Accuracy Performance Research Articles

A Comprehensive Review of Processing-in-Memory Architectures for Deep Neural Networks

This comprehensive review explores the advancements in processing-in-memory (PIM) techniques and chiplet-based architectures for deep neural networks (DNNs). It addresses the challenges of monolithic chip architectures and highlights the benefits of chiplet-based designs in terms of scalability and flexibility. This review emphasizes dataflow-awareness, communication optimization, and thermal considerations in PIM-enabled manycore architectures. It discusses tailored dataflow requirements for different machine learning workloads and presents a heterogeneous PIM system for energy-efficient neural network training. Additionally, it explores thermally efficient dataflow-aware monolithic 3D (M3D) NoC architectures for accelerating CNN inferencing. Overall, this review provides valuable insights into the development and evaluation of chiplet and PIM architectures, emphasizing improved performance, energy efficiency, and inference accuracy in deep learning applications.

Text sentiment analysis and classification based on bidirectional long and short term memory networks

In this study, we explored the method of text sentiment analysis and classification using bi-directional long and short-term memory networks, which brings new ideas to the field of sentiment classification. In the text pre-processing stage, we processed chat phrases, non-alphanumeric characters, deactivated words, numbers, special characters, blanks, and addresses to convert the raw text data into numerical data that can be processed by the model. Subsequently, word frequency statistics were conducted for texts of different mood types to analyse the frequency of occurrence of different words in texts of different mood types. By introducing a bi-directional long and short-term memory network model for training, we observed the accuracy and loss value performance on the training and validation sets. The accuracy of the training set increases from the initial 94.6% to 99.5%, and the accuracy of the validation set gradually grows to 99.4%; at the same time, the loss value of the training set decreases from the initial value of 0.2075 to 0.0194, and the loss value of the validation set gradually increases from the initial value of 0.0538 to 0.0372. This indicates that the bi-directional long short-term memory network achieves a significant and stable performance improvement. The confusion matrix results for the test set show that our model achieves more than 99% prediction accuracy on the test set. This implies that our proposed method can efficiently classify textual sentiment with high prediction accuracy. This research result helps to improve the effectiveness of the application of sentiment analysis techniques in the fields of social media opinion monitoring and product review analysis, and provides a useful reference for further deepening the research on text sentiment analysis. In conclusion, this study achieves the optimisation of the text sentiment classification task by means of a bidirectional long and short-term memory network model, and achieves satisfactory results. In the future, the model performance and generalisation ability can be further improved by combining more corpus and optimising the model structure, so as to be better applied in real-world scenarios.

Intelligent Identification of Surrounding Rock Grades Based on a Self-Developed Rock Drilling Test System

The classification of surrounding rock is crucial for formulating safe tunnel construction plans and support measures. However, the complex geological environment of tunnels presents a challenge in obtaining accurate drilling parameters for rock mass classification. This paper presents the development of a rock drilling testing system, which includes a propulsion speed acquisition system, oil pressure acquisition system, air pressure acquisition system, and an automatic data acquisition system. This system enables real-time, high-precision automatic collection and storage of parameters such as propulsion speed, with data collected twice per second for each parameter. Leveraging the Qingdao Metro Line 6 as a case study, we conducted rock mass drilling and constructed a rock mass classification database. By employing kernel density estimation and Pearson correlation analysis, we quantified the correlation between rock mass classification and the drilling parameters. The results indicated that relying on a single drilling parameter is insufficient for accurately determining rock mass classification. Both impact pressure and rotational pressure showed the strongest correlation with rock mass classification, each with a correlation coefficient below −0.8 (indicating a strong negative correlation). Outlier values of drilling parameters were excluded using the interval method. Based on the remaining data, we established an intelligent rock mass classification model using the random forest algorithm. This model demonstrated good accuracy and generalization performance, with an average accuracy exceeding 0.9. The proposed rock drilling testing system, combined with the intelligent rock mass classification model, forms an integrated system for the intelligent identification of rock mass grades. This system has significant implications for the intelligent and safe construction of drill-and-blast tunnels.

Evaluating and Enhancing Artificial Intelligence Models for Predicting Student Learning Outcomes

Predicting student outcomes is an essential task and a central challenge among artificial intelligence-based personalised learning applications. Despite several studies exploring student performance prediction, there is a notable lack of comprehensive and comparative research that methodically evaluates and compares multiple machine learning models alongside deep learning architectures. In response, our research provides a comprehensive comparison to evaluate and improve ten different machine learning and deep learning models, either well-established or cutting-edge techniques, namely, random forest, decision tree, support vector machine, K-nearest neighbours classifier, logistic regression, linear regression, and state-of-the-art extreme gradient boosting (XGBoost), as well as a fully connected feed-forward neural network, a convolutional neural network, and a gradient-boosted neural network. We implemented and fine-tuned these models using Python 3.9.5. With a keen emphasis on prediction accuracy and model performance optimisation, we evaluate these methodologies across two benchmark public student datasets. We employ a dual evaluation approach, utilising both k-fold cross-validation and holdout methods, to comprehensively assess the models’ performance. Our research focuses primarily on predicting student outcomes in final examinations by determining their success or failure. Moreover, we explore the importance of feature selection using the ubiquitous Lasso for dimensionality reduction to improve model efficiency, prevent overfitting, and examine its impact on prediction accuracy for each model, both with and without Lasso. This study provides valuable guidance for selecting and deploying predictive models for tabular data classification like student outcome prediction, which seeks to utilise data-driven insights for personalised education.

Systematic Review of Newborn Screening Programmes for Spinal Muscular Atrophy

Spinal muscular atrophy (SMA) is a genetic neuromuscular disorder causing the degeneration of motor neurons in the spinal cord. Recent studies suggest greater effectiveness of treatment in the presymptomatic stage. This systematic review synthesises findings from 37 studies (and 3 overviews) of newborn screening for SMA published up to November 2023 across 17 countries to understand the methodologies used; test accuracy performance; and timing, logistics and feasibility of screening. All studies screened for the homozygous deletion of SMN1 exon 7. Most (28 studies) used RT-PCR as the initial test on dried blood spots (DBSs), while nine studies also reported second-tier tests on DBSs for screen-positive cases. Babies testing positive on DBSs were referred for confirmatory testing via a range of methods. Observed SMA birth prevalence ranged from 1 in 4000 to 1 in 20,000. Most studies reported no false-negative or false-positive cases (therefore had a sensitivity and specificity of 100%). Five studies reported either one or two false-negative cases each (total of six cases; three compound heterozygotes and three due to system errors), although some false-negatives may have been missed due to lack of follow-up of negative results. Eleven studies reported false-positive cases, some being heterozygous carriers or potentially related to heparin use. Time to testing and treatment varied between studies. In conclusion, several countries have implemented newborn screening for SMA in the last 5 years using a variety of methods. Implementation considerations include processes for timely initial and confirmatory testing, partnerships between screening and neuromuscular centres, and timely treatment initiation.

A particle swarm optimization-based deep clustering algorithm for power load curve analysis

To address the inflexibility of the convolutional autoencoder (CAE) in adjusting the network structure and the difficulty of accurately delineating complex class boundaries in power load data, a particle swarm optimization deep clustering method (DC-PSO) is proposed. First, a particle swarm optimization algorithm for automatically searching the optimal network architecture and hyperparameters of CAE (AHPSO) is proposed to obtain better reconstruction performance. Then, an end-to-end deep clustering model based on a reliable sample selection strategy is designed for the deep clustering algorithm to accurately delineate the category boundaries and further improve the clustering effect. The experimental results show that the DC-PSO algorithm exhibits high clustering accuracy and higher performance for the power load profile clustering.

Few-Shot Conditional Learning: Automatic and Reliable Device Classification for Medical Test Equipment

The limited availability of specialized image databases (particularly in hospitals, where tools vary between providers) makes it difficult to train deep learning models. This paper presents a few-shot learning methodology that uses a pre-trained ResNet integrated with an encoder as a backbone to encode conditional shape information for the classification of neonatal resuscitation equipment from less than 100 natural images. The model is also strengthened by incorporating a reliability score, which enriches the prediction with an estimation of classification reliability. The model, whose performance is cross-validated, reached a median accuracy performance of over 99% (and a lower limit of 73.4% for the least accurate model/fold) using only 87 meta-training images. During the test phase on complex natural images, performance was slightly degraded due to a sub-optimal segmentation strategy (FastSAM) required to maintain the real-time inference phase (median accuracy 87.25% ). This methodology proves to be excellent for applying complex classification models to contexts (such as neonatal resuscitation) that are not available in public databases. Improvements to the automatic segmentation strategy prior to the extraction of conditional information will allow a natural application in simulation and hospital settings.

Advancements of Glucose Monitoring Biosensor: Current State, Generations of Technological Progress, and Innovation Dynamics.

Glucose monitoring is essential for managing diabetes, and continuous glucose monitoring biosensors can offer real-time monitoring with little invasiveness. However, challenges remain in improving sensor accuracy, selectivity, and overall performance. This article aims to review current trends and recent advancements in glucose-monitoring biosensors while evaluating their benefits and limitations for diabetes monitoring. An analysis of current literature on transdermal glucose sensors was conducted, focusing on detection techniques, novel nanomaterials, and integrated sensor systems. Recent research has led to advancements in electrochemical, optical, electromagnetic, and sonochemical sensors for transdermal glucose detection. The use of novel nanomaterials and integrated sensor designs has improved sensitivity, selectivity, and accuracy. However, issues like calibration requirements, motion artifacts, and skin irritation persist. Transdermal glucose sensors show promise for non-invasive, convenient diabetes monitoring but require further enhancements to address limitations in accuracy, reliability, and biocompatibility. Continued research and innovation focusing on sensor materials, designs, and surface chemistry is needed to optimize biosensor performance and utility. The study offers a comprehensive analysis of the present status of technological advancement and highlights areas that need more research.

Integrating metaheuristics and artificial intelligence for healthcare: basics, challenging and future directions

Accurate and rapid disease detection is necessary to manage health problems early. Rapid increases in data amount and dimensionality caused challenges in many disciplines, with the primary issues being high computing costs, memory costs, and low accuracy performance. These issues will arise since Machine Learning (ML) classifiers are mostly used in these fields. However, noisy and irrelevant features have an impact on ML accuracy. Therefore, to choose the best subset of features and decrease the dimensionality of the data, Metaheuristics (MHs) optimization algorithms are applied to Feature Selection (FS) using various modalities of medical imaging or disease datasets with different dimensions. The review starts by giving a general overview of the many approaches to AI algorithms, followed by a general overview of the various MH algorithms for healthcare applications, an analysis of MHs boosted AI for healthcare applications, and using a wide range of research databases as a data source for access to numerous field publications. The final section of this review discusses the problems and challenges facing healthcare application development.

Flexible Multimodal Sensors Enhanced by Electrospun Lead-Free Perovskite and PVDF-HFP Composite Form-Stable Mesh Membranes for In Situ Plant Monitoring.

The pH and humidity of the crop environment are essential indicators for monitoring crop growth status. This study reports a lead-free perovskite/polyvinylidene fluoride-hexafluoropropylene composite (LPPC) to enhance the stability and reliability of in situ plant pH and humidity monitoring. The mesh composite membrane of LPPC illustrates a hydrophobic contact angle of 101.982°, a tensile strain of 800%, and an opposing surface potential of less than -184.9 mV, which ensures fast response, high sensitivity, and stability of the sensor during long-term plant monitoring. The LPPC-coated pH electrode possesses a sensitivity of -63.90 mV/pH, which provides a fast response within 5 s and is inert to environmental temperature interference. The LPPC-coated humidity sensor obtains a sensitivity of -145.7 Ω/% RH, responds in 28 s, and works well under varying light conditions. The flexible multimodal sensor coated with an LPPC membrane completed real-time in situ monitoring of soilless strawberries for 17 consecutive days. Satisfactory consistency and accuracy performance are observed. The study provides a simple solution for developing reliable, flexible wearable multiparameter sensors for in situ monitoring of multiple parameters of crop environments.

Lane Line and Object Detection Using Yolo v3

In the contemporary age, creating autonomous vehi- cles is a crucial starting point for the advancement of intelligent transportation systems that rely on sophisticated telecommu- nications network infrastructure, including the upcoming 6g networks. The paper discusses two significant issues, namely, lane detection and obstacle detection (such as road signs, traffic lights, vehicles ahead, etc.) using image processing algorithms. To address issues like low accuracy in traditional image processing methods and slow real-time performance of deep learning-based methods, barriers for smart traffic lane and object detection algorithms are proposed. We initially rectify the distorted image resulting from the camera and then employ a threshold algorithm for the lane detection algorithm. The image is obtained by extracting a specific region of interest and applying an inverse perspective transform to obtain a top-down view. Finally, we apply the sliding window technique to identify pixels that belong to each lane and modify it to fit a quadratic equation. The Yolo algorithm is well-suited for identifying various types of obstacles, making it a valuable tool for solving identification problems. Finally, we utilize real-time videos and a straightforward dataset to conduct simulations for the proposed algorithm. The simula- tion outcomes indicate that the accuracy of the proposed method for lane detection is 97.91% and the processing time is 0.0021 seconds. The proposal for detecting obstacles has an accuracy rate of 81.90% and takes only 0.022 seconds to process. Compared to the conventional image processing technique, the proposed method achieves an average accuracy of 89.90% and execution time of 0.024 seconds, demonstrating a robust capability against noise. The findings demonstrate that the suggested algorithm can be implemented in self-driving car systems, allowing for efficient and fast processing of the advanced network.

Patient-ventilator asynchrony classification in mechanically ventilated patients: Model-based or machine learning method?

Background and objectivePatient-ventilator asynchrony (PVA) is associated with poor clinical outcomes and remains under-monitored. Automated PVA detection would enable complete monitoring standard observational methods do not allow. While model-based and machine learning PVA approaches exist, they have variable performance and can miss specific PVA events. This study compares a model and rule-based algorithm with a machine learning PVA method by retrospectively validating both methods using an independent patient cohort. MethodsHysteresis loop analysis (HLA) which is a rule-based method (RBM) and a tri-input convolutional neural network (TCNN) machine learning model are used to classify 7 different types of PVA, including: 1) flow asynchrony; 2) reverse triggering; 3) premature cycling; 4) double triggering; 5) delayed cycling; 6) ineffective efforts; and 7) auto triggering. Class activation mapping (CAM) heatmaps visualise sections of respiratory waveforms the TCNN model uses for decision making, improving result interpretability. Both PVA classification methods were used to classify incidence in an independent retrospective clinical cohort of 11 mechanically ventilated patients for validation and performance comparison. ResultsSelf-validation with the training dataset shows overall better HLA performance (accuracy, sensitivity, specificity: 97.5 %, 96.6 %, 98.1 %) compared to the TCNN model (accuracy, sensitivity, specificity: 89.5 %, 98.3 %, 83.9 %). In this study, the TCNN model demonstrates higher sensitivity in detecting PVA, but HLA was better at identifying non-PVA breathing cycles due to its rule-based nature. While the overall AI identified by both classification methods are very similar, the intra-patient distribution of each PVA type varies between HLA and TCNN. ConclusionThe collective findings underscore the efficacy of both HLA and TCNN in PVA detection, indicating the potential for real-time continuous monitoring of PVA. While ML methods such as TCNN demonstrate good PVA identification performance, it is essential to ensure optimal model architecture and diversity in training data before widespread uptake as standard care. Moving forward, further validation and adoption of RBM methods, such as HLA, offers an effective approach to PVA detection while providing clear distinction into the underlying patterns of PVA, better aligning with clinical needs for transparency, explicability, adaptability and reliability of these emerging tools for clinical care.

3D-CEUS/MRI-CEUS fusion imaging vs 2D-CEUS after locoregional therapies for hepatocellular carcinoma: a multicenter prospective study of therapeutic responseevaluation.

To compare the diagnostic accuracy of 3D contrast-enhanced ultrasound (CEUS)/MRI-CEUS fusion imaging with 2D-CEUS in assessing the response of hepatocellular carcinoma (HCC) to locoregional therapies in a multicenter prospective study. A consecutive series of patients with HCC scheduled for locoregional treatment were enrolled between April 2021 and March 2023. Patients were randomly divided into 3D-CEUS/MRI-CEUS fusion imaging group (3D/fusion group) or 2D-CEUS group (2D group). CEUS was performed 1 week before and 4-6 weeks after locoregional treatment. Contrast-enhanced MRI (CE-MRI) 4-6 weeks after treatment was set as the reference standard. CEUS images were evaluated for the presence or absence of viable tumors. Diagnostic performance criteria, including sensitivity, specificity, accuracy, and area under the curve (AUC), were determined for each modality. A total of 140 patients were included, 70 patients in the 2D group (mean age, 60.2 ± 10.4 years) and 70 patients in the 3D/fusion group (mean age, 59.8 ± 10.6 years). The sensitivity of the 3D/fusion group was 100.0% (95% CI: 75.9, 100.0), higher than that of the 2D group (55.6%, 95% CI: 22.7, 84.7; p = 0.019). The specificity of the 3D/fusion group was 96.3% (95% CI: 86.2, 99.4), which was comparable to that of the 2D group (98.4%, 95% CI: 90.0, 99.9; p = 0.915). The AUC of the 3D/fusion group was 0.98 (95% CI: 0.95, 1.00), higher than that of the 2D group (0.77, 95% CI: 0.56, 0.98; p = 0.020). 3D-CEUS/MRI-CEUS fusion imaging exhibits superior diagnostic accuracy in evaluating the treatment response to locoregional therapies for HCC. 3D-CEUS/MRI-CEUS fusion imaging can be applied for post-treatment assessment of residual tumors in HCC undergoing locoregional treatment, offering potential benefits in terms of accurate diagnosis and clinical management. Evaluating for HCC recurrence following locoregional therapy is important. 3D-CEUS/MRI-CEUS fusion imaging achieved a higher sensitivity than 2D-CEUS in assessing residual tumors after locoregional therapies. 3D-CEUS/MRI-CEUS fusion imaging can help clinicians intervene early in residual HCC lesions after locoregional treatment.

The use of image classification to estimate flamingo abundance from aerial, drone and satellite imagery

The Lesser Flamingo Phoeniconaias minor is a Near Threatened species known to be highly gregarious and that can concentrate in large numbers at core foraging and breeding sites, yet is also known to disperse widely in search of suitable foraging habitat. The density of flamingos at occupied sites, together with frequent dispersal between foraging sites, poses significant challenges to obtaining reliable counts and associated population estimates. Our study tested the use of unmanned aerial vehicles (drones) imagery and high-resolution satellite imagery for producing density estimates of Lesser Flamingos at the only known breeding site in South Africa (Kamfers Dam, Northern Cape Province). We implemented a supervised classification approach using multiple supervised classification algorithms, including xgbTree, randomForest, nnet and GBM. The four supervised classification models performed very well at classifying flamingos from both drone and satellite imagery, with classification accuracies ranging between 0.98 and 0.99. The inclusion of an additional predictor (other than RGB bands) yielded the highest variable importance score and greatly increased model accuracy and predictive performance. Estimates of adult flamingos from drone imagery acquired across nesting mudflats on 5 February (n = 1 320) and 14 February (n = 330) 2019 suggested a 75% decline in incubating adults. Furthermore, a total of 16 134 (15 973–16 295) flamingos were estimated from drone imagery on 1 March 2019, and 17 291 (17 118–17 290) flamingos from satellite imagery on 19 March 2019. The modelled size of the creche across dates suggested a loss of 234 birds (4.5%) during the period of assessment, with the highest count estimating 5 218 juveniles. Our study successfully implemented machine learning classifiers that were able to detect and derive highly accurate counts of Lesser Flamingos from both drone and satellite imagery. This use of machine learning could greatly improve the efficiency of monitoring and associated conservation efforts underway for the Lesser Flamingo.

The use of image classification to estimate flamingo abundance from aerial, drone and satellite imagery

The Lesser Flamingo Phoeniconaias minor is a Near Threatened species known to be highly gregarious and that can concentrate in large numbers at core foraging and breeding sites, yet is also known to disperse widely in search of suitable foraging habitat. The density of flamingos at occupied sites, together with frequent dispersal between foraging sites, poses significant challenges to obtaining reliable counts and associated population estimates. Our study tested the use of unmanned aerial vehicles (drones) imagery and high-resolution satellite imagery for producing density estimates of Lesser Flamingos at the only known breeding site in South Africa (Kamfers Dam, Northern Cape Province). We implemented a supervised classification approach using multiple supervised classification algorithms, including xgbTree, randomForest, nnet and GBM. The four supervised classification models performed very well at classifying flamingos from both drone and satellite imagery, with classification accuracies ranging between 0.98 and 0.99. The inclusion of an additional predictor (other than RGB bands) yielded the highest variable importance score and greatly increased model accuracy and predictive performance. Estimates of adult flamingos from drone imagery acquired across nesting mudflats on 5 February (n = 1 320) and 14 February (n = 330) 2019 suggested a 75% decline in incubating adults. Furthermore, a total of 16 134 (15 973–16 295) flamingos were estimated from drone imagery on 1 March 2019, and 17 291 (17 118–17 290) flamingos from satellite imagery on 19 March 2019. The modelled size of the creche across dates suggested a loss of 234 birds (4.5%) during the period of assessment, with the highest count estimating 5 218 juveniles. Our study successfully implemented machine learning classifiers that were able to detect and derive highly accurate counts of Lesser Flamingos from both drone and satellite imagery. This use of machine learning could greatly improve the efficiency of monitoring and associated conservation efforts underway for the Lesser Flamingo.

Accuracy of prediction from multi-environment trials for new locations using pedigree information and environmental covariates: the case of sorghum (Sorghum bicolor (L.) Moench) breeding

Key messagesWe investigate a method of extracting and fitting synthetic environmental covariates and pedigree information in multilocation trial data analysis to predict genotype performances in untested locations.Plant breeding trials are usually conducted across multiple testing locations to predict genotype performances in the targeted population of environments. The predictive accuracy can be increased by the use of adequate statistical models. We compared linear mixed models with and without synthetic covariates (SCs) and pedigree information under the identity, the diagonal and the factor-analytic variance-covariance structures of the genotype-by-location interactions. A comparison was made to evaluate the accuracy of different models in predicting genotype performances in untested locations using the mean squared error of predicted differences (MSEPD) and the Spearman rank correlation between predicted and adjusted means. A multi-environmental trial (MET) dataset evaluated for yield performance in the dry lowland sorghum (Sorghum bicolor (L.) Moench) breeding program of Ethiopia was used. For validating our models, we followed a leave-one-location-out cross-validation strategy. A total of 65 environmental covariates (ECs) obtained from the sorghum test locations were considered. The SCs were extracted from the ECs using multivariate partial least squares analysis and subsequently fitted in the linear mixed model. Then, the model was extended accounting for pedigree information. According to the MSEPD, models accounting for SC improve predictive accuracy of genotype performances in the three of the variance-covariance structures compared to others without SC. The rank correlation was also higher for the model with the SC. When the SC was fitted, the rank correlation was 0.58 for the factor analytic, 0.51 for the diagonal and 0.46 for the identity variance-covariance structures. Our approach indicates improvement in predictive accuracy with SC in the context of genotype-by-location interactions of a sorghum breeding in Ethiopia.

Selection of pre-trained weights for transfer learning in automated cytomegalovirus retinitis classification

Cytomegalovirus retinitis (CMVR) is a significant cause of vision loss. Regular screening is crucial but challenging in resource-limited settings. A convolutional neural network is a state-of-the-art deep learning technique to generate automatic diagnoses from retinal images. However, there are limited numbers of CMVR images to train the model properly. Transfer learning (TL) is a strategy to train a model with a scarce dataset. This study explores the efficacy of TL with different pre-trained weights for automated CMVR classification using retinal images. We utilised a dataset of 955 retinal images (524 CMVR and 431 normal) from Siriraj Hospital, Mahidol University, collected between 2005 and 2015. Images were processed using Kowa VX-10i or VX-20 fundus cameras and augmented for training. We employed DenseNet121 as a backbone model, comparing the performance of TL with weights pre-trained on ImageNet, APTOS2019, and CheXNet datasets. The models were evaluated based on accuracy, loss, and other performance metrics, with the depth of fine-tuning varied across different pre-trained weights. The study found that TL significantly enhances model performance in CMVR classification. The best results were achieved with weights sequentially transferred from ImageNet to APTOS2019 dataset before application to our CMVR dataset. This approach yielded the highest mean accuracy (0.99) and lowest mean loss (0.04), outperforming other methods. The class activation heatmaps provided insights into the model's decision-making process. The model with APTOS2019 pre-trained weights offered the best explanation and highlighted the pathologic lesions resembling human interpretation. Our findings demonstrate the potential of sequential TL in improving the accuracy and efficiency of CMVR diagnosis, particularly in settings with limited data availability. They highlight the importance of domain-specific pre-training in medical image classification. This approach streamlines the diagnostic process and paves the way for broader applications in automated medical image analysis, offering a scalable solution for early disease detection.

Adaptive GN block-based model for seismic response prediction of train-bridge coupled systems

Deep learning (DL) methods has found extensive application in predicting structural responses. However, conventional neural networks often lack the capacity to effectively capture inter-data relationships, relying solely on training data for end-to-end prediction. Consequently, these methods may struggle with generalization, particularly when faced with new structural configurations absent from the training dataset. To overcome this limitation, this study adopts a graph-based framework, representing structures as graphs facilitates the transfer of features between different components, mimicking the transfer of forces in the real world. Subsequently, a Graph Neural Network Blocks (GN Blocks) model is introduced. Leveraging the holistic inference structure inherent in graph representations, the GN Block-based model as an adaptive model exhibits superior generalization performance in response analysis tasks involving unknown structural forms. Illustrated through the application to train-bridge coupled (TBC) systems, the GN Block-based model demonstrates high prediction accuracy and robust generalization performance, underscoring its suitability for tackling complex structural response prediction.

Explainable artificial intelligence (XAI) for predicting the need for intubation in methanol-poisoned patients: a study comparing deep and machine learning models

The need for intubation in methanol-poisoned patients, if not predicted in time, can lead to irreparable complications and even death. Artificial intelligence (AI) techniques like machine learning (ML) and deep learning (DL) greatly aid in accurately predicting intubation needs for methanol-poisoned patients. So, our study aims to assess Explainable Artificial Intelligence (XAI) for predicting intubation necessity in methanol-poisoned patients, comparing deep learning and machine learning models. This study analyzed a dataset of 897 patient records from Loghman Hakim Hospital in Tehran, Iran, encompassing cases of methanol poisoning, including those requiring intubation (202 cases) and those not requiring it (695 cases). Eight established ML (SVM, XGB, DT, RF) and DL (DNN, FNN, LSTM, CNN) models were used. Techniques such as tenfold cross-validation and hyperparameter tuning were applied to prevent overfitting. The study also focused on interpretability through SHAP and LIME methods. Model performance was evaluated based on accuracy, specificity, sensitivity, F1-score, and ROC curve metrics. Among DL models, LSTM showed superior performance in accuracy (94.0%), sensitivity (99.0%), specificity (94.0%), and F1-score (97.0%). CNN led in ROC with 78.0%. For ML models, RF excelled in accuracy (97.0%) and specificity (100%), followed by XGB with sensitivity (99.37%), F1-score (98.27%), and ROC (96.08%). Overall, RF and XGB outperformed other models, with accuracy (97.0%) and specificity (100%) for RF, and sensitivity (99.37%), F1-score (98.27%), and ROC (96.08%) for XGB. ML models surpassed DL models across all metrics, with accuracies from 93.0% to 97.0% for DL and 93.0% to 99.0% for ML. Sensitivities ranged from 98.0% to 99.37% for DL and 93.0% to 99.0% for ML. DL models achieved specificities from 78.0% to 94.0%, while ML models ranged from 93.0% to 100%. F1-scores for DL were between 93.0% and 97.0%, and for ML between 96.0% and 98.27%. DL models scored ROC between 68.0% and 78.0%, while ML models ranged from 84.0% to 96.08%. Key features for predicting intubation necessity include GCS at admission, ICU admission, age, longer folic acid therapy duration, elevated BUN and AST levels, VBG_HCO3 at initial record, and hemodialysis presence. This study as the showcases XAI's effectiveness in predicting intubation necessity in methanol-poisoned patients. ML models, particularly RF and XGB, outperform DL counterparts, underscoring their potential for clinical decision-making.

Explainable artificial intelligence (XAI) for predicting the need for intubation in methanol-poisoned patients: a study comparing deep and machine learning models.

The need for intubation in methanol-poisoned patients, if not predicted in time, can lead to irreparable complications and even death. Artificial intelligence (AI) techniques like machine learning (ML) and deep learning (DL) greatly aid in accurately predicting intubation needs for methanol-poisoned patients. So, our study aims to assess Explainable Artificial Intelligence (XAI) for predicting intubation necessity in methanol-poisoned patients, comparing deep learning and machine learning models. This study analyzed a dataset of 897 patient records from Loghman Hakim Hospital in Tehran, Iran, encompassing cases of methanol poisoning, including those requiring intubation (202 cases) and those not requiring it (695 cases). Eight established ML (SVM, XGB, DT, RF) and DL (DNN, FNN, LSTM, CNN) models were used. Techniques such as tenfold cross-validation and hyperparameter tuning were applied to prevent overfitting. The study also focused on interpretability through SHAP and LIME methods. Model performance was evaluated based on accuracy, specificity, sensitivity, F1-score, and ROC curve metrics. Among DL models, LSTM showed superior performance in accuracy (94.0%), sensitivity (99.0%), specificity (94.0%), and F1-score (97.0%). CNN led in ROC with 78.0%. For ML models, RF excelled in accuracy (97.0%) and specificity (100%), followed by XGB with sensitivity (99.37%), F1-score (98.27%), and ROC (96.08%). Overall, RF and XGB outperformed other models, with accuracy (97.0%) and specificity (100%) for RF, and sensitivity (99.37%), F1-score (98.27%), and ROC (96.08%) for XGB. ML models surpassed DL models across all metrics, with accuracies from 93.0% to 97.0% for DL and 93.0% to 99.0% for ML. Sensitivities ranged from 98.0% to 99.37% for DL and 93.0% to 99.0% for ML. DL models achieved specificities from 78.0% to 94.0%, while ML models ranged from 93.0% to 100%. F1-scores for DL were between 93.0% and 97.0%, and for ML between 96.0% and 98.27%. DL models scored ROC between 68.0% and 78.0%, while ML models ranged from 84.0% to 96.08%. Key features for predicting intubation necessity include GCS at admission, ICU admission, age, longer folic acid therapy duration, elevated BUN and AST levels, VBG_HCO3 at initial record, and hemodialysis presence. This study as the showcases XAI's effectiveness in predicting intubation necessity in methanol-poisoned patients. ML models, particularly RF and XGB, outperform DL counterparts, underscoring their potential for clinical decision-making.