Convolutional Neural Network Model Research Articles

FACNN: fuzzy-based adaptive convolution neural network for classifying COVID-19 in noisy CXR images.

COVID-19 detection using chest X-rays (CXR) has evolved as a significant method for early diagnosis of the pandemic disease. Clinical trials and methods utilize X-ray images with computer and intelligent algorithms to improve detection and classification precision. This article thus proposes a fuzzy-based adaptive convolution neural network (FACNN) model to improve the detection precision by confining the false rates. The feature extraction process between the successive regions is validated using a fuzzy process that classifies labeled and unknown pixels. The membership functions are derived based on high precision features for detection and false rate suppression process. The convolution neural network process is responsible for increasing detection precision through recurrent training based on feature availability. This availability analysis is verified using fuzzy derivatives under local variances. Based on variance-reduced features, the appropriate regions with labeled and unknown features are used for normal or infected classification. Thus, the proposed FACNN improves accuracy, precision, and feature extraction by 14.36%, 8.74%, and 12.35%, respectively. This model reduces the false rate and extraction time by 10.35% and 10.66%, respectively.

Automatic Optic Nerve Assessment From Transorbital Ultrasound Images: A Deep Learning-based Approach.

Transorbital Ultrasonography (TOS) is a promising imaging technology that can be used to characterize the structures of the optic nerve and the potential alterations that may occur in those structures as a result of an increase in intracranial pressure (ICP) or the presence of other disorders such as multiple sclerosis (MS) and hydrocephalus. In this paper, the primary objective is to develop a fully automated system that is capable of segmenting and calculating the diameters of structures that are associated with the optic nerve in TOS images. These structures include the optic nerve diameter sheath (ONSD) and the optic nerve diameter (OND). A fully convolutional neural network (FCN) model that has been pre-trained serves as the foundation for the segmentation method. The method that was developed was utilized to collect 464 different photographs from 110 different people, and it was accomplished with the assistance of four distinct pieces of apparatus. An examination was carried out to compare the outcomes of the automatic measurements with those of a manual operator. Both OND and ONSD have a typical inaccuracy of -0.12 0.32 mm and 0.14 0.58 mm, respectively, when compared to the operator. The Pearson correlation coefficient (PCC) for OND is 0.71, while the coefficient for ONSD is 0.64, showing that there is a positive link between the two measuring tools. A conclusion may be drawn that the technique that was developed is automatic, and the average error (AE) that was reached for the ONSD measurement is compatible with the ranges of inter-operator variability that have been discovered in the literature.

Toward confident prostate cancer detection using ultrasound: a multi-center study.

Deep learning-based analysis of micro-ultrasound images to detect cancerous lesions is a promising tool for improving prostate cancer (PCa) diagnosis. An ideal model should confidently identify cancer while responding with appropriate uncertainty when presented with out-of-distribution inputs that arise during deployment due to imaging artifacts and the biological heterogeneity of patients and prostatic tissue. Using micro-ultrasound data from 693 patients across 5 clinical centers who underwent micro-ultrasound guided prostate biopsy, we train and evaluate convolutional neural network models for PCa detection. To improve robustness to out-of-distribution inputs, we employ and comprehensively benchmark several state-of-the-art uncertainty estimation methods. PCa detection models achieve performance scores up to average AUROC with a 10-fold cross validation setup. Models with uncertainty estimation obtain expected calibration error scores as low as , indicating that confident predictions are very likely to be correct. Visualizations of the model output demonstrate that the model correctly identifies healthy versus malignant tissue. Deep learning models have been developed to confidently detect PCa lesions from micro-ultrasound. The performance of these models, determined from a large and diverse dataset, is competitive with visual analysis of magnetic resonance imaging, the clinical benchmark to identify PCa lesions for targeted biopsy. Deep learning with micro-ultrasound should be further studied as an avenue for targeted prostate biopsy.

Enhancing Wind Farm Reliability: A Field of View Enhanced Convolutional Neural Network-Based Model for Fault Diagnosis and Prevention

Wind farms play a crucial role in renewable energy generation, but their reliability is often compromised by complex environmental and equipment conditions. This study proposes a field of view enhanced convolutional neural network (CNN) model for fault diagnosis and prevention in wind farms. The model is developed by collecting and processing wind farm fault data and compared with support vector machine (SVM) and k-nearest neighbor (KNN) models. The results showed that the proposed CNN model outperformed the other models in terms of convergence speed (17 iterations to reach the minimum loss), fault diagnosis accuracy (99.3% and 99.2% for inner and outer circle faults, respectively), and stable power output improvement. The model’s application to maintenance scheduling and economic benefit analysis in a real wind farm case demonstrated its high consistency and accuracy in fault prediction and maintenance optimization. The proposed approach has the potential to enhance wind farm reliability, efficiency, and economy by enabling accurate fault diagnosis, early warning, and preventive maintenance.

Evaluating Deep Learning Techniques for Detecting Aneurysmal Subarachnoid Hemorrhage: A Comparative Analysis of Convolutional Neural Network and Transfer Learning Models.

Machine learning (ML) and deep learning (DL) techniques offer a promising multidisciplinary solution for subarachnoid hemorrhage (SAH) detection. The novel transfer learning approach mitigates the time constraints associated with the traditional techniques and demonstrates a superior performance. This study aims to evaluate the effectiveness of convolutional neural networks (CNNs) and CNN-based transfer learning models in differentiating between aneurysmal subarachnoid hemorrhage (aSAH) and nonaneurysmal subarachnoid hemorrhage (naSAH). Data from Istanbul Ümraniye Training and Research Hospital, which included 15,600 DICOM images from 123 patients with aSAH and 7,793 images from 80 patients with naSAH, were used. The study employed four models: Inception-V3, EfficientNetB4, single-layer CNN, and three-layer CNN. Transfer learning models were customized by modifying the last three layers and using the Adam optimizer. The models were trained on Google Collaboratory and evaluated based on metrics such as F-score, precision, recall, and accuracy. EfficientNetB4 demonstrated the highest accuracy (99.92%), with a better F-score (99.82%), recall (99.92%), and precision (99.90%) than the other methods. The single- and three-layer CNNs and the transfer learning models produced comparable results. No overfitting was observed, and robust models were developed. CNN-based transfer learning models can accurately diagnose the etiology of SAH from CT images and is a valuable tool for clinicians. This approach could reduce the need for invasive procedures such as digital subtraction angiography, leading to more efficient medical resource utilization and improved patient outcomes.

Amharic spoken digits recognition using convolutional neural network

Spoken digits recognition (SDR) is a type of supervised automatic speech recognition, which is required in various human–machine interaction applications. It is utilized in phone-based services like dialing systems, certain bank operations, airline reservation systems, and price extraction. However, the design of SDR is a challenging task that requires the development of labeled audio data, the proper choice of feature extraction method, and the development of the best performing model. Even if several works have been done for various languages, such as English, Arabic, Urdu, etc., there is no developed Amharic spoken digits dataset (AmSDD) to build Amharic spoken digits recognition (AmSDR) model for the Amharic language, which is the official working language of the government of Ethiopia. Therefore, in this study, we developed a new AmSDD that contains 12,000 utterances of 0 (Zaero) to 9 (zet’enyi) digits which were recorded from 120 volunteer speakers of different age groups, genders, and dialects who repeated each digit ten times. Mel frequency cepstral coefficients (MFCCs) and Mel-Spectrogram feature extraction methods were used to extract trainable features from the speech signal. We conducted different experiments on the development of the AmSDR model using the AmSDD and classical supervised learning algorithms such as Linear Discriminant Analysis (LDA), K-Nearest Neighbors (KNN), Support Vector Machine (SVM), and Random Forest (RF) as the baseline. To further improve the performance recognition of AmSDR, we propose a three layers Convolutional Neural Network (CNN) architecture with Batch normalization. The results of our experiments show that the proposed CNN model outperforms the baseline algorithms and scores an accuracy of 99% and 98% using MFCCs and Mel-Spectrogram features, respectively.

Driver Distraction Detection Using Machine Learning

The problem of driver distraction has become serious, and this is due to physical distractions that can be dangerous on the roads. Convolutional Neural Networks (CNNs) are used in this research to introduce a new approach to identifying physical distractions. To train the designed CNN model, a large dataset of labeled images that depict different physical distractions such as eating, texting, and manipulating controls is utilized. We make use of the spatial hierarchies learned by CNNs that enable us to extract important information for distraction identification with high efficiency. The experimental results validate the proposed method’s ability to recognize physical distractions in real driving scenarios effectively. Incorporating such models into in-car systems could therefore result in reduced distractions and consequently fewer accidents and additionally offer immediate warnings or interventions thus improving driver safety considerably. Key Words: Distraction, Machine Learning, Convolution Neural Network, Road accidents.

Composition and state prediction of lithium-ion cathode via convolutional neural network trained on scanning electron microscopy images

High-throughput materials research is strongly required to accelerate the development of safe and high energy-density lithium-ion battery (LIB) applicable to electric vehicle and energy storage system. The artificial intelligence, including machine learning with neural networks such as Boltzmann neural networks and convolutional neural networks (CNN), is a powerful tool to explore next-generation electrode materials and functional additives. In this paper, we develop a prediction model that classifies the major composition (e.g., 333, 523, 622, and 811) and different states (e.g., pristine, pre-cycled, and 100 times cycled) of various Li(Ni, Co, Mn)O2 (NCM) cathodes via CNN trained on scanning electron microscopy (SEM) images. Based on those results, our trained CNN model shows a high accuracy of 99.6% where the number of test set is 3840. In addition, the model can be applied to the case of untrained SEM data of NCM cathodes with functional electrolyte additives.

Automated detection of anterior crossbite on intraoral images and videos utilizing deep learning.

Malocclusion has emerged as a burgeoning global public health concern. Individuals with an anterior crossbite face an elevated risk of exhibiting characteristics such as a concave facial profile, negative overjet, and poor masticatory efficiency. In response to this issue, we proposed a convolutional neural network (CNN)-based model designed for the automated detection and classification of intraoral images and videos. A total of 1865 intraoral images were included in this study, 1493 (80%) of which were allocated for training and 372 (20%) for testing the CNN. Additionally, we tested the models on 10 videos, spanning a cumulative duration of 124 seconds. To assess the performance of our predictions, metrics including accuracy, sensitivity, specificity, precision, F1-score, area under the precision-recall (AUPR) curve, and area under the receiver operating characteristic (ROC) curve (AUC) were employed. The trained model exhibited commendable classification performance, achieving an accuracy of 0.965 and an AUC of 0.986. Moreover, it demonstrated superior specificity (0.992 vs. 0.978 and 0.956, P < 0.05) in comparison to assessments by two orthodontists. Conversely, the CNN model displayed diminished sensitivity (0.89 vs. 0.96 and 0.92, P < 0.05) relative to the orthodontists. Notably, the CNN model accomplished a perfect classification rate, successfully identifying 100% of the videos in the test set. The deep learning (DL) model exhibited remarkable classification accuracy in identifying anterior crossbite through both intraoral images and videos. This proficiency holds the potential to expedite the detection of severe malocclusions, facilitating timely classification for appropriate treatment and, consequently, mitigating the risk of complications.

Fabric pilling image segmentation by embedding dual-attention mechanism U-Net network

The initial step in fabric pilling rating is to segment and localize the pilling region, which is a crucial and challenging task. This paper presents a fabric puckering image segmentation method that is integrated into a U-Net network with a dual-attention mechanism. We have enhanced the fully convolutional neural network (U-Net) model by incorporating the dual-attention mechanism. This modification has resulted in a powerful feature extraction capability, enabling the objective and accurate segmentation of the fabric puckering region. We refer to this improved model as the dual-attention U-Net. The network model for fabric pilling feature extraction adopts the improved VGG16 model architecture as its encoding part. The model parameters are initialized with VGG16 pre-training weights to accelerate convergence speed. Second, the feature fusion between the corresponding layers of the encoding part and the decoding part was fed into the dual-attention mechanism module to strengthen the weight values of the fabric pilling region adaptively, which made the model more focused on the fabric pilling target region; Third, the dual-attention U-Net model was trained by data augmentation and migration learning strategies to prevent overfitting; Finally, the performance of the model was evaluated with the collected fabric pilling data set. The results of the experiments indicate that the claimed dual-attention U-Net model is superior to the typical U-Net model in terms of accuracy and precision. The dual-attention U-Net model achieved 99.03% accuracy for fabric pilling segmentation.

Deep Learning Approach for Gait Detection for Precise Stimulation of FES to Correct Foot Drop

Automatic detection of foot lift is one of the most important events of Functional Electrical Stimulation (FES). The FES system is used for the correction of Foot Drop (FD). FD is a condition where a person is unable to lift their foot from the ground due to complications that may arise after a stroke or spinal cord injury. It is crucial to accurately detect the patient’s foot lift event when correcting FD through FES as the pulse should only be applied when the person lifts their foot. The FES system applies the electrical pulse based on the input of the foot-lift detection sensor. A conventional FES system employs a sensor that is affixed on the heel to detect the lifting of the foot, but the connecting cables make the patient uncomfortable. To address this problem, IMU (Inertial Measurement Unit)-based sensors have been used, but they have some disadvantages, such as false triggering, low accuracy, and calibration. In this paper, we have presented an algorithm for detecting foot-lift events with high accuracy using a single IMU sensor through the application of deep learning techniques. We have recorded data from 10 healthy people and 10 foot drop patients. We have implemented Artificial Neural Network (ANN), K-Nearest Neighbour (KNN), and Convolutional Neural Network (CNN) models on these data and compared the results of these three models. The proposed algorithm aims to improve the precision of stimulation in the FES system.

A Deep Learning Model with Axial Attention for Radar Echo Extrapolation

ABSTRACT Radar echo extrapolation is an important approach in precipitation nowcasting which utilizes historical radar echo images to predict future echo images. In this paper, we introduce the self-attention mechanism into Trajectory Gated Recurrent Unit (TrajGRU) model. Under the sequence-to-sequence framework, we have developed a novel convolutional recurrent neural network called Self-attention Trajectory Gated Recurrent Unit (SA-TrajGRU), which incorporates the self-attention mechanism. The SA-TrajGRU model which combines spatiotemporal variant structure in TrajGRU and self-attention module is simple and effective. We evaluate our approach on the Moving MNIST-2 dataset and the CIKM AnalytiCup 2017 radar echo dataset. The experimental results show that the performance of the proposed SA-TrajGRU model is comparable to other convolutional recurrent neural network models. HSS and CSI scores of the SA-TrajGRU model are higher than scores of other models under the radar echo threshold of 25 dBZ, indicating that the SA-TrajGRU model has the most accurate prediction results under this threshold.

Soft computing machine learning applications for assessing regional-scale landslide susceptibility in the Nepal Himalaya

PurposeThis research is aimed at preparing landslide susceptibility using spatial analysis and soft computing machine learning techniques based on convolutional neural networks (CNNs), artificial neural networks (ANNs) and logistic regression (LR) models.Design/methodology/approachUsing the Geographical Information System (GIS), a spatial database including topographic, hydrologic, geological and landuse data is created for the study area. The data are randomly divided between a training set (70%), a validation (10%) and a test set (20%).FindingsThe validation findings demonstrate that the CNN model (has an 89% success rate and an 84% prediction rate). The ANN model (with an 84% success rate and an 81% prediction rate) predicts landslides better than the LR model (with a success rate of 82% and a prediction rate of 79%). In comparison, the CNN proves to be more accurate than the logistic regression and is utilized for final susceptibility.Research limitations/implicationsLand cover data and geological data are limited in largescale, making it challenging to develop accurate and comprehensive susceptibility maps.Practical implicationsIt helps to identify areas with a higher likelihood of experiencing landslides. This information is crucial for assessing the risk posed to human lives, infrastructure and properties in these areas. It allows authorities and stakeholders to prioritize risk management efforts and allocate resources more effectively.Social implicationsThe social implications of a landslide susceptibility map are profound, as it provides vital information for disaster preparedness, risk mitigation and landuse planning. Communities can utilize these maps to identify vulnerable areas, implement zoning regulations and develop evacuation plans, ultimately safeguarding lives and property. Additionally, access to such information promotes public awareness and education about landslide risks, fostering a proactive approach to disaster management. However, reliance solely on these maps may also create a false sense of security, necessitating continuous updates and integration with other risk assessment measures to ensure effective disaster resilience strategies are in place.Originality/valueLandslide susceptibility mapping provides a proactive approach to identifying areas at higher risk of landslides before any significant events occur. Researchers continually explore new data sources, modeling techniques and validation approaches, leading to a better understanding of landslide dynamics and susceptibility factors.

Optimizing Convolutional Neural Networks, XGBoost, and Hybrid CNN-XGBoost for Precise Red Tilapia (Oreochromis niloticus Linn.) Weight Estimation in River Cage Culture with Aerial Imagery

Accurate feeding management in aquaculture relies on assessing the average weight of aquatic animals during their growth stages. The traditional method involves using a labor-intensive approach and may impact the well-being of fish. The current research focuses on a unique way of estimating red tilapia’s weight in cage culture via a river, which employs unmanned aerial vehicle (UAV) and deep learning techniques. The described approach includes taking pictures by means of a UAV and then applying deep learning and machine learning algorithms to them, such as convolutional neural networks (CNNs), extreme gradient boosting (XGBoost), and a Hybrid CNN-XGBoost model. The results showed that the CNN model achieved its accuracy peak after 60 epochs, showing accuracy, precision, recall, and F1 score values of 0.748 ± 0.019, 0.750 ± 0.019, 0.740 ± 0.014, and 0.740 ± 0.019, respectively. The XGBoost reached its accuracy peak with 45 n_estimators, recording values of approximately 0.560 ± 0.000 for accuracy and 0.550 ± 0.000 for precision, recall, and F1. Regarding the Hybrid CNN-XGBoost model, it demonstrated its prediction accuracy using both 45 epochs and n_estimators. The accuracy value was around 0.760 ± 0.019, precision was 0.762 ± 0.019, recall was 0.754 ± 0.019, and F1 was 0.752 ± 0.019. The Hybrid CNN-XGBoost model demonstrated the highest accuracy compared to using standalone CNN and XGBoost models and could reduce the time required for weight estimation by around 11.81% compared to using the standalone CNN. Although the testing results may be lower than those from previous laboratory studies, this discrepancy is attributed to the real-world testing conditions in aquaculture settings, which involve uncontrollable factors. To enhance accuracy, we recommend increasing the sample size of images and extending the data collection period to cover one year. This approach allows for a comprehensive understanding of the seasonal effects on evaluation outcomes.

Detection and Classification of Wilting in Soybean Crop using Cutting-edge Deep Learning Techniques

Background: This paper employs deep learning in the classification of soybean wilting, a plant health indicator affected by external pressures, using a Convolutional Neural Network (CNN) with a pre-trained model. It highlights the promise of deep learning in agriculture by examining the relevance of wilting, evolution in the agricultural sector and applications in crop wellness monitoring. Methods: A CNN is used in the study to classify soybean withering, with special attention to the VGG16 pre-trained model. Deep learning’s ability to interpret complex data patterns is harnessed for intelligent and accurate wilting detection. A smart detection system tailored for soybean wilting is developed, incorporating recent advancements and addressing associated challenges. Result: The CNN model, notably VGG16, achieves 76% overall accuracy in distinguishing healthy and wilted soybean leaves, signifying a transformative shift in soybean crop health management. The approach offers a precise, efficient and sustainable solution supported by state-of-the-art CNN technology, advancing soybean cultivation practices.

Abstract PS09-09: Multiomics profiling and molecular classification refine precision treatment strategies for HER2-positive breast cancer

Abstract Background: Anti-HER2 targeted therapy has achieved a series of breakthroughs. However, the current treatment strategy regarding HER2-positive breast cancer remains indiscriminate and lacks specificity, which limits the further improvement of overall treatment response and may lead to overtreatment and extra cost for some patients. Our study aims to reveal the molecular heterogeneity of HER2-positive breast cancer to guide a more precise treatment. Patients and methods: We selected HER2-positive breast cancer patients treated at Fudan University Shanghai Cancer Center between 2013 and 2014 and conducted genomic, transcriptomic, proteomic and metabolomic profiling. We then applied a non-negative matrix factorization algorithm on transcriptomic data to obtain an unsupervised classification. And we further studied the correlation between subtypes and corresponding treatment strategies in multiple cohorts of adjuvant and neoadjuvant therapy. For clinical accessibility, we developed convolutional neural network models through deep learning algorithm based on digital pathology to identify different subtypes. Additionally, we explored novel treatment strategies using the patient-derived organoids (PDOs) models. Results: We established a novel multiomics cohort of 180 HER2- breast cancer patients and classified them into four clinically significant molecular subtypes: (1) A classical HER2-enriched (HER2-CLA, N=51) subtype characterized by strong ERBB2 signaling and remarkable sensitivity to anti-HER2-targeted therapy (pathologic complete response with dual-targeted therapy: 93%). (2) an immunomodulatory (HER2-IM, N=36) subtype characterized by an immune-activated microenvironment and excellent prognosis with current treatment (no relapse in 97% of patients with a median follow-up of 86 months). Tumors of this subtype were therefore candidates for de-escalatory treatment. (3) A luminal-like (HER2-LUM, N=55) subtype distinguished by activated estrogen receptor signaling and (4) a basal/mesenchymal-like (HER2-BM, N=38) subtype enriched in activated receptor tyrosine kinase pathways. HER2-LUM and HER2- BM showed limited benefit from anti-HER2 therapy, and thus, add-on therapies might be needed. The overall area under the curve (AUC) of the convolutional neural network model based on digital pathology for identifying different subtypes is 0.77. In the exploration of novel treatment strategies, we found in the PDO model that the HER2-LUM subtype is more sensitive to a treatment regimen combining standard (chemotherapy and targeted therapy) with subsequent endocrine therapy and CDK4/6 inhibitors compared to other subtypes. Additionally, the HER2-BM subtype demonstrated greater sensitivity to treatment with a combination of EGFR inhibitors, PDGFR inhibitors or VEGFR inhibitors. Conclusion: We uncovered a high degree of molecular heterogeneity in HER2-positive breast cancer and illustrated its impact on treatment response. More precise treatment can be given according to the characteristics of different subtypes, which may achieve good efficacy and simultaneously reduce overtreatment and extra cost. The comprehensive profiling of HER2-positive breast cancers could also serve as an important resource for further exploration. Key Words: HER2-positive breast cancer cohort; molecular classification; targeted therapy; precision treatment; de-escalatory treatment. Citation Format: Yu-Wei Li, Ding Ma, Xiang-Rong Wu, Lei-Jie Dai, Shen Zhao, Yu-Zheng Xu, Xi Jin, Xiao Yi, Ying Wang, Cai-Jin Lin, Yi-Fan Zhou, Tong Fu, Wen-Tao Yang, Ming Li, Hong Lv, Yi-Zhou Jiang, Zhi-Ming Shao. Multiomics profiling and molecular classification refine precision treatment strategies for HER2-positive breast cancer [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PS09-09.

Upsampling Monte Carlo Reactor Simulation Tallies in Depleted Sodium-Cooled Fast Reactor Assemblies Using a Convolutional Neural Network

The computational demand of neutron Monte Carlo transport simulations can increase rapidly with the spatial and energy resolution of tallied physical quantities. Convolutional neural networks have been used to increase the resolution of Monte Carlo simulations of light water reactor assemblies while preserving accuracy with negligible additional computational cost. Here, we show that a convolutional neural network can also be used to upsample tally results from Monte Carlo simulations of sodium-cooled fast reactor assemblies, thereby extending the applicability beyond thermal systems. The convolutional neural network model is trained using neutron flux tallies from 300 procedurally generated nuclear reactor assemblies simulated using OpenMC. Validation and test datasets included 16 simulations of procedurally generated assemblies, and a realistic simulation of a European sodium-cooled fast reactor assembly was included in the test dataset. We show the residuals between the high-resolution flux tallies predicted by the neural network and high-resolution Monte Carlo tallies on relative and absolute bases. The network can upsample tallies from simulations of fast reactor assemblies with diverse and heterogeneous materials and geometries by a factor of two in each spatial and energy dimension. The network’s predictions are within the statistical uncertainty of the Monte Carlo tallies in almost all cases. This includes test assemblies for which burnup values and geometric parameters were well outside the ranges of those in assemblies used to train the network.

Prediction model based on artificial intelligence for identifying risk of coronary atherosclerotic heart disease in computed tomography

ObjectiveTo analyze the application value of artificial intelligence (AI) in coronary computed tomography angiography (CCTA) image processing and diagnosis of coronary atherosclerotic heart disease (CHD). MethodsA total of 80 patients with suspected CHD in our hospital were selected for CCTA examination and blood lipid examination. The convolutional neural networks (CNN) model of coronary artery plaque detection was constructed, and the data set was randomly divided into training set and test set after pretreatment of lipid characteristics and image characteristics. The prediction efficiency and accuracy of the model were evaluated. ResultsIn the data set, the lipid indexes LDL-C, TC, and TG of patients in the CHD group were significantly higher than those in the Non-CHD group (P < 0.05). The average processing and diagnosis time of the AI model was (187.19 ± 18.79) s, which was significantly shorter than the average time of doctors (989.07 ± 50.40) s, and the difference was statistically significant (P < 0.05). There was no significant difference in the detection of calcified plaque, non-calcified plaque, and mixed plaque between doctors and AI models (P > 0.05). However, 5 plaques were misdiagnosed in the AI model (3.38%). The area under the curve (AUC) value of the CNN recognition model-based AI and manual recognition of doctors for the CHD were 0.870 (95% CI: 0.698–0.931) and 0.870 (95% CI: 0.691–0.926) (P < 0.001). ConclusionAI integrated with lipid parameters has certain clinical value in CCTA image processing efficiency and plaque diagnosis, and can be used as an effective auxiliary tool to analyze and diagnose CHD.

Vision-aided grasp classification: design and evaluation of compact CNN for prosthetic hands.

Powered prosthetic hands capable of executing various grasp patterns are highly sought-after solutions for upper limb amputees. A crucial requirement for such prosthetic hands is the accurate identification of the intended grasp pattern and subsequent activation of the prosthetic digits accordingly. Vision-based grasp classification techniques offer improved coordination between amputees and prosthetic hands without physical contact. Deep learning methods, particularly Convolutional Neural Networks (CNNs), are utilized to process visual information for classification. The key challenge lies in developing a model that can effectively generalize across various object shapes and accurately classify grasp classes. To address this, a compact CNN model named GraspCNet is proposed, specifically designed for grasp classification in prosthetic hands. The use of separable convolutions reduces the computational burden, making it potentially suitable for real-time applications on embedded systems. The GraspCNet model is designed to learn and generalize from object shapes, allowing it to effectively classify unseen objects beyond those included in the training dataset. The proposed model was trained and tested using various standard object data sets. A cross-validation strategy has been adopted to perform better in seen and unseen object class scenarios. The average accuracy achieved was 82.22% and 75.48% in the case of seen, and unseen object classes respectively. In computer-based real-time experiments, the GraspCNet model achieved an accuracy of 69%. A comparative analysis with state-of-the-art techniques revealed that the proposed GraspCNet model outperformed most benchmark techniques and demonstrated comparable performance with the DcnnGrasp method. The compact nature of the GraspCNet model suggests its potential for integration with other sensing modalities in prosthetic hands.

Epileptic Seizure Detection from Decomposed EEG Signal through 1D and 2D Feature Representation and Convolutional Neural Network

Electroencephalogram (EEG) has emerged as the most favorable source for recognizing brain disorders like epileptic seizure (ES) using deep learning (DL) methods. This study investigated the well-performed EEG-based ES detection method by decomposing EEG signals. Specifically, empirical mode decomposition (EMD) decomposes EEG signals into six intrinsic mode functions (IMFs). Three distinct features, namely, fluctuation index, variance, and ellipse area of the second order difference plot (SODP), were extracted from each of the IMFs. The feature values from all EEG channels were arranged in two composite feature forms: a 1D (i.e., unidimensional) form and a 2D image-like form. For ES recognition, the convolutional neural network (CNN), the most prominent DL model for 2D input, was considered for the 2D feature form, and a 1D version of CNN was employed for the 1D feature form. The experiment was conducted on a benchmark CHB-MIT dataset as well as a dataset prepared from the EEG signals of ES patients from Prince Hospital Khulna (PHK), Bangladesh. The 2D feature-based CNN model outperformed the other 1D feature-based models, showing an accuracy of 99.78% for CHB-MIT and 95.26% for PHK. Furthermore, the cross-dataset evaluations also showed favorable outcomes. Therefore, the proposed method with 2D composite feature form can be a promising ES detection method.