Version Of Dataset Research Articles

Exploring the links among brain iron accumulation, cognitive performance, and dietary intake in older adults: A longitudinal MRI study

This study evaluated longitudinal brain iron accumulation in older adults, its association with cognition, and the role of specific nutrients in mitigating iron accumulation. MRI-based, quantitative susceptibility mapping estimates of brain iron concentration were acquired from seventy-two healthy older adults (47 women, ages 60–86) at a baseline timepoint (TP1) and a follow-up timepoint (TP2) 2.5–3.0 years later. Dietary intake was evaluated at baseline using a validated questionnaire. Cognitive performance was assessed at TP2 using the uniform data set (Version 3) neuropsychological tests of episodic memory (MEM) and executive function (EF). Voxel-wise, linear mixed-effects models, adjusted for longitudinal gray matter volume alterations, age, and several non-dietary lifestyle factors revealed brain iron accumulation in multiple subcortical and cortical brain regions, which was negatively associated with both MEM and EF performance at T2. However, consumption of specific dietary nutrients at TP1 was associated with reduced brain iron accumulation. Our study provides a map of brain regions showing iron accumulation in older adults over a short 2.5-year follow-up and indicates that certain dietary nutrients may slow brain iron accumulation.

MMIFR: Multi-modal industry focused data repository

In the rapidly advancing field of industrial automation, the availability of robust and diverse datasets is crucial for the development and evaluation of machine learning models. The data repository consists of four distinct versions of datasets: MMIFR-D, MMIFR-FS, MMIFR-OD and MMIFR-P. The MMIFR-D dataset comprises a comprehensive assemblage of 5907 images accompanied by corresponding textual descriptions, notably facilitating the application of industrial equipment classification. In contrast, the MMIFR-FS dataset serves as an alternative variant characterized by the inclusion of 129 distinct classes and 5907 images, specifically catering to the task of few-shot learning within the industrial domain. MMIFR-OD is another alternative variant, comprising 8,839 annotation instances across 128 distinct categories, is predominantly utilized for object detection tasks. Additionally, the MMIFR-P dataset consists of 142 textual-visual information pairs, making it suitable for detecting pairs of industrial equipment. Furthermore, we conduct a comprehensive comparative analysis of our dataset in relation to other datasets used in industrial settings. Benchmark performances for different industrial tasks on our data repository are provided. The proposed multimodal dataset, MMIFR, can be utilized for research in industrial automation, quality control, safety monitoring, and other relevant domains.

Adaptive Bug Localization Framework for Precision-Driven Bug Localization in Software Engineering

Software development always looks for automated methods to improve productivity and accuracy in issue detection. The paper conducts a comparative examination of several machine-learning techniques to tackle the bug localization difficulty. Our study compared the performance of Logistic Regression (LR), Random Forest Classifier (RFC), Support Vector Machine (SVM), Gradient Boosting Classifier (GBC), and Adaptive Bug Localization System (ABLS) on five dataset versions. The results demonstrate the superior performance of ensemble learning methods. The ABLS model regularly beats other models regarding F1 score, accuracy, and recall, indicating its strong potential for precise problem localization. The study highlights the necessity of continuously adapting models to tackle idea drift in dynamic datasets. Our research suggests a path for future endeavours involving improving feature engineering and integrating real-time online learning to sustain high performance in bug localization activities.

Multi-Object Tracking in Agricultural Applications using a Vision Transformer for Spatial Association

This paper introduces a Multi-Object Tracking (MOT) framework for agricultural applications that estimates global positions in pixel coordinates using the local feature matching transformer — LoFTR. We design an efficient tracker that augments the capabilities of a state-of-the-art tracking algorithm by incorporating a novel association strategy based on spatial information of targets leaving and returning the camera field of view. We evaluate our framework using the publicly available LettuceMOT benchmark dataset and an adapted version of the AppleMOTS benchmark dataset that we denominate AppleMOT. Our experimental results demonstrate that our method outperforms cutting-edge algorithms for robotic plant tracking in the LettuceMOT dataset. The evaluation metrics show average improvements of up to 25% compared to the best publicly available results, demonstrating the benefits of our spatial association approach. For the AppleMOT dataset, we obtained bounding-box-based MOT evaluation metrics comparable to the segmentation-based (MOTS) counterparts presented in the original AppleMOTS paper. These findings highlight the effectiveness and potential of our approach in addressing the unique challenges posed by agricultural environments.

Presenting a Long-Term, Reprocessed Dataset of Global Sea Surface Temperature Produced Using the OSTIA System

Over the past few decades, the oceans have stored the majority of the excess heat in the climate system resulting from anthropogenic emissions. An accurate, long-term sea surface temperature (SST) dataset is essential for monitoring and researching the changes to the global oceans. A variety of SST datasets have been produced by various institutes over the years, and here, we present a new SST data record produced originally within the Copernicus Marine Environment Monitoring Service (which is therefore named CMEMS v2.0) and assess: (1) its accuracy compared to independent observations; (2) how it compares with the previous version (named CMEMS v1.2); and (3) its performance during two major volcanic eruptions. By comparing both versions of the CMEMS datasets using independent in situ observations, we show that both datasets are within the target accuracy of 0.1 K, but that CMEMS v2.0 is closer to the ground truth. The uncertainty fields generated by the two analyses were also compared, and CMEMS v2.0 was found to provide a more accurate estimate of its own uncertainties. Frequency and vector analysis of the SST fields determined that CMEMS v2.0 feature resolution and horizontal gradients were also superior, indicating that it resolved oceanic features with greater clarity. The behavior of the two analyses during two volcanic eruption events (Mt. Pinatubo and El Chichón) was examined. A comparison with the HadSST4 gridded in situ dataset suggested a cool bias in the CMEMS v2.0 dataset versus the v1.2 dataset following the Pinatubo eruption, although a comparison with sparser buoy-only observations yielded less clear results. No clear impact of the El Chichón eruption (which was a smaller event than Mt. Pinatubo) on CMEMS v2.0 was found. Overall, with the exception of a few specific and extreme events early in the time series, CMEMS v2.0 possesses high accuracy, resolution, and stability and is recommended to users.

Does Differentially Private Synthetic Data Lead to Synthetic Discoveries?

Synthetic data have been proposed as a solution for sharing anonymized versions of sensitive biomedical datasets. Ideally, synthetic data should preserve the structure and statistical properties of the original data, while protecting the privacy of the individual subjects. Differential Privacy (DP) is currently considered the gold standard approach for balancing this trade-off. The aim of this study is to investigate how trustworthy are group differences discovered by independent sample tests from DP-synthetic data. The evaluation is carried out in terms of the tests' Type I and Type II errors. With the former, we can quantify the tests' validity, i.e., whether the probability of false discoveries is indeed below the significance level, and the latter indicates the tests' power in making real discoveries. We evaluate the Mann-Whitney U test, Student's t-test, chi-squared test, and median test on DP-synthetic data. The private synthetic datasets are generated from real-world data, including a prostate cancer dataset (n = 500) and a cardiovascular dataset (n = 70,000), as well as on bivariate and multivariate simulated data. Five different DP-synthetic data generation methods are evaluated, including two basic DP histogram release methods and MWEM, Private-PGM, and DP GAN algorithms. A large portion of the evaluation results expressed dramatically inflated Type I errors, especially at levels of ϵ ≤ 1. This result calls for caution when releasing and analyzing DP-synthetic data: low p-values may be obtained in statistical tests simply as a byproduct of the noise added to protect privacy. A DP Smoothed Histogram-based synthetic data generation method was shown to produce valid Type I error for all privacy levels tested but required a large original dataset size and a modest privacy budget (ϵ ≥ 5) in order to have reasonable Type II error levels.

Risk Assessment and Diagnosis Code Prediction from Electronic Health Records (EHR) using deep learning

Knowledge in the medical field is often expressed by distinct and subjective norms. Much research and development efforts have focused on using deep learning algorithms to predict the likelihood of illnesses from EHR in recent years. When it comes to risk prediction, deep learning-based techniques outperform more conventional machine learning models. But nothing in the literature fully accounts for what doctors already know, such as the connections between illnesses and their risk factors. This research examines the use of Multi-layer Perceptron models for the categorisation of diagnoses in electronic health records. The raw data and a modified version of the EHR dataset are used to train two MLPs with distinct topologies. For comparative purposes, a Random Forest is used as a baseline. To phenotype patients using their electronic health records, we provide a deep learning method. Predictive modelling of chronic illnesses is the particular scenario used to verify the suggested model on a real-world EHR data warehouse. Many deep learning applications on EHRs have been effective, and there is still a lot of potential to be realised. It has been discovered that deep learning models can learn from the limited EHR dataset, but not to a level where they outperform the baseline model.

Effect of aerobic exercise program on neuropathic pain and quality of life in person with paraplegia: study protocol for a randomized controlled trial

BackgroundIndividuals with spinal cord injury (SCI) often suffer from neuropathic pain which is often disabling and negatively affects function, participation, and quality of life (QoL). Pharmacological treatments lack efficacy in neuropathic pain reduction hence studying alternatives to drug treatment is necessary. Preclinical evidence of various aerobic exercises has shown positive effects on neuropathic pain but scientific studies investigating its effect in the SCI human population are limited.MethodologyThis study is a double-blind, parallel, two-group, randomized controlled trial with an interventional study design that aims to evaluate the effectiveness of aerobic exercise program on neuropathic pain and quality of life (QoL) in individuals with chronic paraplegia. Thirty individuals with chronic paraplegia with the neurological level of injury from T2 to L2 will be recruited from the rehabilitation department at a super specialty hospital based on the inclusion criteria. Using a 1:1 allocation ratio, the participants will be randomly assigned to one of the two groups. The intervention group will perform high-intensity interval training (HIIT) aerobic exercise using an arm ergometer based on their peak heart rate, and the control group will perform free-hand arm aerobic exercise. In both groups, the intervention will be delivered as 30-min sessions, four times a week for 6 weeks.Outcome measuresInternational Spinal Cord Injury Pain Basic Data Set Version 3.0 will be used for diagnosing and assessing neuropathic pain and its interference with day-to-day activities, mood, and sleep. The International Spinal Cord Society (ISCoS) QoL basic data set will be used to assess QoL, and 6-min push test distance will be used to assess peak heart rate and aerobic capacity.DiscussionThe effectiveness of the aerobic exercise program will be assessed based on the changes in neuropathic pain score and its interference with day-to-day activities, mood, sleep, QoL, and aerobic capacity after 3 weeks mid-intervention and after 6 weeks post-intervention. The trial will provide new knowledge about the effectiveness of the aerobic exercise program in improving neuropathic pain and QoL in individuals with chronic paraplegia.Trial registrationClinical Trials Registry-India CTRI/2023/08/056257. Registered on 8 August 2023.

Hate speech detection in low-resourced Indian languages: An analysis of transformer-based monolingual and multilingual models with cross-lingual experiments

Abstract Warning: This paper is based on hate speech detection and may contain examples of abusive/ offensive phrases. Cyberbullying, online harassment, etc., via offensive comments are pervasive across different social media platforms like ™Twitter, ™Facebook, ™YouTube, etc. Hateful comments must be detected and eradicated to prevent harassment and violence on social media. In the Natural Language Processing (NLP) domain, the most prevalent task is comment classification, which is challenging, and language models based on transformers are at the forefront of this advancement. This paper intends to analyze the performance of language models based on transformers like BERT, ALBERT, RoBERTa, and DistilBERT on the Indian hate speech datasets over binary classification. Here, we utilize the existing datasets, i.e., HASOC (Hindi and Marathi) and HS-Bangla. So, we evaluate several multilingual language models like MuRIL-BERT, XLM-RoBERTa, etc., few monolingual language models like RoBERTa-Hindi, Maha-BERT (Marathi), Bangla-BERT (Bangla), Assamese-BERT (Assamese), etc., and perform cross-lingual experiment also. For further analyses, we perform multilingual, monolingual, and cross-lingual experiments on our Hate Speech Assamese (HS-Assamese) (Indo-Aryan language family) and Hate Speech Bodo (HS-Bodo) (Sino-Tibetan language family) dataset (HS dataset version 2) also and achieved a promising result. The motivation of the cross-lingual experiment is to encourage researchers to learn about the power of the transformer. Note that no pre-trained language models are currently available for Bodo or any other Sino-Tibetan languages.

A web-based mpox skin lesion detection system using state-of-the-art deep learning models considering racial diversity

The recent ‘Mpox’ outbreak, formerly known as ‘Monkeypox’, has been a significant public health concern and spread to over 110 countries globally. The challenge of clinically diagnosing mpox early on is due, in part, to its similarity to other types of rashes. Computer-aided screening tools have been proven valuable in cases where Polymerase Chain Reaction (PCR) based diagnosis is not immediately available. Deep learning methods excel at learning complex data representations, but their efficacy largely depends on adequate training data. To address this challenge, we present the “Mpox Skin Lesion Dataset Version 2.0 (MSLD v2.0)” as a follow-up to the previously released open-access dataset, one of the first to contain mpox lesion images. This dataset contains images of patients with mpox and five other non-mpox classes (chickenpox, measles, hand-foot-mouth disease, cowpox, and healthy). We benchmark the performance of several state-of-the-art deep learning models, including VGG16, ResNet50, DenseNet121, MobileNetV2, EfficientNetB3, InceptionV3, and Xception, to classify mpox and other infectious skin diseases. We leverage transfer learning implemented with pre-trained weights generated from the HAM10000 dataset, an extensive collection of pigmented skin lesion images. To reduce the impact of racial bias, we employ a color space data augmentation method to increase skin color variability during training and present a rigorous assessment of the technique’s effectiveness in skin lesion classification; we achieved the best overall accuracy of 83.59±2.11%. Finally, the developed models are incorporated within a prototype web application to analyze uploaded skin images by a user and determine whether a subject is a suspected mpox patient. We believe that the presented tool can be of significant impact in screening of mpox in case of a future outbreak.

CBAM VGG16: An efficient driver distraction classification using CBAM embedded VGG16 architecture

Driver monitoring systems (DMS) are crucial in autonomous driving systems (ADS) when users are concerned about driver/vehicle safety. In DMS, the significant influencing factor of driver/vehicle safety is the classification of driver distractions or activities. The driver’s distractions or activities convey meaningful information to the ADS, enhancing the driver/ vehicle safety in real-time vehicle driving. The classification of driver distraction or activity is challenging due to the unpredictable nature of human driving. This paper proposes a convolutional block attention module embedded in Visual Geometry Group (CBAM VGG16) deep learning architecture to improve the classification performance of driver distractions. The proposed CBAM VGG16 architecture is the hybrid network of the CBAM layer with conventional VGG16 network layers. Adding a CBAM layer into a traditional VGG16 architecture enhances the model’s feature extraction capacity and improves the driver distraction classification results. To validate the significant performance of our proposed CBAM VGG16 architecture, we tested our model on the American University in Cairo (AUC) distracted driver dataset version 2 (AUCD2) for cameras 1 and 2 images. Our experiment results show that the proposed CBAM VGG16 architecture achieved 98.65% classification accuracy for camera 1 and 97.85% for camera 2 AUCD2 datasets. The CBAM VGG16 architecture also compared the driver distraction classification performance with DenseNet121, Xception, MoblieNetV2, InceptionV3, and VGG16 architectures based on the proposed model’s accuracy, loss, precision, F1 score, recall, and confusion matrix. The drivers’ distraction classification results indicate that the proposed CBAM VGG16 has 3.7% classification improvements for AUCD2 camera 1 images and 5% for camera 2 images compared to the conventional VGG16 deep learning classification model. We also tested our proposed architecture with different hyperparameter values and estimated the optimal values for best driver distraction classification. The significance of data augmentation techniques for the data diversity performance of the CBAM VGG16 model is also validated in terms of overfitting scenarios. The Grad-CAM visualization of our proposed CBAM VGG16 architecture is also considered in our study, and the results show that VGG16 architecture without CBAM layers is less attentive to the essential parts of the driver distraction images. Furthermore, we tested the effective classification performance of our proposed CBAM VGG16 architecture with the number of model parameters, model size, various input image resolutions, cross-validation, Bayesian search optimization and different CBAM layers. The results indicate that CBAM layers in our proposed architecture enhance the classification performance of conventional VGG16 architecture and outperform the state-of-the-art deep learning architectures.

The impact of evolving synoptic weather patterns on multi-scale transport and sources of persistent high-concentration ozone pollution event in the Yangtze River Delta, China

High-concentration ozone pollution pose threats to ecosystems and human health. However, there is limited research on the impact of the alternating evolution of synoptic weather patterns (SWPs) on the multi-scale transport processes and sources of ozone. From June 14 to 18, 2018, a rare consecutive ozone pollution plagued in Hefei and broader Yangtze River Delta region (YRD). This study investigates the meteorological factors and sources using in-situ observational data and WRF-Chem model simulations. Analysis reveals a northeastern low-pressure system moving from north to south generated a cold front. This moving cold front facilitated the vertical transport of warm air masses carrying high-concentration ozone originating from North China. Subsequently, Ozone-rich air masses (ORMs) were transported over the YRD, influenced by the eastward movement of the Mongolian high-pressure system. Based on WRF-Chem model with NOx tagging mechanisms and WRF-FLEXPART backward simulations, it is confirmed that a notable atmospheric transport originated from North China region (NCR) to Hefei, especially on June 15. As the Mongolian high-pressure weakens and shifts east-southward, it carried ORMs generated by NOx emissions from the YRD, accumulating over the sea within the range of 120°E to 126°E and 25°N to 30°N. Both WRF-chem model results and TRopospheric Ozone and Precursors from Earth System Sounding (TROPESS) Chemistry Reanalysis dataset Version 2 (TCR-2) revealed the existence of ORMs in this geographic range. Subsequently, the ORMs carried out to sea by the weakened high-pressure system were reintroduced inland, influenced by southeast winds brought about by the peripheral circulation of typhoon “Gaemi”. In summary, the alternating evolution of SWPs significantly influences multi-scale ozone transport from both the NCR and the YRD regions, making substantial contributions to this prolonged episode. These findings offer valuable insights for improving regional ozone pollution prevention and control mechanisms.

B-TTDb: A Database of Turkish Tweets for Predicting the Top One Hundred Emojis

Emoji prediction is an important research task that focuses on finding the most appropriate emoji(s) quickly and effortlessly for a specific text. Now that Turkish is on the list of the top 20 most spoken languages in the world and there are a considerable number of Turkish-speaking social media users, studying emoji prediction in Turkish holds significant value. In this study, a Turkish tweets database, named Bitirim's Turkish Tweets Database (B-TTDb), was constructed for academic and industrial studies based on the prediction of the top 100 emojis. B-TTDb consists of four datasets. The first dataset includes raw tweets, the second dataset is the organized version of the first dataset, the third dataset is the pre-processed version of the second dataset, and the last one is the organized version of the third dataset. The last one is the final version and it is named Bitirim's Dataset (B-D). It includes a total of 158,201 unique tweets belonging to the top 100 emoji classes. For database validation, experiments were conducted on B-D with popular machine learning algorithms for the top 10, 20, 50, and 100 emojis. This study could be considered as the first study that contributes to the literature by the first validated large database of Turkish tweets that includes such a large number of emojis. In addition, B-TTDb could be a basis as well as motivation for various further studies.

Question Answering System Based on Bideirectional Long-Short-Term Memory (Bilstm)

In the modern world, Q&A systems are essential for promoting better communication between people and technology. These systems play an important role in collecting information quickly and efficiently, and this leads to great progress in learning, teaching and development in many areas of life. Using deep learning techniques, this research addresses the problem of excellent prediction of the questions that need to be answered. We created a question-and-answer system using "Bidirectional Long Short-Term Memory (BiLSTM)", a modern neural network known for its accuracy and results in text analysis and natural language understanding. This technique is more effective in understanding questions and producing very accurate answers because of its special ability to pay attention to preceding and following information in a sentence. Preprocessing was used to remove unnecessary, unimportant and time-consuming data. The "Stanford Question Answering Dataset version 2 (SQuAD 2.0)"was used, which is considered one of the important datasets used in the field of machine learning and natural language processing. The following evaluation metrics were used to evaluate the model’s performance: “Mean Average Precision(MAP), Mean Reciprocal Rank(MRR), Recall, Precision, Loss, F1 Score, and Exact Match (EM). The results, based on 150 epochs (EPOGs) and 128 batch sizes with a cleaned dataset split into 70% training and 30% test/validation (15% each), are as follows:"Precision (0.966), Loss (0.591), F1-score (0.966), Recall (0.967), EM (0.967), MRR(0.918), MAP (0.776), and accuracy(0.966)". Interestingly, the highest performance was observed when using the accuracy measure.

Attenuating majority attack class bias using hybrid deep learning based IDS framework

In real-time application domains, like finance, healthcare and defence, delay in service or stealing information may lead to unrecoverable consequences. So, early detection of intrusion is important to prevent security breaches. In recent days, anomaly-based intrusion detection using Hybrid Deep Learning approaches are becoming more popular. The most used benchmark datasets in the literature are NSL-KDD and UNSW-NB15, and these datasets are imbalanced. The models built on imbalanced datasets may lead to biased results towards majority classes by neglecting the minority class, even though they are equally important. In many cases, high accuracy is achieved for majority classes in the imbalanced datasets. But, the class-level performances are poor with respect to the minority class. The class balancing will also play an important role in attenuating the bias in prediction for imbalanced datasets. In this paper, a Hybrid Deep Learning Based Intrusion Detection (HDLBID) framework is proposed with CNN-BiLSTM combination. The four techniques, namely, Random Oversampling (ROS), ADASYN, SMOTE, and SMOTE-Tomek, are used for class balancing in the proposed HDLBID framework. The proposed HDLBID with SMOTE-Tomek achieves an overall accuracy of 99.6% with NSL-KDD and 89.02% for UNSW-NB15. It results in an improvement of 13.67% for NSL-KDD and 10.62% for UNSW-NB15 over the existing recent related models. In the proposed HDLBID, in addition to overall accuracy, the class-level F1 score is also calculated. A comparative study is presented to show the effectiveness of balancing dataset compared to imbalanced dataset, and observed that the SMOTE-Tomek class balancing comparatively performed well. An improvement of 37.43% is observed in the U2R class of the NSL-KDD dataset and 61.65% improvement is seen in the Worms class of the UNSW-NB15 dataset, both with SMOTE-Tomek class balancing. Therefore, the proposed HDLBID with SMOTE-Tomek class balancing reports the best results in terms of overall accuracy compared to existing recent related approaches. Also, in terms of class-level analysis, HDLBID reports best results with SMOTE-Tomek over imbalanced version of datasets.

Monocular Depth Estimation Based on Dilated Convolutions and Feature Fusion

Depth estimation represents a prevalent research focus within the realm of computer vision. Existing depth estimation methodologies utilizing LiDAR (Light Detection and Ranging) technology typically obtain sparse depth data and are associated with elevated hardware expenses. Multi-view image-matching techniques necessitate prior knowledge of camera intrinsic parameters and frequently encounter challenges such as depth inconsistency, loss of details, and the blurring of edges. To tackle these challenges, the present study introduces a monocular depth estimation approach based on an end-to-end convolutional neural network. Specifically, a DNET backbone has been developed, incorporating dilated convolution and feature fusion mechanisms within the network architecture. By integrating semantic information from various receptive fields and levels, the model’s capacity for feature extraction is augmented, thereby enhancing its sensitivity to nuanced depth variations within the image. Furthermore, we introduce a loss function optimization algorithm specifically designed to address class imbalance, thereby enhancing the overall predictive accuracy of the model. Training and validation conducted on the NYU Depth-v2 (New York University Depth Dataset Version 2) and KITTI (Karlsruhe Institute of Technology and Toyota Technological Institute) datasets demonstrate that our approach outperforms other algorithms in terms of various evaluation metrics.

Development of an early detection and automatic targeting system for cotton weeds using an improved lightweight YOLOv8 architecture on an edge device

Traditional means of weed removal, such as human work or the use of pesticides, frequently require significant amounts of effort, incur high expenses, and can negatively impact the environment. This study introduces a modified version of the YOLOv8 nano architecture that is suitable for running on edge devices for real-time applications. The proposed model uses an augmented version of the well-known CottonWeedDet12 dataset consisting of a total of 16,944 images with characteristic annotations to develop a model capable of correctly distinguishing 12 different cotton weed classes with an increased mean average precision of 97.6 % that is about 1.2 % more than the model trained using original, unaugmented dataset. The final selected model uses a convolutional block attention module (CBAM) and a unique C3Ghost block within the YOLOv8 backbone, which together increase the model's reliability for more accurate predictions with reduced computational complexity. Upon training with the augmented dataset, the proposed model with only 3.6 million parameters was able to achieve an mAP@50 score of 97.6 %, which surpasses all previous studies conducted using this dataset. Additionally, a high F1 score of 94.4 % proves that the model has a good balance between recall and precision. Class Activation Map (CAM) approaches such as EigenCAM, Grad-CAM++, and LayerCAM explainable AI (XAI) showed promising results for each of the customized models upon testing their interpretability for cotton weed detection. Furthermore, based on this model, a fast and cost-efficient targeting system was developed using a yaw-pitch mechanism for automatic weed tracking and herbicide spraying.

DNA barcoding is currently unreliable for species identification in most crayfishes.

DNA barcoding is commonly used for species identification. Despite this, there has not been a comprehensive assessment of the utility of DNA barcoding in crayfishes (Decapoda: Astacidea). Here we examined the extent to which local barcoding gaps (used for species identification) and global barcoding gaps (used for species discovery) exist among crayfishes, and whether global gaps met a previously suggested 10× threshold (mean interspecific difference being 10× larger than mean intra specific difference). We examined barcoding gaps using publicly available mitochondrial COI sequence data from the National Center for Biotechnology Information's nucleotide database. We created two versions of the COI datasets used for downstream analyses: one focused on the number of unique haplotypes (N H) per species, and another that focused on total number of sequences (N S; i.e., including redundant haplotypes) per species. A total of 81 species were included, with 58 species and five genera from the family Cambaridae and 23 species from three genera from the family Parastacidae. Local barcoding gaps were present in only 30 species (20 Cambaridae and 10 Parastacidae species). We detected global barcoding gaps in only four genera (Cambarus, Cherax, Euastacus, and Tenuibranchiurus), which were all below (4.2× to 5.2×) the previously suggested 10× threshold. We propose that a ~5× threshold would be a more appropriate working hypothesis for species discovery. While the N H and N S datasets yielded largely similar results, there were some discrepant inferences. To understand why some species lacked a local barcoding gap, we performed species delimitation analyses for each genus using the N H dataset. These results suggest that current taxonomy in crayfishes may be inadequate for the majority of examined species, and that even species with local barcoding gaps present may be in need of taxonomic revisions. Currently, the utility of DNA barcoding for species identification and discovery in crayfish is quite limited, and caution should be exercised when mitochondrial-based approaches are used in place of taxonomic expertise. Assessment of the evidence for local and global barcoding gaps is important for understanding the reliability of molecular species identification and discovery, but outcomes are dependent on the current state of taxonomy. As this improves (e.g., via resolving species complexes, possibly elevating some subspecies to the species-level status, and redressing specimen misidentifications in natural history and other collections), so too will the utility of DNA barcoding.

ROCOv2: Radiology Objects in COntext Version 2, an Updated Multimodal Image Dataset

Automated medical image analysis systems often require large amounts of training data with high quality labels, which are difficult and time consuming to generate. This paper introduces Radiology Object in COntext version 2 (ROCOv2), a multimodal dataset consisting of radiological images and associated medical concepts and captions extracted from the PMC Open Access subset. It is an updated version of the ROCO dataset published in 2018, and adds 35,705 new images added to PMC since 2018. It further provides manually curated concepts for imaging modalities with additional anatomical and directional concepts for X-rays. The dataset consists of 79,789 images and has been used, with minor modifications, in the concept detection and caption prediction tasks of ImageCLEFmedical Caption 2023. The dataset is suitable for training image annotation models based on image-caption pairs, or for multi-label image classification using Unified Medical Language System (UMLS) concepts provided with each image. In addition, it can serve for pre-training of medical domain models, and evaluation of deep learning models for multi-task learning.

An accurate transformer-based model for transition-based dependency parsing of free word order languages

Transformer models are the state-of-the-art in Natural Language Processing (NLP) and the core of the Large Language Models (LLMs). We propose a transformer-based model for transition-based dependency parsing of free word order languages. We have performed experiments on five treebanks from the Universal Dependencies (UD) dataset version 2.12. Our experiments show that a transformer model, trained with the dynamic word embeddings performs better than a multilayer perceptron trained on the state-of-the-art static word embeddings even if the dynamic word embeddings have a vocabulary size ten times smaller than the static word embeddings. The results show that the transformer trained on dynamic word embeddings achieves an unlabeled attachment score (UAS) of 84.17% for Urdu language which is ≈3.6% and ≈1.9% higher than the UAS scores of 80.56857% and 82.26859% achieved by the multilayer perceptron (MLP) using two static state-of-the-art word embeddings. The proposed approach is investigated for Arabic, Persian and Uyghur languages, in addition to Urdu, for UAS scores and the results suggest that the proposed solution outperform the MLP-based approaches.