by leveraging supervised machine learning algorithms such as random forest and boosting, is designed for multi-class code classification. Experimental results revealed that the proposed ensemble models show promising performance compared to traditional models that rely solely on unstructured or structured clinical data, emphasizing their suitability for real-world deployment.

Turn-table micro-CT scanner for dynamic perfusion imaging in mice: design, implementation, and evaluation

Objective.This study introduces a novel desktop micro-CT scanner designed for dynamic perfusion imaging in mice, aimed at enhancing preclinical imaging capabilities with high resolution and low radiation doses.Approach.The micro-CT system features a custom-built rotating table capable of both circular and helical scans, enabled by a small-bore slip ring for continuous rotation. Images were reconstructed with a temporal resolution of 3.125 s and an isotropic voxel size of 65µm, with potential for higher resolution scanning. The system's static performance was validated using standard quality assurance phantoms. Dynamic performance was assessed with a custom 3D-bioprinted tissue-mimetic phantom simulating single-compartment vascular flow. Flow measurements ranged from 1.51to 9 ml min-1, with perfusion metrics such as time-to-peak, mean transit time, and blood flow index calculated.In vivoexperiments involved mice with different genetic risk factors for Alzheimer's and cardiovascular diseases to showcase the system's capabilities for perfusion imaging.Main Results.The static performance validation confirmed that the system meets standard quality metrics, such as spatial resolution and uniformity. The dynamic evaluation with the 3D-bioprinted phantom demonstrated linearity in hemodynamic flow measurements and effective quantification of perfusion metrics.In vivoexperiments highlighted the system's potential to capture detailed perfusion maps of the brain, lungs, and kidneys. The observed differences in perfusion characteristics between genotypic mice illustrated the system's capability to detect physiological variations, though the small sample size precludes definitive conclusions.Significance.The turn-table micro-CT system represents a significant advancement in preclinical imaging, providing high-resolution, low-dose dynamic imaging for a range of biological and medical research applications. Future work will focus on improving temporal resolution, expanding spectral capabilities, and integrating deep learning techniques for enhanced image reconstruction and analysis.

Self-adjusted Nitrous Oxide During Urodynamic Studies Reduces Patient Pain Without Compromising Study Quality: A Randomized Controlled Trial

Background and objectiveA urodynamic study (UDS) is a routine clinic procedure that can cause significant discomfort for certain patients, with no satisfactory analgesic alternatives currently available. Our aim was to evaluate the effectiveness of low-dose self-adjusted nitrous oxide (SANO), titrated to the patient’s desired effect, on standard metrics for bladder function and on patient-reported pain and anxiety. MethodsWe conducted a single-institution, double-blind, randomized crossover trial in adults undergoing UDS. Each patient underwent two consecutive UDS runs, randomized to receive oxygen during the first run followed by SANO during the second run, or vice versa. UDS outcomes (capacity, detrusor strength, residual volume) and patient subjective outcomes (Visual Analog Scale for pain and anxiety, operator assessment of verbal feedback) were compared between the two runs. Secondary analyses were performed to compare outcomes during the first UDS run and adjust for treatment order. A paired Wilcoxon signed rank-sum test and McNemar’s χ2 test were used to compare continuous and categorical variables, respectively. Adverse events were recorded. Key findings and limitationsNineteen patients were randomized (10 to oxygen for the first run, 9 to SANO for the first run). UDS outcomes did not differ between the two arms. Patients reported significantly less pain during the SANO run than during the oxygen run (p = 0.046). Verbal feedback was significantly better with SANO (p = 0.001). Most patients (15/19, 79%) stated that they would prefer to receive SANO during future UDS. There were no significant complications. Conclusions and clinical implicationsSANO oxide is a safe and effective means of preserving standard adult UDS metrics while significantly reducing patient-reported pain. Patient summaryUrodynamic tests (UDS) for evaluation of lower urinary tract symptoms can cause discomfort and pain. Our study shows that nitrous oxide gas self-adjusted by patients had no effect on UDS test outcomes or on verbal feedback during the procedure, and reduced discomfort and pain in comparison to oxygen. Nitrous oxide may an attractive option for patients who are reluctant to undergo UDS.

Minimal Networks on Balls and Spheres for Almost Standard Metrics

We study the existence of minimal networks in the unit sphere Sd\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$${\ extbf{S}}^d$$\\end{document} and the unit ball Bd\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$${\ extbf{B}}^d$$\\end{document} of Rd\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$${\ extbf{R}}^d$$\\end{document} endowed with Riemannian metrics close to the standard ones. We employ a finite-dimensional reduction method, modelled on the configuration of θ\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\ heta $$\\end{document}-networks in Sd\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$${\ extbf{S}}^d$$\\end{document} and triods in Bd\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$${\ extbf{B}}^d$$\\end{document}, jointly with the Lusternik–Schnirelmann category.

Deep multimodal saliency parcellation of cerebellar pathways: Linking microstructure and individual function through explainable multitask learning.

Parcellation of human cerebellar pathways is essential for advancing our understanding of the human brain. Existing diffusion magnetic resonance imaging tractography parcellation methods have been successful in defining major cerebellar fibre tracts, while relying solely on fibre tract structure. However, each fibre tract may relay information related to multiple cognitive and motor functions of the cerebellum. Hence, it may be beneficial for parcellation to consider the potential importance of the fibre tracts for individual motor and cognitive functional performance measures. In this work, we propose a multimodal data-driven method for cerebellar pathway parcellation, which incorporates both measures of microstructure and connectivity, and measures of individual functional performance. Our method involves first training a multitask deep network to predict various cognitive and motor measures from a set of fibre tract structural features. The importance of each structural feature for predicting each functional measure is then computed, resulting in a set of structure-function saliency values that are clustered to parcellate cerebellar pathways. We refer to our method as Deep Multimodal Saliency Parcellation (DeepMSP), as it computes the saliency of structural measures for predicting cognitive and motor functional performance, with these saliencies being applied to the task of parcellation. Applying DeepMSP to a large-scale dataset from the Human Connectome Project Young Adult study (n = 1065), we found that it was feasible to identify multiple cerebellar pathway parcels with unique structure-function saliency patterns that were stable across training folds. We thoroughly experimented with all stages of the DeepMSP pipeline, including network selection, structure-function saliency representation, clustering algorithm, and cluster count. We found that a 1D convolutional neural network architecture and a transformer network architecture both performed comparably for the multitask prediction of endurance, strength, reading decoding, and vocabulary comprehension, with both architectures outperforming a fully connected network architecture. Quantitative experiments demonstrated that a proposed low-dimensional saliency representation with an explicit measure of motor versus cognitive category bias achieved the best parcellation results, while a parcel count of four was most successful according to standard cluster quality metrics. Our results suggested that motor and cognitive saliencies are distributed across the cerebellar white matter pathways. Inspection of the final k = 4 parcellation revealed that the highest-saliency parcel was most salient for the prediction of both motor and cognitive performance scores and included parts of the middle and superior cerebellar peduncles. Our proposed saliency-based parcellation framework, DeepMSP, enables multimodal, data-driven tractography parcellation. Through utilising both structural features and functional performance measures, this parcellation strategy may have the potential to enhance the study of structure-function relationships of the cerebellar pathways.

Spatial attention guided cGAN for improved salient object detection

Recent research shows that Conditional Generative Adversarial Networks (cGANs) are effective for Salient Object Detection (SOD), a challenging computer vision task that mimics the way human vision focuses on important parts of an image. However, implementing cGANs for this task has presented several complexities, including instability during training with skip connections, weak generators, and difficulty in capturing context information for challenging images. These challenges are particularly evident when dealing with input images containing small salient objects against complex backgrounds, underscoring the need for careful design and tuning of cGANs to ensure accurate segmentation and detection of salient objects. To address these issues, we propose an innovative method for SOD using a cGAN framework. Our method utilizes encoder-decoder framework as the generator component for cGAN, enhancing the feature extraction process and facilitating accurate segmentation of the salient objects. We incorporate Wasserstein-1 distance within the cGAN training process to improve the accuracy of finding the salient objects and stabilize the training process. Additionally, our enhanced model efficiently captures intricate saliency cues by leveraging the spatial attention gate with global average pooling and regularization. The introduction of global average pooling layers in the encoder and decoder paths enhances the network's global perception and fine-grained detail capture, while the channel attention mechanism, facilitated by dense layers, dynamically modulates feature maps to amplify saliency cues. The generated saliency maps are evaluated by the discriminator for authenticity and gives feedback to enhance the generator's ability to generate high-resolution saliency maps. By iteratively training the discriminator and generator networks, the model achieves improved results in finding the salient object. We trained and validated our model using large-scale benchmark datasets commonly used for salient object detection, namely DUTS, ECSSD, and DUT-OMRON. Our approach was evaluated using standard performance metrics on these datasets. Precision, recall, MAE and Fβ score metrics are used to evaluate performance. Our method achieved the lowest MAE values: 0.0292 on the ECSSD dataset, 0.033 on the DUTS-TE dataset, and 0.0439 on the challenging and complex DUT-OMRON dataset, compared to other state-of-the-art methods. Our proposed method demonstrates significant improvements in salient object detection, highlighting its potential benefits for real-life applications.

Identifying arguments within a text: Categorizing errors and their impact in arguments' relation prediction

The automatic identification of argument units within a text is a crucial task, as it is the first step that should be performed by an end-to-end argument mining system. In this work, we propose an approach for categorizing errors in predicted argument units, which allows the evaluation of segmentation models from an argumentative perspective. We assess the ability of several models to generalize knowledge across different text domains and, through the proposed categorization, we show differences in their behavior that may not be noticeable using standard classification metrics. Furthermore, we assess how the errors in predicted argument units impact on a task that rely on accurate unit identification, an aspect that has not been studied in previous research, and that helps to evaluate the usability of an imperfect segmentation model beyond the segmentation task itself.

Aerosol climatology, variability, and trends over the Indo-Gangetic Plain in CMIP6 models

We evaluate the performance of Coupled Model Intercomparison Project Phase 6 (CMIP6) models in simulating aerosol optical depth (AOD) climatology, variability, and trends over the Indo-Gangetic Plain (IGP) considered for the period of space-borne MODIS satellite observations during 2003–2014. The multi-model ensemble (MME) mean of CMIP6 models, is considered when better correlations (≥0.5) of models against the MODIS-derived AOD observations using standard statistical skill metrics. Analysis suggests that the CMIP6 MME successfully reproduces the spatial distribution of AOD climatology and its seasonal variability, albeit with some discrepancies. In particular, CMIP6 MME underestimates AOD spatial distributions across the IGP region (∼ bais varies between 0.2 and 0.4 during different seasons). CMIP6 MME captures coarse-mode aerosol distribution similar to AERONET over the IGP, while it has limited skill in capturing the fine-mode aerosol distribution, which could be due to uncertainties pertinent to anthropogenic aerosol emissions. Furthermore, the CMIP6 MME captures the seasonal evolution of aerosols over the IGP and its sub-regions (i.e., eastern and western IGP regions). Notably, the CMIP6 MME successfully captures the dominance among various aerosol species and their contributions to total AOD over the IGP. The CMIP6 MME mimic the observed AOD trends (varies between 0.01 and 0.03 year−1) well across the study regions, except the western IGP, where the MME is unable to replicate declining AOD trends during the pre-monsoon and monsoon seasons compared to MODIS. By conducting a principal component analysis on AOD data, it is apparent that the CMIP6 MME captures AOD variances that are comparable to MODIS observations, albeit with some discrepancies. This finding of the present research will help in policymaking, and mitigation strategies in the region.

Nearest neighbor future captioning: generating descriptions for possible collisions in object placement tasks

Domestic service robots (DSRs) that support people in everyday environments have been widely investigated. However, their ability to predict and describe future risks resulting from their own actions remains insufficient. In this study, we focus on the linguistic explainability of DSRs. Most existing methods do not explicitly model the region of possible collisions; thus, they do not properly generate descriptions of these regions. In this paper, we propose the Nearest Neighbor Future Captioning Model that introduces the Nearest Neighbor Language Model for future captioning of possible collisions, which enhances the model output with a nearest neighbors retrieval mechanism. Furthermore, we introduce the Collision Attention Module that attends regions of possible collisions, which enables our model to generate descriptions that adequately reflect the objects associated with possible collisions. To validate our method, we constructed a new dataset containing samples of collisions that can occur when a DSR places an object in a simulation environment. The experimental results demonstrated that our method outperformed baseline methods, based on the standard metrics. In particular, on CIDEr-D, the baseline method obtained 25.09 points, whereas our method obtained 33.08 points.