Classification Method Research Articles

Approaches and processes for paediatric chest X-ray classification used in the SHINE TB treatment-shortening trial.

<sec><title>INTRODUCTION</title>SHINE (Shorter Treatment for Minimal Tuberculosis in Children) was the first Phase 3 paediatric TB treatment-shortening trial. Robust chest X-ray (CXR) classification methods were integral to excluding severe disease for trial eligibility and to retrospectively adjudicating TB status at baseline. We describe and critically evaluate the CXR classification approaches and processes used in the SHINE trial.</sec><sec><title>METHODS</title>Children with non-severe TB were randomised to 4- vs 6-months anti-TB treatment. Radiologically non-severe TB was defined on CXR. CXRs were systematically interpreted by on-site clinicians prospectively for eligibility determination and retrospectively by experts to inform adjudication of baseline TB status and disease severity.</sec><sec><title>RESULTS</title>A screening CXR was successfully obtained from all 1,204 enrolled children; 1,134 CXRs from children with intra-thoracic TB were reviewed by expert readers. Compared with the expert panel, enrolling clinicians classified more CXRs as abnormal and 'typical TB' and all as radiologically non-severe. The expert panel retrospectively classified 71/1,134 (6%) CXRs as severe. Of these, 4 (5.6%) had unfavourable outcomes compared with 34 (3.0%) in the trial overall.</sec><sec><title>DISCUSSION</title>Using CXRs to classify radiological disease severity and inform eligibility decisions in real-time by local enrolling clinicians was feasible and safe in this large paediatric TB trial. Retrospective central expert CXR review was successful. Refinement of the CXR methods for the classification of both disease severity and TB status could support standardised implementation in routine care and research.</sec>.

Roodmus: a toolkit for benchmarking heterogeneous electron cryo-microscopy reconstructions.

Conformational heterogeneity of biological macromolecules is a challenge in single-particle averaging (SPA). Current standard practice is to employ classification and filtering methods that may allow a discrete number of conformational states to be reconstructed. However, the conformation space accessible to these molecules is continuous and, therefore, explored incompletely by a small number of discrete classes. Recently developed heterogeneous reconstruction algorithms (HRAs) to analyse continuous heterogeneity rely on machine-learning methods that employ low-dimensional latent space representations. The non-linear nature of many of these methods poses a challenge to their validation and interpretation and to identifying functionally relevant conformational trajectories. These methods would benefit from in-depth benchmarking using high-quality synthetic data and concomitant ground truth information. We present a framework for the simulation and subsequent analysis with respect to the ground truth of cryo-EM micrographs containing particles whose conformational heterogeneity is sourced from molecular dynamics simulations. These synthetic data can be processed as if they were experimental data, allowing aspects of standard SPA workflows as well as heterogeneous reconstruction methods to be compared with known ground truth using available utilities. The simulation and analysis of several such datasets are demonstrated and an initial investigation into HRAs is presented.

CLIP-FSSC: A Transferable Visual Model for Fish and Shrimp Species Classification Based on Natural Language Supervision

Fish and shrimp species classification is a practical need in the field of aquaculture. Traditional classification method includes extracting modal features of the single image and training them on downstream datasets. However, this method has the disadvantage of requiring manual annotated image data and significant training time. To address these issues, this paper introduces a method named CLIP-FSSC (Contrastive Language–Image Pre-training for Fish and Shrimp Species Classification) for zero-shot prediction using a pre-trained model. The proposed method aims to classify fish and shrimp species in the field of aquaculture using a multimodal pre-trained model that utilizes semantic text description as an image supervision signal for transfer learning. In the downstream fish dataset, we use natural language labels for three types of fish - grass carp, common carp, and silver carp. We extract text category features using a transformer and compare the results obtained from three different CLIP-based backbones for the image modality - vision transformer, Resnet50, and Resnet101. We compare the performance of these models with previous methods that performed well. After performing zero-shot predictions on samples of the three types of fish, we achieve similar or even better classification accuracy than models trained on downstream fish datasets. Our experiment results show an accuracy of 98.77%, and no new training process is required. This proves that using the semantic text modality as the label for the image modality can effectively classify fish species. To demonstrate the effectiveness of this method on other species in the field of aquaculture, we collected two sets of shrimp data - prawn and cambarus. Through zero-shot prediction, we achieve the highest classification accuracy of 92.00% for these two types of shrimp datasets. Overall, our results demonstrate that using a multimodal pre-trained model with semantic text description as an image supervision signal for transfer learning can effectively classify fish and shrimp species with high accuracy, while reducing the need for manual annotation and training time.

Evaluation of Secondary Buckling and Ultimate Strength Behaviour According to Boundary Conditions of Curved Plate Subjected to Compressive Load

This paper analyzed the buckling and ultimate strength characteristics of plate members with curvature, which are essentially used in ships and marine structures, through numerical analysis. Major structural members are affected by rotational velocity depending on the rigidity of structures placed around them, and it is very important to check how this affects buckling and ultimate strength. The aspect ratio of the curved plate was adopted as 2.5, 3.0, and 3.5 to reflect the plate thickness range and high-strength steel characteristics actually used. In the case of no curvature, and in the case of curvature, 5 degrees, 10 degrees, 15 degrees, 20 degrees, 30 degrees, and 45 degrees were considered. Under certain thickness and curvature conditions, secondary buckling occurs and the buckling state of the plate changes rapidly. Therefore, the final strength is evaluated lower than in the case where there is no curvature. Buckling and ultimate strength were high in a clamped condition around the curved plate, and there was no specificity in behavior depending on the aspect ratio change. In the four-side simply supported condition, the in-plane stiffness gradient changes depending on the aspect ratio due to the secondary buckling effect, and this characteristic is a phenomenon that is not considered at all in the existing classification method.

E-commerce clothing review analysis by advanced ML Algorithms

A significant aspect in the rapid ascent of Bangladesh's e-commerce sector in recent years has been the significance of consumer evaluations. By examining these reviews, readers can gain enhanced insight into consumer happiness and product quality. This research analyses the sentiment of online clothes reviews with advanced machine-learning techniques. The fundamental purpose of evaluating several machine learning models, such as KNN, RF, XGB, Multi, LSTM, and CNN, is to identify the most effective method for sentiment classification. Following the compilation of an extensive dataset of clothing assessments and the execution of data preparation tasks, including cleaning, tokenization, and stop word elimination, we utilised a combination of deep learning and machine learning models for classification purposes. KNN had the highest accuracy in forecasting the future attitudes of the models evaluated. The findings indicate in the 5000 datasets that KNN is the most effective algorithm for assessing the sentiment of online purchasing reviews. KNN reached at 0.91% Accuracy. This project's automated and scalable methodology may enable online businesses to evaluate client feedback and make more informed decisions. Future research initiatives encompass augmenting the dataset, experimenting with various methods, and incorporating these models into AI-driven mobile and online applications for real-time sentiment analysis.

Quantitative characterisation of debris cloud-induced pitting damage in AL-Whipple shields of spacecraft using infrared thermography

ABSTRACT Quantitatively evaluating the debris cloud-induced pitting damage in the AL-Whipple structure of spacecraft, is still a challenge, let alone microstructural plastic deformation usually accompanied by geometric cratering defects, which may decrease material stress-carrying capacity and structural life. With this motivation, a dedicated modelling technique is proposed to gain an insight into various modalities (i.e., geometric and plastic defects) of pitting damage-induced thermal modulation for infrared inspection. Then, a sequence of infrared images is obtained and analysed using the gap smoothing algorithm and naive Bayes classification methods. On this basis, a quantitative characterization framework is proposed to characterize the damage modalities and their morphologies. This method is validated using a 3D numerical model which contains pitting damage acquired from the real-world structure. Modality classification and morphology evaluation results are well consistent with the artificial defects, in terms of the modality, diameter, and depth. Then they‘re experimentally corroborated, in which the severity and distribution of pitting damage are precisely characterized, in terms of the crater number and morphologies, as well as the size of plastic regions around these craters, which are consistent with the actual observations. This work proposes a non-destructive strategy for pitting damage evaluation, which is engineering-friendly.

BiodiViz: Leveraging NER and RE for Automated Knowledge Graph Generation in Biodiversity Research

In biodiversity research, the integration of machine learning and data visualization is increasingly important for uncovering valuable insights from academic literature. This study introduces an innovative knowledge graph application, BiodiViz, designed to translate intricate text into intuitive visual representations, fostering a deeper comprehension of biodiversity relationships. BiodiViz uses the top-performing Named Entity Recognition (NER) and Relation Extraction (RE) models to automatically generate a comprehensive knowledge graph for biodiversity research. The NER model extracts and categorizes entities like organisms, phenomena, and habitats, while the RE model identifies relationships such as "have," "occur in," and "influence" from the BiodivNERE dataset (Abdelmageed et al. 2022). These entities and relationships are organized into nodes and edges within a graph. Researchers input text into BiodiViz, producing a visual knowledge graph that simplifies the analysis of complex biodiversity data, reducing manual effort and enhancing efficiency. Named Entity Recognition &amp; Relation Extraction BiodiViz leverages advanced Bidirectional Encoder Representations from Transformers (BERT)-based Large Language Models (LLMs) (Rogers et al. 2020), fine-tuned specifically for NER and RE tasks using the BiodivNERE dataset. The fine-tuning process involved various models, including BERT (Devlin et al. 2019), ELECTRA (Clark et al. 2020), and BiodivBERT (Abdelmageed et al. 2023). These models were evaluated for performance using the results of their F1-score as the main metric, which is the harmonic mean of precision (the proportion of true positive results among all positive predictions) and recall (the proportion of true positive results among all actual positives), with BiodivBERT achieving an F1-score of 77.16% for the NER task, while BERT excelled in the RE task with an F1-score of 81.28%. Rigorous hyperparameter optimization further enhanced the performance of BiodivBERT in the RE task by 3.38%. The BiodivNERE corpora by Abdelmageed et al. (2022) were used to fine-tune several models for NER and RE tasks in the biodiversity domain. The first corpus from the BiodivNERE corpora is BiodivNER, which is a gold standard dataset (manually labelled test corpora) for evaluating NER tasks. The fine-tuning process employed the token classification method from the Hugging Face library (Hugging Face 2023b), which assigns labels to each token in a sequence. Experiments were conducted with a batch size of four, meaning the model processes four examples/rows of data at a time before making an update to improve its learning. This is due to the constraints of the NVIDIA® GeForce RTX™ 3060 graphics processor. (NVIDIA 2024) Model performance was evaluated using the seqeval library (Nakayama 2018), focusing on accuracy, precision, recall, and F1 scores. For text classification, the second corpus, BiodivRE, was utilized, following previous research recommendations to explore fine-tuning settings for BiodivBERT. Hyperparameter optimization (Feurer and Hutter 2019) was conducted using Hugging Face’s Trainer API with an Optuna backend (Hugging Face 2023a), concentrating on learning rate and the number of training epochs (i.e., the number of complete passes through the entire dataset during model training). The BiodiViz Knowledge Graph Application The fine-tuned NER and RE models with the best F1-scores—BiodivBERT and BERT, respectively—were integrated into the knowledge graph application. Fig. 1 illustrates the flowchart of the application pipeline. Each sentence in the input text will go through the NER model to identify and label the entities within the sentence. Subsequently, these labeled entities, together with the original sentence, will be input into the RE model. The RE model will analyze every pair of entities for a potential relation and output the type of relation they share. The application will then utilize this data to create a graph with appropriate labels and color-coding. An example of the application's user interface with the knowledge graph is shown in Fig. 2. This study highlights the practical application of machine learning and data visualization in advancing biodiversity research, emphasizing the importance of developing user-friendly tools to support scientific exploration and discovery. The BiodiViz application, including the code and resources, is available on GitHub*1, providing an accessible tool for biodiversity researchers to streamline their analyses.

Prediction of grain shoal reservoirs via seismic forward modeling and waveform classification: Application to the northern slope of the central Sichuan Uplift, Sichuan Basin, SW China

Grain shoals are one of the principal reservoir facies of carbonate rocks in oil and gas exploration domains. With the discovery of oil and gas in the grain shoal reservoirs of the Longwangmiao Formation in the central Sichuan paleo-uplift, the characteristics and distribution of grain shoal reservoirs are still unclear in the northern slope of the central Sichuan paleo-uplift. Through integrated analysis of wireline logs and high-quality 3D seismic data, combined with seismic forward modeling and waveform classification, reservoir seismic facies, reservoir classification, and the reservoir distribution of different grades are investigated. The results show that (1) under the thin thickness (i.e., the thickness is around 80 m) of the Longwangmiao Formation (i) when reservoirs are not developed, the seismic waveform is manifested as a single peak, and the top interface of the Longwangmiao Formation corresponds to 1/8 [Formula: see text] above the single peak. (ii) When a single set of reservoirs is merely developed in the second member of the Longwangmiao Formation (SMLF), the waveform displays a single peak, and the top peak moves downward. In addition, when multiple sets of reservoirs are developed in the SMLF, the seismic reflections of multiple reservoirs are consistent with those of a single reservoir with the same cumulative thickness. (iii) When the reservoirs are developed in the first member of the Longwangmiao Formation (FMLF) and SMLF, the single peak moves downward, and either the reservoir thickness of the SMLF is greater than that of the FMLF or the reservoir velocity of the SMLF is lower than that of the FMLF. (2) Under the thick thickness (i.e., the thickness is approximately 110 m) of the Longwangmiao Formation, (i) when reservoirs are not developed, the waveform of the Longwangmiao Formation exhibits a double peak, and the top interface of the Longwangmiao Formation corresponds to the extreme value of the upper peak. (ii) When only a single set of reservoirs or multiple sets of reservoirs are developed in the SMLF, the Longwangmiao Formation exhibits a single peak, and the top interface of the Longwangmiao Formation corresponds to the trough. (iii) With the reservoir development in the first and second members of the Longwangmiao Formation, the waveform of the Longwangmiao Formation shows an asymmetric low-frequency peak, and either the reservoir thickness of the SMLF is greater than that of the FMLF or the reservoir velocity of the SMLF is lower than that of the FMLF. (3) Using the waveform classification method, seismic waveforms corresponding to the Longwangmiao Formation could be summarized into four types (i.e., two types of waveforms associated with reservoir development and two types of waveforms associated with reservoir un-development). Two reservoir waveforms are distributed as two strips oriented approximately east–west and spread in the central study area, according to the planar study results that combine seismic attributes and waveform classification. The reservoir stripes near well PS1 represent the better-developed reservoir zones in the study area.

Single-Channel Sleep EEG Classification Method Based on LSTM and Hidden Markov Model

Background: The single-channel sleep EEG has the advantages of convenient collection, high-cost performance, and easy daily use, and it has been widely used in the classification of sleep stages. Methods: This paper proposes a single-channel sleep EEG classification method based on long short-term memory and a hidden Markov model (LSTM-HMM). First, the single-channel EEG is decomposed using wavelet transform (WT), and multi-domain features are extracted from the component signals to characterize the EEG characteristics fully. Considering the temporal nature of sleep stage changes, this paper uses a multi-step time series as the input for the model. After that, the multi-step time series features are input into the LSTM. Finally, the HMM improves the classification results, and the final prediction results are obtained. Results: A complete experiment was conducted on the Sleep-EDFx dataset. The results show that the proposed method can extract deep information from EEG and make full use of the sleep stage transition rule. The proposed method shows the best performance in single-channel sleep EEG classification; the accuracy, macro average F1 score, and kappa are 82.71%, 0.75, and 0.76, respectively. Conclusions: The proposed method can realize single-channel sleep EEG classification and provide a reference for other EEG classifications.

Serum proteomic profiling reveals potential inflammatory biomarkers in long-COVID patients: a comparative analysis with healthy controls

Abstract Background Long-COVID (also known as post-acute sequelae of SARS-CoV-2 infection, PASC), is defined as the persistence, or the new onset, of symptoms after recovery from the acute phase of COVID-19. It consists of a multiorgan syndrome expressing itself with a wide variety of cardiovascular and non-cardiovascular symptoms such us chest pain, fatigue, shortness of breath, autonomic dysfunction, cognitive impairment and sleep disorders even in the absence of a clear evidence of organ dysfunction. Some previous studies have suggested the potential role of an abnormal inflammatory response after SARS-CoV-2 infection, however the specific underlying mechanisms remain unclear. In this study we performed a proteomic analysis of the serum of 80 long-COVID patients to identify potential biomarkers that may act as indicators or therapeutic targets for long COVID. Methods and results We compared the levels of the proteins in both long-COVID group and a control group using multivariate and univariate analysis. Both PLS-DA and PCA analysis showed a clear distinction between the two groups; clustering analysis also clearly separated control from long-COVID patients. Proteins were then filtered based on (a) statistical analysis, considering proteins with a VIP&amp;gt;1 (PLSDA) and p&amp;lt;0.05 (t-test); (b) coefficient of variation (CV), where only proteins with CV&amp;lt;10 were considered; (c) and fold change (FC), for which proteins with FC&amp;lt;0.67 or FC&amp;gt;1.5 were selected for further analysis. The selected proteins were then subjected to a ROC curve analysis to identify antithrombin as a potential biomarker with an area under the curve (AUC) near 1. Other proteins, displaying an AUC &amp;gt; 0.98, were also identified as potential biomarkers (Sirtuin 1, NatriureticPeptide B, Hemopexin Arachidonate 5-Lipoxygenase). In addition, a multivariate ROC method, using linear SVM as classification method, identified that a combination of ten proteins would also result in an AUC near 1. Finally, a correlation analysis between the proteins of interest and the clinical manifestations was performed. A positive correlation (above 0.5) was found between dyspnea and Glucocorticoid Receptor and between memory deficit and Antithrombin. Arachidonate 5-Lipoxygenase was negatively correlated (below -0.5) with tremors and hair loss, MannoseBinding Lectin 2 was found to be positively linked to asthenia, Antithrombin with concentration deficit and Endothelin-2 with muscles aches. Conclusions The identified biomarkers are associated with inflammatory processes, corroborating literature evidence that long-COVID patients develop an inflammatory state that damages many tissues. Further studies are needed to validate this data in larger cohorts to identify specific biomarkers, and guide future preventive strategies or treatments to address long COVID and its cardiovascular and non-cardiovascular sequelae.

Cause and Characteristics of Changes in Mesoscale Convective Systems within a Convection-Permitting Regional Climate Model

Abstract Changes in the frequency and intensity of mesoscale convective systems (MCS) are assessed using convection-permitting regional climate model simulations. We present a novel classification method that relates MCSs to the magnitude of their large-scale forcing environments to better understand the changing nature of MCSs across different forcing environments in a possible future climate scenario. Overall, the annual frequency, intensity, and amount of precipitation associated with MCSs are projected to increase for broad portions of the eastern conterminous United States (CONUS). Furthermore, changes in the characteristics of MCSs show larger, longer-lived, and faster MCSs particularly over the Midwest and Southeast. Seasonal examination of this response reveals a robust intensification of March–May (MAM) MCSs. The higher frequency of MAM MCSs are found to occur in weakly forced synoptic environments. Increased rainfall amounts are explained by an intensification of MCSs due to changing thermodynamics and enhanced lower-tropospheric moisture transport into the central CONUS by the North Atlantic Subtropical High. Differences in mid-latitude storm tracks are favorable towards the enhanced development of MCSs associated with strong baroclinic forcing within the Midwest and Northern Plains, but are only realized in the presence of strong changes in lower-tropospheric moisture transport. Overall, these results suggest a shift in the behavior of MAM MCSs that more closely corresponds to the behavior of June–August (JJA) MCSs in the historical climate. The increased occurrence of MCSs in weakly forced environments, particularly in MAM, deserves further investigation.

Automatic categorization of medical documents in Afaan Oromo using ensemble machine learning techniques

Automatic medical document classification using machine learning techniques can enhance the productivity of healthcare services by reducing processing time and cost. This work proposes an ensemble learning approach to develop a model that classifies electronic medical documents in Afaan Oromo. The main tasks in this work are preparing the corpus, pre-processing, training the models, and the classification process. We used the term frequency-inverse document frequency (TF-IDF) and bag of words (BOW) feature extraction methods. An ensemble technique in this work is that it creates multiple individual classifier predictions from naïve Bayes, random forest, SVM, and logistic regression and then combines them to advance a reliable and more accurate classifier. Evaluation measures were employed using accuracy, F1-score, recall, and precision for performance comparison. The efficiency of the proposed method is compared with the two existing boosting approaches, namely gradient boosting and adaboost. The experimental result shows the efficiency of BOW feature extraction over TF-IDF in this work on our dataset. These results also illustrated the effectiveness of the proposed model by scoring 94.81% accuracy and 94.84% F1-score. This work significantly contributes to the technological enhancement of service delivery, managing documents through classification methods, and advancing the data processing systems in healthcare sectors.

Deep learning approach to predict autism spectrum disorder: a systematic review and meta-analysis

BackgroundThe use of the deep learning (DL) approach has been suggested or applied to identify childhood autism spectrum disorder (ASD). The capacity to predict ASD, however, differs across investigations. Our study’s objective was to conduct a meta-analysis to determine the DL for ASD in children’s classification accuracy.MethodsEligibility criteria were designed according to the purpose of the meta-analysis; PubMed, EMBASE, Cochrane Library, and Web of Science Database were searched for articles published up to April 16, 2023, on the accuracy of DL methods for ASD classification. Using the Revised Tool for the Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) to assess the quality of the included studies. Sensitivity, specificity, areas under the curve (AUC), summary receiver operating characteristic (SROC), and corresponding 95% confidence intervals (CIs) were compiled by using the bivariate random-effects models.ResultsA total of 11 predictive trials based on DL models were included, involving 9495 ASD patients from 6 different databases. According to bivariate random-effects models’ results, the overall sensitivity, specificity, and AUC of the DL technique for ASD were, 0.95 (95% CI = 0.88–0.98), 0.93 (95% CI = 0.85–0.97), and 0.98 (95%CI: 0.97–0.99), respectively. Subgroup analysis results found that different datasets did not cause heterogeneity (meta-regression P = 0.55). The Kaggle dataset’s sensitivity and specificity were 0.94 (95%CI: 0.82-1.00) and 0.91 (95%CI: 0.76-1.00), and with 0.97 (95%CI: 0.92-1.00) and 0.97 (95%CI: 0.92-1.00) for ABIDE dataset.ConclusionsDL techniques has satisfactory sensitivity, specificity, and AUC in ASD classification. However, the major heterogeneity of the included studies limited the effectiveness of this meta-analysis. Further trials need to be performed to demonstrate the clinical practicability of DL diagnosis.

Deep learning and Diagnostic value of cardiac cine MRI for assessing patent foramen ovale in patients with cerebral disorders compared with cTCD

Abstract Purpose To investigate the clinical feasibility of cardiac MRI (CMRI) cine technology using deeplearning without contrast for diagnosis and classification of patent foramen ovales (PFOs) in patients with cerebral disorders. Materials and Methods Fourty-four patients (24 males and 20 females; mean age, 41.3 years; range, 21-64 years) with cerebral disorders underwent both contrast transcranial doppler sonography (cTCD) and non-enhanced CMRI examinations between October 2019 and March 2020. and datas from 335 patients were collected by 2023.CMRI was performed with a 3.0T MR scanner by using OBL FIESTA CINE 4CH sequence. The scanning direction was perpendicular to the interatrial septum (IAS) slice. The obtained MR images were analyzed by AW station 4.7 and the pseudo-color coding was also performed according to the different phases. The condition of the blood shunt was observed and recorded by using deep learning, including measuring the length and width of the PFO and observing whether it was complicated with IAS aneurysm or secondary septum thickening. Results Thirty-nine of 44 patients with cerebral disorders were confirmed with positive RLS by cTCD test, and 37 of the 39 patients were diagnosed with a PFO at CMRI. Two of 5 patients with negative cTCD test were also diagnosed with a PFO at CMRI. Compared with cTCD as a standard, CMRI had a sensitivity of 94.9%, specificity of 60.0%, accuracy of 90.9%, positive predictive value of 94.9%, negative predictive value of 60.0%, and AUC of 0.774. Based on the pseudo-color coding, the right-to-left color jet was observed in 34 patients, and the two-way shunting was found in 5 patients. The IAS aneurysm and secondary septum thickening were found in 5 and 3 patients (11.4% and 6.8%). The maximum diameters of PFOs ranged from 1.7 mm to 16.8 mm and the mean diameter was 5.4±3.4 mm. Conclusion By using deep learning,CMRI cine technology without contrast provided a noninvasive and excellent method for PFO identification, evaluation and classification, with both high sensibility (92.85%) and concordance (90.9%) when compared to cTCD.PFO types

Left ventricular hypertrophy, wall thickness, and histological myocyte alteration in end-stage heart failure: a call for a standard definition

Abstract Objective This study aimed to examine the relationship between different left ventricular hypertrophy (LVH) classification criteria according to the American Society of Echocardiography (ASE), and anatomical measurement of LV wall thickness (LVWT) and histological evidence of myocyte hypertrophy. Methods Data from 146 consecutive patients (47 years; 75% were male) who have undergone heart transplantation were analysed. Fixed ventricles were dissected and measured in weight after both atria, great vessels, and epicardial fat were removed. The fixed biventricular weight was calculated as 87% of total weight (the effect of fixed formalin reduced the total weight by 3%). Anatomical LVWT was measured with a ruler at LV free wall where maximal thickness is located. A comprehensive echocardiogram was performed in all patients before heart transplantation. LVM and relative wall thickness (RWT) (2 PWT/LVEDD) were measured by M-mode echocardiography in parasternal long-axis view. Patients were considered to have LVH when the LVM index exceeds 95 g/m2 in women and exceeds 115 g/m2 in men. Patients were considered to have concentric LVH when RWT &amp;gt; 0.42. LVWT was measured using 2D echocardiography perpendicular to the LV long axis. Results Of 146 explanted hearts, dilated cardiomyopathy (CM), cardiac amyloid, and hypertrophic CM were diagnosed in 53 (47%), 1 (1%), and 8 (5%) patients, respectively. Anatomic ventricular mass correlated with echocardiographic LVM (r=0.5, p &amp;lt; 0.01). Anatomic LVWT modestly correlated with echocardiographic LVWT (r=0.2, p=0.01). Of 146 patients, echocardiographic normal LV geometry, concentric LV remodelling, concentric LVH, and eccentric LVH were present in 22 (15%), 6 (4%), 14 (10%), and 104 (71%), respectively. Increased LVWT (≥ 13 mm) assessed by 2-D echocardiography was present in normal LV geometry (5%), concentric LV remodelling (5%), concentric LVH (57%), and eccentric LVH (33%) (p &amp;lt;0.001) in those classified by ASE guidelines. Anatomic LVWT (≥ 13 mm) was present in normal LV geometry (14%), concentric LV remodelling 5%), concentric LVH (14%), and eccentric LVH (80%) (p &amp;lt;0.001) in those classified by ASE guidelines. Histological myocyte hypertrophy was present in normal LV geometry (12%), concentric LV remodelling (2%), concentric LVH (11%), and eccentric LVH (70%) (p &amp;lt;0.001) in those classified by ASE guidelines. Conclusions 2D-derived echocardiographic increased LVWT in patients with LVH defined by increased LVM index was present in approximately 30-50%. Furthermore, anatomical increased LVWT in those with LVH classified by ASE guidelines was present in a varying range of 10-70%. Patients with concentric LVH had a low prevalence of histological myocyte hypertrophy (approximately 10%). Inconsistencies in classification methods reveal the need for more robust LVH terminologies and definition that is consistent across cardiac imaging and anatomico-pathological examination, particularly in patients with CM.

Bidirectional Mamba with Dual-Branch Feature Extraction for Hyperspectral Image Classification

The hyperspectral image (HSI) classification task is widely used in remote sensing image analysis. The HSI classification methods based on convolutional neural networks (CNNs) have greatly improved the classification performance. However, they cannot well utilize the sequential properties of spectral features and face the challenge of increasing computational cost with the increase in network depth. To address these shortcomings, this paper proposes a novel network with a CNN-Mamba architecture, called DBMamba, which uses a bidirectional Mamba to process spectral feature sequences at a linear computational cost. In the DBMamba, principal component analysis (PCA) is first used to extract the main features of the data. Then, a dual-branch CNN structure, with the fused features from spectral–spatial features by 3D-CNN and spatial features by 2D-CNN, is used to extract shallow spectral–spatial features. Finally, a bidirectional Mamba is used to effectively capture global contextual information in features and significantly enhance the extraction of spectral features. Experimental results on the Indian Pines, Salinas, and Pavia University datasets demonstrate that the classification performance surpasses that of many cutting-edge methods, improving by 1.04%, 0.15%, and 0.09%, respectively, over the competing SSFTT method. The research in this paper enhances the existing knowledge on HSI classification and provides valuable insights for future research in this field.

Artificial Intelligence-Based Segmentation and Classification of Plant Images with Missing Parts and Fractal Dimension Estimation

Existing research on image-based plant classification has demonstrated high performance using artificial intelligence algorithms. However, limited camera viewing angles can cause parts of the plant to be invisible in the acquired images, leading to an inaccurate classification. However, this issue has not been addressed by previous research. Hence, our study aims to introduce a method to improve classification performance by taking these limitations into account; specifically, we incorporated both segmentation and classification networks structured as shallow networks to expedite the processing times. The proposed shallow plant segmentation network (Shal-PSN) performs adversarial learning based on a discriminator network; and a shallow plant classification network (Shal-PCN) with applied residual connections was also implemented. Moreover, the fractal dimension estimation is used in this study for analyzing the segmentation results. Additionally, this study evaluated the performance of the proposed Shal-PSN that achieved the dice scores (DSs) of 87.43% and 85.71% with PlantVillage and open leaf image (OLID-I) open datasets, respectively, in instances where 40–60% of plant parts were missing. Moreover, the results demonstrate that the proposed method increased the classification accuracy from 41.16% to 90.51% in the same instances. Overall, our approach achieved superior performance compared to the existing state-of-the-art classification methods.

General Industrial Process Optimization Method to Leverage Machine Learning Applied to Injection Moulding

ABSTRACTThe development of machine learning (ML) technologies is driving changes across many sectors. In industrial settings, this is called the fourth industrial revolution and encompasses several technologies pushing the boundaries of industrial automation. In this study, a general industrial process optimization (GIPO) methodology is formulated in the context of Industry 4.0 and tested on an industrial injection moulding machine (IMM). GIPO aims to encourage the practical inclusion of industrial artificial intelligence at all levels of the manufacturing process, while enabling industrial equipment to adapt to a changing processing environment. Special attention is given to the generality of the methodology so that it can be extended to other applications. In the example case study presented here, GIPO combines K‐nearest neighbours classification and nearest neighbours optimization methods to optimize an injection moulding process effectively. Practical implementation conducted on the IMM demonstrates a novel methodology to leverage data mining and ML methods in a real‐world setting to improve production quality, production time and energy cost.

YOLOv8-Based Estimation of Estrus in Sows Through Reproductive Organ Swelling Analysis Using a Single Camera

Accurate and efficient estrus detection in sows is crucial in modern agricultural practices to ensure optimal reproductive health and successful breeding outcomes. A non-contact method using computer vision to detect a change in a sow’s vulva size holds great promise for automating and enhancing this critical process. However, achieving precise and reliable results depends heavily on maintaining a consistent camera distance during image capture. Variations in camera distance can lead to erroneous estrus estimations, potentially resulting in missed breeding opportunities or false positives. To address this challenge, we propose a robust six-step methodology, accompanied by three stages of evaluation. First, we carefully annotated masks around the vulva to ensure an accurate pixel perimeter calculation of its shape. Next, we meticulously identified keypoints on the sow’s vulva, which enabled precise tracking and analysis of its features. We then harnessed the power of machine learning to train our model using annotated images, which facilitated keypoint detection and segmentation with the state-of-the-art YOLOv8 algorithm. By identifying the keypoints, we performed precise calculations of the Euclidean distances: first, between each labium (horizontal distance), and second, between the clitoris and the perineum (vertical distance). Additionally, by segmenting the vulva’s size, we gained valuable insights into its shape, which helped with performing precise perimeter measurements. Equally important was our effort to calibrate the camera using monocular depth estimation. This calibration helped establish a functional relationship between the measurements on the image (such as the distances between the labia and from the clitoris to the perineum, and the vulva perimeter) and the depth distance to the camera, which enabled accurate adjustments and calibration for our analysis. Lastly, we present a classification method for distinguishing between estrus and non-estrus states in subjects based on the pixel width, pixel length, and perimeter measurements. The method calculated the Euclidean distances between a new data point and reference points from two datasets: “estrus data” and “not estrus data”. Using custom distance functions, we computed the distances for each measurement dimension and aggregated them to determine the overall similarity. The classification process involved identifying the three nearest neighbors of the datasets and employing a majority voting mechanism to assign a label. A new data point was classified as “estrus” if the majority of the nearest neighbors were labeled as estrus; otherwise, it was classified as “non-estrus”. This method provided a robust approach for automated classification, which aided in more accurate and efficient detection of the estrus states. To validate our approach, we propose three evaluation stages. In the first stage, we calculated the Mean Squared Error (MSE) between the ground truth keypoints of the labia distance and the distance between the predicted keypoints, and we performed the same calculation for the distance between the clitoris and perineum. Then, we provided a quantitative analysis and performance comparison, including a comparison between our previous U-Net model and our new YOLOv8 segmentation model. This comparison focused on each model’s performance in terms of accuracy and speed, which highlighted the advantages of our new approach. Lastly, we evaluated the estrus–not-estrus classification model by defining the confusion matrix. By using this comprehensive approach, we significantly enhanced the accuracy of estrus detection in sows while effectively mitigating human errors and resource wastage. The automation and optimization of this critical process hold the potential to revolutionize estrus detection in agriculture, which will contribute to improved reproductive health management and elevate breeding outcomes to new heights. Through extensive evaluation and experimentation, our research aimed to demonstrate the transformative capabilities of computer vision techniques, paving the way for more advanced and efficient practices in the agricultural domain.

Mapping the Knowledge Structure of Image Recognition in Cultural Heritage: A Scientometric Analysis Using CiteSpace, VOSviewer, and Bibliometrix

The application of image recognition techniques in the realm of cultural heritage represents a significant advancement in preservation and analysis. However, existing scholarship on this topic has largely concentrated on specific methodologies and narrow categories, leaving a notable gap in broader understanding. This study aims to address this deficiency through a thorough bibliometric analysis of the Web of Science (WoS) literature from 1995 to 2024, integrating both qualitative and quantitative approaches to elucidate the macro-level evolution of the field. Our analysis reveals that the integration of artificial intelligence, particularly deep learning, has significantly enhanced digital documentation, artifact identification, and overall cultural heritage management. Looking forward, it is imperative that research endeavors expand the application of these techniques into multidisciplinary domains, including ecological monitoring and social policy. Additionally, this paper examines non-invasive identification methods for material classification and damage detection, highlighting the role of advanced modeling in optimizing the management of heritage sites. The emergence of keywords such as ‘ecosystem services’, ‘models’, and ‘energy’ in the recent literature underscores a shift toward sustainable practices in cultural heritage conservation. This trend reflects a growing recognition of the interconnectedness between heritage preservation and environmental sciences. The heightened awareness of environmental crises has, in turn, spurred the development of image recognition technologies tailored for cultural heritage applications. Prospective research in this field is anticipated to witness rapid advancements, particularly in real-time monitoring and community engagement, leading to the creation of more holistic tools for heritage conservation.