Classification Model Research Articles

Defining the optimal setting for transcriptomic analyses on blood samples for response prediction in immunotherapy-treated NSCLC patients

Transcriptomic profiling of blood immune cells offers a promising alternative to invasive, sampling bias-prone tissue-based biomarker assays for predicting immune checkpoint inhibitor (ICI) therapy response in non-small cell lung cancer (NSCLC) patients. However, the optimal analytical approach to identify systemic correlates of response still needs to be explored. We collected peripheral blood mononuclear cells and whole blood (WB) samples from 33 ICI-treated NSCLC patients before ICI treatment and at the first response evaluation. After bulk polyadenylated RNA-sequencing, we assessed differences in gene expression profiles between non-responders and responders using differential expression analysis, single sample gene set enrichment analysis (ssGSEA), and cell type deconvolution. We evaluated gene expression values, ssGSEA scores, and deconvolved cell type proportions to distinguish non-responders from responders via ROC curve (AUC) analysis, training a logistic regression classification model. Gene expression values and deconvolved proportions yielded the best results with WB samples after treatment (AUC = 0.87 and 0.85, respectively). Overall, ssGSEA scores showed superior classification performance across all sample types and timepoints (AUC > 0.7). In conclusion, transcriptomic analysis through ssGSEA demonstrated the best performance as a non-invasive biomarker for predicting clinical benefit in ICI-treated NSCLC patients, with gene expression and deconvolution on post-treatment WB samples also showing promising results.

AGCNAF: predicting disease-gene associations using GCN and multi-head attention to fuse the similarity features

Abstract Association prediction between diseases and genes is a critical step in revealing the molecular mechanisms of diseases and developing drug treatment strategies. With the explosive growth of data in the biomedical field, how to effectively utilize these data for accurate prediction has become a hotspot and challenge in current research. To overcome the limitations of current prediction methods in dealing with complex biological network structures and feature extraction, this study proposes AGCNAF, a method that combines an unsupervised Graph Convolutional Network (GCN) and a multi-head attention mechanism. The metagraph-guided random walk strategy enables AGCNAF to capture local and high-order topological structures in the graph, while GCN is responsible for realizing deep feature extraction of these structures. By incorporating similarity features through the multi-attention mechanism, AGCNAF achieves effective integration of global and local features, which significantly improves the prediction performance. By incorporating similarity features through the multi-attention mechanism, AGCNAF achieves effective integration of global and local features, which significantly improves the prediction performance. By utilizing the machine learning binary classification model for prediction, the experimental results through five-fold cross-validation show that AGCNAF demonstrates significant advantages in prediction performance compared to existing methods, with its AUC and AUPR reaching 0.9686 and 0.9709, respectively, and the AUC up to 0.9812 under specific conditions. 
To verify the practical application value of AGCNAF, this study also conduct case studies on Alzheimer's disease, lung cancer, and breast cancer. The results further confirm the excellent performance of AGCNAF in identifying potential disease-gene associations, which opens up new possibilities for future disease-gene research.

Advanced Image Classification Using Convolutional Neural Networks

Deep learning is essential for computer vision and object detection. This project explores the use of Convolutional Neural Networks (CNNs) for image classification using the CIFAR-10 dataset, which is widely used and provides a robust benchmark for evaluating image classification models. The CNN model, comprising various convolutional, pooling, and fully connected layers, is trained on a portion of the dataset and tested on another portion to evaluate its performance. Techniques such as data augmentation, dropout, and specific activation functions are employed to enhance the model’s efficacy. The study details the CNN architecture and reports the results, including accuracy metrics, to demonstrate the model’s effectiveness.

A Multi-label Artificial Intelligence Approach for Improving Breast Cancer Detection With Mammographic Image Analysis.

Breast cancer remains a major global health concern. This study aimed to develop a deep-learning-based artificial intelligence (AI) model that predicts the malignancy of mammographic lesions and reduces unnecessary biopsies in patients with breast cancer. In this retrospective study, we used deep-learning-based AI to predict whether lesions in mammographic images are malignant. The AI model learned the malignancy as well as margins and shapes of mass lesions through multi-label training, similar to the diagnostic process of a radiologist. We used the Curated Breast Imaging Subset of Digital Database for Screening Mammography. This dataset includes annotations for mass lesions, and we developed an algorithm to determine the exact location of the lesions for accurate classification. A multi-label classification approach enabled the model to recognize malignancy and lesion attributes. Our multi-label classification model, trained on both lesion shape and margin, demonstrated superior performance compared with models trained solely on malignancy. Gradient-weighted class activation mapping analysis revealed that by considering the margin and shape, the model assigned higher importance to border areas and analyzed pixels more uniformly when classifying malignant lesions. This approach improved diagnostic accuracy, particularly in challenging cases, such as American College of Radiology Breast Imaging-Reporting and Data System categories 3 and 4, where the breast density exceeded 50%. This study highlights the potential of AI in improving the diagnosis of breast cancer. By integrating advanced techniques and modern neural network designs, we developed an AI model with enhanced accuracy for mammographic image analysis.

B-TBM: A Novel Deep Learning Model with Enhanced Loss Function for HAZOP Risk Classification Using Natural Language Statistical Laws

HAZOP is a paradigm of industrial safety, and the introduction of deep learning-based HAZOP text categorization marks the arrival of an intelligent era of safety analysis. However, existing risk analysis methods have limitations in processing complex texts and extracting deep risk features. To solve this problem, this paper proposes a novel HAZOP risk event classification model based on BERT, BiLSTM, and TextCNN. The complexity of HAZOP text is revealed by introducing statistical laws of natural language, such as Zipf’s law and Heaps’ law, and the outputs of different levels of BERT are further combined linearly to collaborate with BiLSTM and TextCNN to capture long-term dependency and local contextual information for a more accurate classification task. Meanwhile, an improved loss function is proposed to effectively solve the deficiencies of the traditional cross-entropy loss function in the mislabeling process and improve the generalization ability of the model. It is experimentally demonstrated that the accuracy of the model is improved by 3% to 4% compared to the traditional BERT model in the task of severity and possibility classification of HAZOP reports. This study not only improves the accuracy and efficiency of HAZOP risk analysis, but also provides new ideas and methods for the application of natural language processing in industrial safety.

Application of Multi-Input Multi-Output Brain-Computer Interface in Cognitive Assessment for Alzheimer’s Disease

This research aims to augment human cognitive abilities, particularly memory function, through the development of innovative prosthetics. The hippocampus, a crucial brain region involved in memory encoding and recall, has been the focal point of numerous studies seeking to develop prosthetics that can restore or enhance memory function in individuals with impairments such as Alzheimer’s disease. This review examines two studies that delve into the development and application of hippocampal neural prosthetics, highlighting the potential of these prosthetics to revolutionize the treatment of memory impairments. One study utilized static neural stimulation patterns to enhance memory for specific information content, while another developed a nonlinear Multiple Input Multiple Output (MIMO) dynamic model to predict and stimulate neural activity patterns during memory encoding. Both studies demonstrated significant improvements in memory performance. Additionally, this paper discusses the clinical significance and application prospects of hippocampal neural prosthetics, including the potential for implantable prosthetics and the widespread applicability of MIMO stimulation. Furthermore, the paper explores advanced modeling and classification techniques for neural ensembles and visual memory decoding, including Volterra-type hierarchical modeling, optimal multi-unit stimulation patterns, and a sparse classification model for decoding visual memories. Finally, the paper presents the implementation and evaluation of neural prosthesis systems, including an application-specific integrated circuit (ASIC) design for a hippocampal prosthesis and a high-performance, scalable system architecture for real-time estimation of neural activity. These advancements hold promise for future research and applications in neural prosthetics for memory enhancement.

Comparison of Machine Learning Models and Feature Importance Investigation of Intelligent Fault Diagnosis Methods for Robots Based on Datasets Across Various Distributions

Intelligent fault detection is an important component of the industrial and automation fields. However, conventional research on intelligent fault detection mainly focuses on industrial production equipment, while there is little research on intelligent fault detection scenarios for robots. Based on a hexapod robot joints dataset, this paper investigates intelligent fault diagnosis tasks in the field of robotics. Firstly, the dataset is clustered into two new datasets with different distributions through k-means method, which is used to simulate the practicality of imbalanced distribution between healthy and fault data. Afterwards, several multioutput machine learning classification models were established to predict robot joints with faults. In two datasets with different distributions, the larger one is used as the training set and the smaller one is used as the testing set. The article compares the performance of these models with prediction accuracy as the main indicator. And based on the results, the paper selects the model with the highest accuracy for further exploration of feature importance. Finally, the article explains the significance of the results and analyzes possible reasons. The experimental results show that the random forest model better overcomes the problem of data distribution differences and has the highest accuracy. In the random forest model, the importance of position data representing position error is higher than that of slope with respect to the axis data representing angle error. This result may be related to the distribution of feature values and the fact that position data contains more crucial information.

Isokinetic Force Modeling in Sports Based on Embedded System of Motion Recognition Model

With the gradual deepening of sports research, sports training methods integrating more scientific and technological means can greatly improve athletes’ professional ability. Combined with the isokinetic strength training method, it can effectively alleviate muscle recovery and body flexibility, which requires an analysis of the level of sports body promotion, and a more efficient training program for various sports parameters. In this study, taking strength enhancement as an example, the convolutional neural network is innovatively integrated into isokinetic strength training. Using the motion data obtained by wearable sensors in the training, we trained the neural network model by real-time analysis of data samples through an embedded system. Further, we optimized the network parameters to realize the extraction and recognition of EMG depth features. Compared with the traditional isokinetic strength classification model, this model lays a foundation for further constructing personalized strength training guidance methods. After the simulation test, the difference of the model’s accuracy requirement before and after the pre-training is 4.6%, and the difference of the test set before and after the pre-intensive training is [Formula: see text]1%. It is 16.2% higher than the TCAM algorithm, and the gap with STA-LSTM remains between [Formula: see text]. It is verified that the attention concentration control scheme can not only reduce the error rate in visual recognition tasks, but also improve the complexity of the neural network model. The optimized lightweight data structure is especially suitable for embedded systems with limited computing power. All kinds of test projects verify its rationality and make full use of the control ability of the embedded system and the characteristics of parallel computing and programmability, which has great practical application value for the development of the Internet of Things system network. The research results have made a thorough study of the physiological guidance of existing strength training, the identification and evaluation methods of exercise-induced muscle fatigue, and the current situation of the existing isokinetic equipment.

Classification of Non-Seismic Tsunami Early Warning Level Using Decision Tree Algorithm

Background: Tsunami caused by volcanic collapse are categorized as non-seismic uncommon events, unlike tsunamis caused by earthquakes, which are common events. The traditional tsunami early warning based on the seismic sensor (e.g. earthquake detectors) may not be applicable to volcanic tsunamis because they do not generate seismic waves. Consequently, these tsunamis cannot be detected in advance, and warnings cannot be issued. New methods should be explored to address these non-seismic tsunamis caused by volcanic collapse. Objective: This study explored the potential of machine learning algorithms in supporting early warning level issuing for non-seismic tsunamis, specifically volcanic tsunamis. The Anak Krakatau volcano event in Indonesia was used as a case study. Methods: This study generated a database of 160 collapse scenarios using numerical simulation as input sequences. A classification model was constructed by defining the worst tsunami elevation and its arrival time at the coast. The database was supervised by labeling the warning levels as targets. Subsequently, a decision tree algorithm was employed to classify the warning levels. Results: The results demonstrated that the classification model performs very well for the Major Tsunami, Minor Tsunami, and Tsunami classes, achieving high precision, recall, and F1-Score with very high accuracy of 98%. However, the macro average indicates uneven performance across classes, as there are instances of ‘No Warning’ in some coastal gauges. Conclusion: To improve the model performance in the ‘No Warning’ class, it is necessary to balance the dataset by including more ‘No Warning’ scenarios, which can be achieved by simulating additional scenarios involving very small-volume collapse. Additionally, exploring additional collapse parameters such as dip angle and outlier volume could contribute to developing a more robust classification model. Keywords: Machine Learning, Classification, Volcanic Tsunamis, Early Warning, Decision Tree

Quantitative texture analysis using machine learning for predicting interpretable pulmonary perfusion from non-contrast computed tomography in pulmonary embolism patients

BackgroundPulmonary embolism (PE) is life-threatening and requires timely and accurate diagnosis, yet current imaging methods, like computed tomography pulmonary angiography, present limitations, particularly for patients with contraindications to iodinated contrast agents. We aimed to develop a quantitative texture analysis pipeline using machine learning (ML) based on non-contrast thoracic computed tomography (CT) scans to discover intensity and textural features correlated with regional lung perfusion (Q) physiology and pathology and synthesize voxel-wise Q surrogates to assist in PE diagnosis.MethodsWe retrospectively collected 99mTc-labeled macroaggregated albumin Q-SPECT/CT scans from patients suspected of PE, including an internal dataset of 76 patients (64 for training, 12 for testing) and an external testing dataset of 49 patients. Quantitative CT features were extracted from segmented lung subregions and underwent a two-stage feature selection pipeline. The prior-knowledge-driven preselection stage screened for robust and non-redundant perfusion-correlated features, while the data-driven selection stage further filtered features by fitting ML models for classification. The final classification model, trained with the highest-performing PE-associated feature combination, was evaluated in the testing cohorts based on the Area Under the Curve (AUC) for subregion-level predictability. The voxel-wise Q surrogate was then synthesized using the final selected feature maps (FMs) and model score maps (MSMs) to investigate spatial distributions. The Spearman correlation coefficient (SCC) and Dice similarity coefficient (DSC) were used to assess the spatial consistency between FMs or MSMs and Q-SPECT scans.ResultsThe optimal model performance achieved an AUC of 0.863 during internal testing and 0.828 on the external testing cohort. The model identified a combination containing 14 intensity and textural features that were non-redundant, robust, and capable of distinguishing between high- and low-functional lung regions. Spatial consistency assessment in the internal testing cohort showed moderate-to-high agreement between MSMs and reference Q-SPECT scans, with median SCC of 0.66, median DSCs of 0.86 and 0.64 for high- and low-functional regions, respectively.ConclusionsThis study validated the feasibility of using quantitative texture analysis and a data-driven ML pipeline to generate voxel-wise lung perfusion surrogates, providing a radiation-free, widely accessible alternative to functional lung imaging in managing pulmonary vascular diseases.Clinical trial numberNot applicable.

Optimisation of artefact detection in photoplethysmography heart rate data: influence of different classifiers in machine learning models

Abstract Background Heart rate (HR) tracking by wrist-worn devices using photoplethysmography (PPG) could assist in continuously following up physical activity. However, the accuracy can be impacted by (motion) artefacts. Machine learning models could help to recognise artefacts in PPG-based HR data. The choice of classifier in these machine learning models is a determing factor for task performance of the model. Purpose This study evaluates and determines the optimal classifier for a new machine learning-based approach to enhance the reliability of artefact detection in PPG-based HR data. Methods A total of 62 participants (27 cardiac rehabilitation patients, 35 healthy athletes) wore both a test device and a reference device measuring HR continuously for 24 hours. A training dataset was prepared, assigning two independent labels (i.e. anomaly and activity) to each HR episode based on the reference device data. Fitbit data were processed using our in-house designed artefact removal procedure, which involves the application of two classification models: one for anomaly detection and another for activity detection. Four distinct classifiers were employed for both models: Balanced Bagging, Balanced Bagging with Random Forest, Balanced Random Forest, and Logistic Regression. Each classifier was evaluated using area under the receiver operating characteristic curve (ROC-AUC), accuracy, sensitivity and specificity. Results Of the 1,647,328 HR data points collected, 103,095 (6.26%) were identified as artefacts. Figure 1 and Figure 2 summarise the performance of the distinct classifiers for the anomaly model and the activity model, respectively. Balanced Bagging and Balanced Bagging with Random Forest consistently demonstrate the highest AUC values and accuracies across both anomaly and activity detection models (anomaly detection: AUC = 0.95, accuracy = 89-85%; activity detection: AUC = 0.98, accuracy = 95%). Comparing these two, Balanced Bagging with Random Forest emerges as the preferred option, given the highest sensitivity in both anomaly detection (93%>86%) and activity detection models (99%>96%). In contrast, Balanced Random Forest and Logistic Regression exhibit inferior performance. In the anomaly detection model, Balanced Random Forest exhibits a lower sensitivity of 75%, while Logistic Regression performs even worse with a sensitivity of 25%. Similarly, in the activity detection model, both Balanced Random Forest and Logistic Regression demonstrate diminished performance. Conclusions Balanced Bagging with Random Forest emerges as the optimal classifier to detect anomalies and activities in continuous PPG-based HR data, thus contributing to the optimisation of our in-house designed procedure for removing artefacts. This processing aims to provide a reliable and automatic way for continuous HR monitoring, which can help monitor and guide physical activities.

Intelligent monitoring of post-processing characteristics in 3D-printed food products: A focus on fermentation process of starch-gluten mixture using NIR and multivariate analysis

The production of three-dimensional (3D)-printed food products requires not only optimal 3D-printing adaptability but also appropriate post-processing characteristics. This study aimed to use near infrared (NIR) spectroscopy to predict the rheological properties of 3D-printed dough, enabling intelligent monitoring of the dough's fermentation process. Utilizing support vector machine (SVM) classification model, the fermentation stages can be classified as under-fermentation, complete fermentation, and over-fermentation. Employing preprocessing methods with Synergy Interval Partial Least Square-Competitive Adaptive Reweighted Sampling (SIPLS-CARS) algorithm, 27, 39, 23, and 27 key wavelengths were filtered from the raw NIR spectral data, corresponding to the prediction of storage modulus (G′), loss modulus (G″), complex viscosity (η∗), and loss factor (tan δ), respectively. Quantitatively, SVM (Support Vector Machine) regression outperformed Partial Least Squares (PLS) with Rc2 values (0.95, 0.94, 0.94) and Rp2 values (0.93, 0.93, 0.94) for G′, G″, and η∗. NIR spectra-based predictive models demonstrated superior performance compared to rheo-fermentation properties models. In summary, these findings show the potential of NIR spectroscopy as a rapid tool for predicting the fermentation progress of 3D-printed doughs.

Transfer learning for probabilistic localization of hidden cracks in concrete structures

AbstractThe utility of discriminative supervised learning models built using multiple training-data sources is investigated for hidden crack localization in concrete. Feed-forward neural network (FFNN) is chosen as the model architecture, and transfer learning is used to assimilate the information obtained from different sources (computational physics simulations and laboratory experiments). The labeled training data consists of values of a damage index and the known locations of hidden cracks. The classification models need to learn how the presence of damage (hidden cracks) affects the damage index at different sensors for different test conditions. To this end, diagnostic FFNN models are built by sequentially adding and training new hidden layers to assimilate labeled information from computer models (different model geometries, test conditions, crack lengths, crack locations) and laboratory experiments on a plain cement slab. These transfer learning-based models are then used to localize damage in concrete specimens that reflect real-world conditions (i.e., specimens with steel reinforcement and randomly distributed aggregate). The actual damage state in these specimens is determined by extracting cores and performing petrographic studies on the extracted cores. The damage probability estimated by transfer learning-based models is compared with the petrographic damage rating index (DRI) to identify the most suitable approach to train the diagnostic models. The transfer learning-based diagnostic methodology shows promise and could be used in various structural health monitoring applications, where sufficient labeled data are typically not available from a single data source.

Diagnostic and prognostic value of ECG-predicted hypertension mediated left ventricular hypertrophy using machine learning

Abstract Background Hypertension is the commonest cause of left ventricular hypertrophy (LVH). Using cardiac magnetic resonance (CMR) imaging, four distinct hypertension mediated LVH phenotypes have been reported: normal LV, LV remodelling, eccentric LVH and concentric LVH. Early detection of hypertensives with LVH can enable timely initiation of therapy however a CMR based screening strategy is costly, particularly in low resource settings. The electrocardiogram (ECG) is an inexpensive screening tool and can detect LVH but its capacity to discriminate between the 4 phenotypes is unknown. Purpose To classify hypertension mediated LVH phenotypes from the ECG using machine learning (ML) and test for associations of ECG predicted LVH phenotypes with incident cardiovascular outcomes. Methods ECG markers with a known physiological association with LVH (including QRS markers, Sokolow-Lyon, and Cornell) were extracted from the 12-lead ECG of 20,439 hypertensive participants in UK Biobank. Three classification models, logistic regression, support vector machine (SVM) and random forest, were trained in 80% of the participants using extracted ECG markers and clinical variables. The remaining 20% individuals were included in the test set for performance measurement. External validation was sought in 877 hypertensives from the Study of Health in Pomerania (SHIP). Subsequently, we tested for associations between the four ECG-predicted LVH phenotypes and incident major adverse cardiovascular events (MACE) and heart failure using Cox proportional hazard regression in the UK Biobank test set with median follow-up of 4.2 years. Results Among UK Biobank participants (mean age ± standard deviation 65 ± 7.4 years), 23,190 had normal LV, 894 LV remodelling, 218 eccentric and 103 concentric LVH. Classification performance of the three models was comparable, but SVM showed consistent results (accuracy 0.79, sensitivity 0.59, specificity 0.87, AUC 0.69) and superior prediction of eccentric LVH (AUC 0.86). In SHIP (55 ± 12.7 years), 704 had normal LV, 134 LV remodelling, 12 eccentric and 24 concentric LVH. In the external validation performed in SHIP, SVM had accuracy of 0.75 (sensitivity 0.51, specificity 0.85, AUC 0.65), with superior prediction of eccentric LVH (AUC 0.76), as shown in Figure 1. Ventricular rate and QRS amplitude (P<0.001) were the features most strongly associated with LVH. In the test set of UK Biobank, setting normal LV as the reference group, the ECG predicted eccentric LVH group was associated with heart failure (Hazard Ratio 3.42, confidence interval 1.06-9.86), there were no associations with MACE (Figure 2). Conclusions ECG-based ML classifiers were able to discriminate the four hypertension mediated LVH phenotypes with external validation demonstrating robustness. This automated approach enhances capabilities of non-specialists and potentially represents an accessible screening strategy for the early detection of hypertensives with LVH.

Research on flood warning model based on dimensionality reduction and integrated neural network

Flood disasters have long-term impacts on human societies and ecosystems, and population growth and urbanization lead to changes in surface conditions that increase the frequency and magnitude of floods. Therefore, understanding and predicting flood probability and constructing a flood warning mechanism are urgent problems. In this paper, a flood warning mechanism based on the probability of flood occurrence is proposed to construct a high-performance and highly interpretable model in a step-by-step manner using machine learning and deep learning methods. First, the correlation level is elucidated by the PCA dimensionality reduction method. Secondly, the K-means clustering method is used to replace the qcut binning method, and the classification model is built by combining the multidimensional data point distribution and classified by polynomial logistic regression. Finally, the integrated neural network model is constructed, the visual model performance is used for full feature optimization, and the performance of the classification model is optimized to achieve the flood prediction effect through the mean square error as the combination of the weights of the integrated model.

Prediction of Incident Diabetic Retinopathy in Adults With Type 1 Diabetes Using Machine Learning Approach: An Exploratory Study.

Early detection and intervention are crucial for preventing vision-threatening diabetic retinopathy (DR) in adults with type 1 diabetes (T1D). This exploratory study uses machine learning on continuous glucose monitoring (CGM) data to identify factors influencing DR and predict high-risk individuals for timely intervention. Between June 2018 and March 2022, adults with T1D with incident DR or no retinopathy (control) were identified. The CGM data were collected retrospectively for up to seven years before the date of defining incident DR or no retinopathy. A mixture of three machine learning algorithms was trained and evaluated in two different scenarios, using different glycemic features extracted from CGM traces (scenario 1), and the two principal components (two PCs; exposure to hyperglycemia and hypoglycemia risk) of those features (scenario 2). Classifiers were evaluated through 10-fold cross-validation using the receiver operating characteristic area under the curve (AUC-ROC) to select the best classification model. The CGM data of 30 adults with incident DR (mean±SD age of 21.2±9.4 years, glycated hemoglobin [HbA1c] of 8.6%±1.0%, and body mass index [BMI] of 24.5±4.8 kg/m2) and 30 adults without DR (age of 41.8±14.7 years, HbA1c of 7.0%±0.9%, and BMI of 26.2±3.6 kg/m2) were included in this analysis. In scenario 2, classifiers outperformed scenario 1, resulting in an average AUC-ROC increase to 0.92 for two of three models, indicating that the two PCs captured vital classification data, representing the most discriminative aspects and enhancing model performance. Machine learning approaches using CGM data may have potential to aid in identifying adults with T1D at risk of DR.

Artificial intelligence derived electrocardiograms: a comparative study of network architectures predicting sex and left ventricular dysfunction

Abstract Background Within cardiovascular medicine, AI-driven electrocardiogram (ECG) analysis has emerged as a powerful tool, capable of predicting clinically significant abnormalities. The increasing growth of wearable technologies for ECG monitoring underscores the importance of advancing AI methodologies. While Convolutional Neural Networks (CNNs) have traditionally been the go-to choice for ECG analysis, the rise of Vision Transformers (ViTs), a novel computer vision paradigm, raises the question of their potential superiority. Purpose To compare the performance of CNNs and ViTs in ECG analysis, and to assess the efficiency of lean models using truncated ECG recordings. Methods We trained a model using 12-lead ECGs as input for the binary classification of two outputs: Sex (Male or Female) and Left ventricular dysfunction (LVD, defined as below 35%), determined by echocardiography within 2 weeks of the ECG. Models were developed using full standard ECGs, single lead and truncated recordings (1,2 and 4 10 seconds). We compared CNNs (PyTorch) and ViTs (HuggingFace) performance. We analyzed the explainability of our findings, representing the AI model's focus of interest on ECGs from normal cases versus those with reduced ejection fraction, thereby validating its diagnostic discernment (Fig 1) Results We identified 150,691 and 29,422 patients with valid ECG tests for the Sex and LVD prediction models, respectively. For the Sex outcome, the AUROCs were 0.911 (95% CI: 0.908-0.914) and 0.898 (95% CI: 0.894-0.901) for the CNN-based model and the ViT-based models, respectively, with the difference statistically significant (P-Value < 0.001). For the LVD outcome, the AUROCs were comparable at 0.878 (95% CI: 0.864-0.892) and 0.866 (95% CI: 0.853-0.879) for the CNN-based model and the ViT-based models, respectively, (P-Value = 0.056). Furthermore, we found that 98.8% for the maximal predictive power of the models was achieved within a 2 second time frame of a 12 lead ECG. Similarly, a single lead based model achieved a relatively high AUC for the prediction of LVD. Conclusions AI-ECG models developed with ViT architecture did not surpass CNN-based models in sex classification and LVD identification. Our study highlights the significance of lean models allowing for rapid prediction within 2-second tracings or using a single lead, demonstrates the potential for streamlined AI-ECG applications in clinical practice.

CNN-Informer: A hybrid deep learning model for seizure detection on long-term EEG

Timely detecting epileptic seizures can significantly reduce accidental injuries of epilepsy patients and offer a novel intervention approach to improve their quality of life. Investigation on seizure detection based on deep learning models has achieved great success. However, there still remain challenging issues, such as the high computational complexity of the models and overfitting caused by the scarce availability of ictal electroencephalogram (EEG) signals for training. Therefore, we propose a novel end-to-end automatic seizure detection model named CNN-Informer, which leverages the capability of Convolutional Neural Network (CNN) to extract EEG local features of multi-channel EEGs, and the low computational complexity and memory usage ability of the Informer to capture the long-range dependencies. In view of the existence of various artifacts in long-term EEGs, we filter those raw EEGs using Discrete Wavelet Transform (DWT) before feeding them into the proposed CNN-Informer model for feature extraction and classification. Post-processing operations are further employed to achieve the final detection results. Our method is extensively evaluated on the CHB-MIT dataset and SH-SDU dataset with both segment-based and event-based criteria. The experimental outcomes demonstrate the superiority of the proposed CNN-Informer model and its strong generalization ability across two EEG datasets. In addition, the lightweight architecture of CNN-Informer makes it suitable for real-time implementation.

Enhancing drug-target interaction predictions in context of neurodegenerative diseases using bidirectional long short-term memory in male Swiss albino mice pharmaco-EEG analysis

Background and ObjectiveEmerging diseases like Parkinson or Alzheimer's, which are not curable, endanger human mental health and are challenging to research. Drug target interactions (DTI) are pivotal in the screening of candidate drugs and focus on a small pool of drug targets. Electroencephalogram shows the responses to psychotropic medicines in the brain bioelectric activity. Synaptic activity can be analyzed by using Local Field Potential recordings obtained from micro-electrodes implanted in the brain. The aim is to evaluate the effects of drug on brain bioelectric activity and increase the drug classification accuracy. The ultimate goal is to advance our understanding of how drugs affect synaptic activity and open the door to more focused treatment for neurodegenerative diseases. MethodsIn this study, Pharmaco-EEG recordings are processed using Advanced neural network models, particularly Convolutional Neural Networks, to assess the effects of medications. The five different medicines used in this study are Ephedrine, Fluoxetine, Kratom, Morphine, and Saline. The signals observed are local field potential signals. To overcome some limits of DTI prediction, we propose Bidirectional Long Short-Term Memory (LSTM) for the categorization of intracranial EEG (i-EEG) data, departing from standard approaches. Similar EEG patterns are presumably caused by drugs that work by homologous pharmacological pathways, producing similar psychotropic effects. To improve accuracy and reduce training loss, our study introduces a bidirectional LSTM model for classification along with Bayesian optimization ResultsHigh recall, precision, and F1-Scores, particularly a 95% F1-Score for morphine, ephedrine, fluoxetine, and saline, suggest good performance in predicting these drug classes. Kratom produces a somewhat lower recall of 94%, but a high F1-Score of 97% and perfect precision of 1.00. The weighted average F1-Score, macro average, and overall accuracy are all consistently high (around 97%), indicating that the model works well throughout the spectrum of drugs. ConclusionsImproved model performance was demonstrated by using a diversified dataset with five drug categories and bidirectional LSTM boosted with Bayesian optimization for hyperparameter tuning. From earlier limited-category models, it represents a substantial advancement.

Chemometric approaches for discriminating manufacturers of Korean handmade paper using infrared spectroscopy

The objective of this study was to identify the manufacturer of Hanji, Korean handmade paper widely used in conservation science. To achieve this, machine learning models utilizing attenuated total reflectance–infrared spectroscopy (ATR–IR) were developed to assess the robustness and effectiveness of the computed models. Principal component analysis (PCA), partial least squares–discriminant analysis (PLS–DA), decision tree (DT), and k-NN models were constructed using IR spectral data, with the spectral region between 1800 and 1500 cm⁻1 identified as the critical input variable through Variable Importance in Projection (VIP) scores. The transformation of the obtained spectra into second derivative spectra proved beneficial in this key spectral region, leading to significant improvements in model performance. Additionally, the application of DBSCAN for outlier detection was effective in refining the dataset, further enhancing the performance of the models. Specifically, the k-NN model, when applied to the selected variables and preprocessed with the second derivative transformation, achieved an F1 score of 0.92. These findings underscore the importance of focusing on the 1800–1500 cm⁻1 spectral range and applying outlier detection techniques, such as DBSCAN, to enhance the robustness and accuracy of the Hanji classification models by eliminating the influence of atypical data points.