ResNet Architecture Research Articles

Deep learning radiomics based on contrast enhanced MRI for preoperatively predicting early recurrence in hepatocellular carcinoma after curative resection

PurposeTo explore the role of deep learning (DL) and radiomics-based integrated approach based on contrast enhanced magnetic resonance imaging (CEMRI) for predicting early recurrence (ER) in hepatocellular carcinoma (HCC) patients after curative resection.MethodsTotal 165 HCC patients (ER, n = 96 vs. non-early recurrence (NER), n = 69) were retrospectively collected and divided into a training cohort (n = 132) and a validation cohort (n = 33). From pretreatment CEMR images, a total of 3111 radiomics features were extracted, and radiomics models were constructed using five machine learning classifiers (logistic regression, support vector machine, k-nearest neighbor, extreme gradient Boosting, and multilayer perceptron). DL models were established via three variations of ResNet architecture. The clinical-radiological (CR), radiomics combined with clinical-radiological (RCR), and deep learning combined with RCR (DLRCR) models were constructed. Model discrimination, calibration, and clinical utilities were evaluated by receiver operating characteristic curve, calibration curve, and decision curve analysis, respectively. The best-performing model was compared with the widely used staging systems and preoperative prognostic indexes.ResultsThe RCR model (area under the curve (AUC): 0.841 and 0.811) and the optimal radiomics model (AUC: 0.839 and 0.804) achieved better performance than the CR model (AUC: 0.662 and 0.752) in the training and validation cohorts, respectively. The optimal DL model (AUC: 0.870 and 0.826) outperformed the radiomics model in the both cohorts. The DL, radiomics, and CR predictors (aspartate aminotransferase (AST) and tumor diameter) were combined to construct the DLRCR model. The DLRCR model presented the best performance over any model, yielding an AUC, an accuracy, a sensitivity, a specificity of 0.917, 0.886, 0.889, and 0.882 in the training cohort and of 0.844, 0.818, 0.800, and 0.846 in the validation cohort, respectively. The DLRCR model achieved better clinical utility compared to the clinical staging systems and prognostic indexes.ConclusionBoth radiomics and DL models derived from CEMRI can predict HCC recurrence, and DL and radiomics-based integrated approach can provide a more effective tool for the precise prediction of ER for HCC patients undergoing resection.

A method for enhancing near-mirror object detection by integrating AWCS and CBAM into ResNet18

ABSTRACT Defects, such as scratches, frequently arise during the manufacturing of near-mirror workpieces, adversely affecting both their longevity and aesthetic appeal. The identification of these near-mirror defects through machine vision is fraught with challenges, including the scarcity of comprehensive datasets, variability in surface curvature, and the diminutive size of the defects. In response to these challenges, this study integrates Visual Parser and deep learning methodologies for the visual detection of near-mirror defects. A dataset conducive to deep learning was developed utilising a sliding window approach. The proposed model incorporates an Adaptive Weighted Convolution Structure (AWCS) and Convolutional Block Attention Module (CBAM) within the ResNet18 architecture. Additionally, we introduce an efficient and precise technique termed Visual Parser, which facilitates the detection and localisation of defect areas in near-mirror workpieces. When combined with a classification model, the Visual Parser allows for the online inference and detection of extensive near-mirror images. Experimental findings indicate that the performance metrics of the proposed ResNet18-AWCS-CBAM (ResNet18-ASCM) model significantly exceed those of established networks. The defect detection and localisation method based on Visual Parser is capable of fulfilling the online inspection requirements pertinent to the production of near-mirror workpieces.

Deep feature response discriminative calibration

Deep neural networks (DNNs) have numerous applications across various domains. Several optimization techniques, such as ResNet and SENet, have been proposed to improve model accuracy. These techniques improve the model performance by adjusting or calibrating feature responses according to a uniform standard. However, they lack the discriminative calibration for different features, thereby introducing limitations in the model output. Therefore, we propose a method that discriminatively calibrates feature responses. The preliminary experimental results indicate that the neural feature response follows a Gaussian distribution. Consequently, we compute confidence values by employing the Gaussian probability density function, and then integrate these values with the original response values. The objective of this integration is to improve the feature discriminability of the neural feature response. Based on the calibration values, we propose a plugin-based calibration module incorporated into a modified ResNet architecture, termed Response Calibration Networks (ResCNet). Extensive experiments on datasets like CIFAR-10, CIFAR-100, SVHN, and ImageNet demonstrate the effectiveness of the proposed approach.

DeepWhaleNet: Climate Change-Aware FFT-Based Deep Neural Network for Passive Acoustic Monitoring

Climate change poses severe risks to the survival of many whale populations, whose habitats and migration patterns are affected by environmental changes. To detect these whales effectively, especially in deep-sea environments, we need to use AI-based techniques to handle the acoustic diversity and variability of different species. However, current methods for whale detection are based on pre- and post-processing steps that reduce their efficiency and generalizability. To address this issue, we present DeepWhaleNet, a novel deep-learning model that automates whale detection in Underwater Passive Acoustic Monitoring datasets. DeepWhaleNet simplifies the detection process by extracting relevant features from raw log-power spectrograms and helps protect these threatened species by supporting conservation efforts. Our model uses a larger short-time Fourier transform as input and a custom ResNet-18 architecture for classification, which enables it to separate whale sounds from noise and capture their temporal and spectral characteristics. We evaluate the performance of DeepWhaleNet and show that it surpasses state-of-the-art methods, achieving an 8.3% improvement in the F-1 score and 21% higher average precision of binary relevance than the baseline method. Moreover, our model demonstrates its versatility and suitability for species-specific retrieval problems through an ablation study on multi-label retrieval problems and a 99.1% recall for Blue Whales.

Improved Speech Authenticity Detection in Chinese–English Bilingual Contexts

The rapid evolution of voice technology has heightened the need for robust detection systems to distinguish between authentic and tampered speech. Recent competitions have significantly advanced the development of countermeasures against spoofing attacks. However, while advancements in detection technologies have been notable, existing methods often focus on a single type of tampering and language. Our contribution lies in developing an improved model that integrates an enhanced ResNet architecture with an LSTM to improve the detection of tampered audio, particularly in challenging multilingual scenarios. In the experiments, we built a hybrid dataset from self-recording Chinese speech and public VCTK2 English samples, enhanced the ResNet model generalization capabilities, and evaluated our approach using the bilingual dataset. Experiment results demonstrate that the proposed approach achieves a superior performance with an equal error rate of 11.62%, even in the face of bilingual conditions, and, more importantly, outperforms the leading models from ASVSpoof 2021 and ADD 2022 competitions. We also employed advanced tampering techniques, including CycleGAN voice conversion and auto splicing, to simulate real-world tampering scenarios and verify the effectiveness of the proposed approach.

Automatic semantic segmentation of breast cancer in DCE-MRI using DeepLabV3+ with modified ResNet50

Research on breast cancer segmentation is essential due to its high prevalence as the most common cancer in women and its occurrence in men as well. Breast cancer involves abnormal cell growth in the breast, highlighting the importance of advanced imaging. Dynamic Contrast Enhanced Magnetic Resonance Imaging (DCE-MRI) is an effective technique for this purpose. Deep learning has significantly influenced medical imaging in recent years, especially in accurately segmenting tumors from MRI images. Two techniques have been proposed for breast tumor segmentation: Dilated ResNet50 (RN50D) and Parallel Layers Added ResNet50 (PLA-RN50). RN50D involves altering the dilation factor of the convolution layer within the residual block of ResNet50. PLA-RN50 entails the integration of parallel layers following the final residual block of the ResNet50 architecture. The modified architectures serve as the backbone for the DeepLabV3+ network. The DeepLabV3+ with RN50D or PLA-RN50D is a powerful and effective architecture that integrates deep feature extraction, multiscale spatial information, and precise segmentation to achieve high accuracy in lesion segmentation for breast DCE-MRI images. The proposed technique is tested on a QIN Breast DCE-MRI dataset comprising 233 images sourced from The Cancer Image Archive. The proposed method achieves a dice score of 0.92. The superior segmentation performance of DeepLabV3+ with PLA-RN50, as compared to its counterparts using ResNet18 and ResNet50, highlights the impactful modifications incorporated in PLA-RN50 for optimizing breast tumor segmentation.

Automated assessment of task-based performance of digital mammography and tomosynthesis systems using an anthropomorphic breast phantom and deep learning-based scoring.

Conventional metrics used for assessing digital mammography (DM) and digital breast tomosynthesis (DBT) image quality, including noise, spatial resolution, and detective quantum efficiency, do not necessarily predict how well the system will perform in a clinical task. A number of existing phantom-based methods have their own limitations, such as unrealistic uniform backgrounds, subjective scoring using humans, and regular signal patterns unrepresentative of common clinical findings. We attempted to address this problem with a realistic breast phantom with random hydroxyapatite microcalcifications and semi-automated deep learning-based image scoring. Our goal was to develop a methodology for objective task-based assessment of image quality for tomosynthesis and DM systems, which includes an anthropomorphic phantom, a detection task (microcalcification clusters), and automated performance evaluation using a convolutional neural network. Experimental 2D and pseudo-3D mammograms of an anthropomorphic inkjet-printed breast phantom with inserted microcalcification clusters were collected on clinical mammography systems to train a signal-present/signal-absent image classifier based on Resnet-18 architecture. In a separate validation study using simulations, this Resnet-18 classifier was shown to approach the performance of an ideal observer. Microcalcification detection performance was evaluated as a function of four dose levels using receiver operating characteristic (ROC) analysis [i.e., area under the ROC curve (AUC)]. To demonstrate the use of this evaluation approach for assessing different technologies, the method was applied to two different mammography systems, as well as to mammograms with re-binned pixels emulating a lower-resolution X-ray detector. Microcalcification detectability, as assessed by the deep learning classifier, was observed to vary with the exposure incident on the breast phantom for both DM and tomosynthesis. At full dose, experimental AUC was 0.96 (for DM) and 0.95 (for DBT), whereas at half dose, it dropped to 0.85 and 0.71, respectively. AUC performance on DM was significantly decreased with an effective larger pixel size obtained with re-binning. The task-based assessment approach also showed the superiority of a newer mammography system compared with an older system. An objective task-based methodology for assessing the image quality of mammography and tomosynthesis systems is proposed. Possible uses for this tool could be quality control, acceptance, and constancy testing, assessing the safety and effectiveness of new technology for regulatory submissions, and system optimization. The results from this study showed that the proposed evaluation method using a deep learning model observer can track differences in microcalcification signal detectability with varied exposure conditions.

Enhancing Semantic Segmentation in High-Resolution TEM Images: A Comparative Study of Batch Normalization and Instance Normalization

Abstract Integrating deep learning into image analysis for transmission electron microscopy (TEM) holds significant promise for advancing materials science and nanotechnology. Deep learning is able to enhance image quality, to automate feature detection, and to accelerate data analysis, addressing the complex nature of TEM datasets. This capability is crucial for precise and efficient characterization of details on the nano—and microscale, e.g., facilitating more accurate and high-throughput analysis of nanoparticle structures. This study investigates the influence of batch normalization (BN) and instance normalization (IN) on the performance of deep learning models for semantic segmentation of high-resolution TEM images. Using U-Net and ResNet architectures, we trained models on two different datasets. Our results demonstrate that IN consistently outperforms BN, yielding higher Dice scores and Intersection over Union metrics. These findings underscore the necessity of selecting appropriate normalization methods to maximize the performance of deep learning models applied to TEM images.

StructNet-DDI: Molecular Structure Characterization-Based ResNet for Prediction of Drug-Drug Interactions.

This study introduces a deep learning framework based on SMILES representations of chemical structures to predict drug-drug interactions (DDIs). The model extracts Morgan fingerprints and key molecular descriptors, transforming them into raw graphical features for input into a modified ResNet18 architecture. The deep residual network, enhanced with regularization techniques, efficiently addresses training issues such as gradient vanishing and exploding, resulting in superior predictive performance. Experimental results show that StructNet-DDI achieved an AUC of 99.7%, an accuracy of 94.4%, and an AUPR of 99.9%, demonstrating the model's effectiveness and reliability. These findings highlight that StructNet-DDI can effectively extract crucial features from molecular structures, offering a simple yet robust tool for DDI prediction.

Classification of Breast Cancer Tumors from Histopathological Images through a Modified ResNet-50 Architecture

The diagnosis of malignant or benign breast cancer tumors from histopathological images is challenging due to human error, which may lead to the patient undergoing additional, often painful, procedures to collect new data. Utilizing a supervised, pre-trained ResNet-50 model as a second opinion for doctors can help eliminate the need for repeated procedures. One main challenge faced by doctors and machine learning models is image blurriness. Applying various data preprocessing and augmentation techniques, such as resizing, Gaussian blurring, histogram equalization, and color space conversions, can improve the model’s performance. The model achieved its best results with an accuracy of 95.61%, precision of 96%, recall of 94%, and an F1-score of 95%.

Improved Breast Cancer Detection using Modified ResNet50-Based on Gradient-Weighted Class Activation Mapping

Breast cancer is a prevalent and serious disease that affects many women around the world every year. Detecting breast cancer early is crucial for improving survival rates and treating the illness effectively. Researchers are exploring various methods, including neural networks and machine learning, to assist in detecting the disease. However, due to limited data availability, leveraging pre-trained models trained on diverse image datasets has become a common practice. This article introduces a novel approach to identifying breast cancer that involves the utilization of a deep learning model utilizing the ResNet50 framework, coupled with heat mapping and gradient-weighted class activation mapping (Grad-CAM). The suggested method was primarily assessed using the FDDM dataset of Subtracted Contrast Enhanced Spectral Mammography (CDD-CESM) images. The outcomes from this model were then contrasted with those of five other well-known models: VGG16, VGG19, MobileNetV2, EfficientNet-B7, and standard ResNet50. The newly proposed model yielded an accuracy of 0.8920, which was better than the other models. Additionally, Grad-CAM showed nearly flawless feature extraction in a breast cancer classification assignment. In the discussion section, the suggested method was utilized with the MIAS dataset to ensure thoroughness and scalability, and to allow for comparison with prior research. The results demonstrated the effectiveness of the suggested approach, with an accuracy of 0.9830 achieved on the MIAS dataset, surpassing previous works. This study significantly enhances the improvement of breast cancer detection through the integration of deep learning, the ResNet50 architecture, and visualization methods including heat maps and Grad-CAM.

Deep learning on chest X-ray and computed tomography scans for detection of COVID-19 as a part of a network-centric digital health stack for future pandemics

Developing a reliable rapid screening protocol for highly infectious diseases like COVID-19 is of paramount interest since it facilitates the isolation of infected patients from the rest of the population. Reverse-transcription polymerase chain reaction (RT-PCR) test is presently the most widely accepted gold-standard test to detect COVID-19. In this method, the RNA of the virus is duplicated by a process called reverse transcription to form DNA for facilitating the copying process. Fluorescent dye is attached to the viral genetic material and copied billions of times through the process called polymerase chain reaction. Enhanced fluorescence is used to identify the presence of genetic material of the virus. These tests are time-consuming and have significant false negatives, i.e., a person with COVID-19 might be categorized as not having the virus. Large-scale RT-PCR testing has its own share of problems such as logistics, availability and affordability in underdeveloped nations, and reliability of the test results. Machine learning algorithms can act as a cheaper supplementary/alternative diagnostic tool for the testing process. In the current study, using publicly available chest X-ray image datasets, different convolutional neural network (CNN)-based models were developed for efficient identification of COVID-19 infected patients, and their efficacies were compared. Key innovations in training the CNNs are discussed. Our results indicate that EfficientNet, SeResNext, and ResNet are best at classifying normal, pneumonia and COVID-19 cases, respectively. The ResNet architecture with transfer learning performed best at detecting COVID-19 with an accuracy of 94%, a rate far superior to that in the RT-PCR test, which is typically in the range of 70 – 80%. This is particularly attractive as an additional noninvasive protocol since such technology-augmented detection is likely to help in reducing the psychological refractory period due to COVID-19 infections. Toward the healthy lung initiative in the post-COVID-19 era, we propose close coupling of the present diagnostic protocols with digital approaches to ensure more reliable personal care within the ambit of large-scale pandemic control mechanisms. Such integration with emerging technological tools can create a benchmark for the first line of defense against future global pandemics.

Investigating hunting in a protected area in Southeast Asia using passive acoustic monitoring with mobile smartphones and deep learning

Hunting is one of the most serious threats to wildlife populations. Guns can be used to hunt both terrestrial and arboreal species. While many methods and techniques have been developed to monitor wildlife populations, few techniques have been developed that can meet the demand needed to monitor threats to wildlife populations and support intervention activities; this is particularly true in Southeast Asia. Here, we used smartphones to record gunshots at 64 sites that were systematically spaced in Chu Mom Ray National Park, Vietnam from July to November 2019. Our specific goals were to: 1) quantify temporal and spatial patterns in gunshots in the national park; 2) investigate the correlation between gunshots and seven environmental variables including forest quality, elevation, and distance to the nearest village or ranger post; and 3) compare the performance of three convolutional neural network (CNN) architectures for automatically detecting gunshots. We manually annotated spectrograms of gunshots using Raven Pro 1.6. Using the manual detection approach, we identified 115 gunshots at 30 sites. The number of gunshots recorded at each recording site over two days varied from 0 to 11. On average, there were about 0.93 gunshots recorded per day per recording site. Hunting activity showed a strong temporal trend. The number of gunshots detected increased gradually from early morning, reaching the peak at noon, with the most gunshots recorded from 10:00 to 14:00 local time, equivalent to 0.18 gunshots per hour per site. No gunshots were detected from 22:00 to 04:00 in the morning. There were no strong relationships between the number of gunshots and all environmental variables, with all the correlation coefficients smaller than 0.3. Comparing three CNN architectures — AlexNet, VGG16, and ResNet18— implemented in the ‘torch for R’ ecosystem, we found that both AlexNet and VGG16 architectures led to acceptable performance for automated detection (F1 score > 0.80), but the ResNet18 architecture did not perform well for this task. We found low generalizability, as the models had relatively low performance on an open dataset from Belize (F1 score ∼ 0.52). To implement effective patrols, protected-area managers should make a hunting-risk map. The effort required for this task can be substantially reduced by using a combination of PAM and automated gunshot detection, however these approaches still require creation of a site-specific training and test dataset, along with manual verification of the detections.

Proposed method for against adversarial images based on ResNet architecture

Our world is becoming increasingly automated due to the application of deep learning/machine learning models to systems, but these systems are vulnerable to adversarial attacks, which create deceptive data to trick them. Without proper defenses, attackers can exploit deep learning systems in facial recognition, self-driving cars, and social media filters. Research on adversarial image generation and methods to against attacks is important. This paper proposes employing the ResNet architecture with adversarial training to against adversarial images. The model is tested on a Hybrid CIFAR-10 dataset, which is designed to improve robustness and accuracy by incorporating GAN-generated images. The proposed model achieves an accuracy of over 95%, which is better than three other state-of-the-art architectures VGG19_bn, ShuffleNetV2, and RepVGG_a2.

Adaptive prototype few-shot image classification method based on feature pyramid

Few-shot learning aims to enable machines to recognize unseen novel classes using limited samples akin to human capabilities. Metric learning is a crucial approach to addressing this challenge, with its performance primarily dependent on the effectiveness of feature extraction and prototype computation. This article introduces an Adaptive Prototype few-shot image classification method based on Feature Pyramid (APFP). APFP employs a novel feature extraction method called FResNet, which builds upon the ResNet architecture and leverages a feature pyramid structure to retain finer details. In the 5-shot scenario, traditional methods for computing average prototypes exhibit limitations due to the typically diverse and uneven distribution of samples, where simple means may inadequately reflect such diversity. To address this issue, APFP proposes an Adaptive Prototype method (AP) that dynamically computes class prototypes of the support set based on the similarity between support set samples and query samples. Experimental results demonstrate that APFP achieves 67.98% and 85.32% accuracy in the 5-way 1-shot and 5-way 5-shot scenarios on the MiniImageNet dataset, respectively, and 84.02% and 94.44% accuracy on the CUB dataset. These results indicate that the proposed APFP method addresses the few-shot learning problem.

Explainable hypergraphs for gait based Parkinson classification

Parkinson Disease (PD) classification using Vertical Ground Reaction Force (VGRF) sensors can help in unobtrusive detection and monitoring of PD patients. State-of-the-art (SOTA) research in PD classification reveals that Deep Learning (DL), at the expense of explainability, performs better than Shallow Learning (SL). In this paper, we introduce a novel explainable weighted hypergraph, where the interconnections of the SOTA features are exploited, leading to more discriminative derived features, and thereby, forming an SL arm. In parallel, we create a DL arm consisting of ResNet architecture to learn the spatio-temporal patterns of the VGRF signals. Probabilities of PD classification scores from the SL and the DL arms are adaptively fused to create a hybrid pipeline. The pipeline achieves an AUC value of 0.979 on the Physionet Parkinson Dataset. This AUC value is found to be superior to the SL as well as the DL arm used in isolation, yielding respective AUCs of 0.878 and 0.852. The proposed pipeline demonstrates explainability through improved permutation feature importance and contrasting examples of use cases, where incorrect misclassification of the DL arm gets rectified by the SL arm and vice versa. We further demonstrate that our solution achieves comparable performance with SOTA methods. To the best of our knowledge, this is the first approach to analyze PD classification with a hypergraph based xAI (Explainable Artificial Intelligence).

Visual interpretation of deep learning model in ECG classification: A comprehensive evaluation of feature attribution methods

Feature attribution methods can visually highlight specific input regions containing influential aspects affecting a deep learning model's prediction. Recently, the use of feature attribution methods in electrocardiogram (ECG) classification has been sharply increasing, as they assist clinicians in understanding the model's decision-making process and assessing the model's reliability. However, a careful study to identify suitable methods for ECG datasets has been lacking, leading researchers to select methods without a thorough understanding of their appropriateness. In this work, we conduct a large-scale assessment by considering eleven popular feature attribution methods across five large ECG datasets using a model based on the ResNet-18 architecture. Our experiments include both automatic evaluations and human evaluations. Annotated datasets were utilized for automatic evaluations and three cardiac experts were involved for human evaluations. We found that Guided Grad-CAM, particularly when its absolute values are utilized, achieves the best performance. When Guided Grad-CAM was utilized as the feature attribution method, cardiac experts confirmed that it can identify diagnostically relevant electrophysiological characteristics, although its effectiveness varied across the 17 different diagnoses that we have investigated.

SSG-Net: A Multi-Branch Fault Diagnosis Method for Scroll Compressors Using Swin Transformer Sliding Window, Shallow ResNet, and Global Attention Mechanism (GAM).

The reliable operation of scroll compressors is crucial for the efficiency of rotating machinery and refrigeration systems. To address the need for efficient and accurate fault diagnosis in scroll compressor technology under varying operating states, diverse failure modes, and different operating conditions, a multi-branch convolutional neural network fault diagnosis method (SSG-Net) has been developed. This method is based on the Swin Transformer, the Global Attention Mechanism (GAM), and the ResNet architecture. Initially, the one-dimensional time-series signal is converted into a two-dimensional image using the Short-Time Fourier Transform, thereby enriching the feature set for deep learning analysis. Subsequently, the method integrates the window attention mechanism of the Swin Transformer, the 2D convolution of GAM attention, and the shallow ResNet's two-dimensional convolution feature extraction branch network. This integration further optimizes the feature extraction process, enhancing the accuracy of fault feature recognition and sensitivity to data variability. Consequently, by combining the global and local features extracted from these three branch networks, the model significantly improves feature representation capability and robustness. Finally, experimental results on scroll compressor datasets and the CWRU dataset demonstrate diagnostic accuracies of 97.44% and 99.78%, respectively. These results surpass existing comparative models and confirm the model's superior recognition precision and rapid convergence capabilities in complex fault environments.

A Hybrid Machine Learning Approach: Analyzing Energy Potential and Designing Solar Fault Detection for an AIoT-Based Solar–Hydrogen System in a University Setting

This research aims to optimize the solar–hydrogen energy system at Kangwon National University’s Samcheok campus by leveraging the integration of artificial intelligence (AI), the Internet of Things (IoT), and machine learning. The primary objective is to enhance the efficiency and reliability of the renewable energy system through predictive modeling and advanced fault detection techniques. Key elements of the methodology include data collection from solar energy production and fault detection systems, energy potential analysis using Transformer models, and fault identification in solar panels using CNN and ResNet-50 architectures. The Transformer model was evaluated using metrics such as Mean Absolute Error (MAE), Mean Squared Error (MSE), and an additional variation of MAE (MAE2). Known for its ability to detect intricate time series patterns, the Transformer model exhibited solid predictive performance, with the MAE and MAE2 results reflecting consistent average errors, while the MSE pointed to areas with larger deviations requiring improvement. In fault detection, the ResNet-50 model outperformed VGG-16, achieving 85% accuracy and a 42% loss, as opposed to VGG-16’s 80% accuracy and 78% loss. This indicates that ResNet-50 is more adept at detecting and classifying complex faults in solar panels, although further refinement is needed to reduce error rates. This study demonstrates the potential for AI and IoT integration in renewable energy systems, particularly within academic institutions, to improve energy management and system reliability. Results suggest that the ResNet-50 model enhances fault detection accuracy, while the Transformer model provides valuable insights for strategic energy output forecasting. Future research could focus on incorporating real-time environmental data to improve prediction accuracy and developing automated AIoT-based monitoring systems to reduce the need for human intervention. This study provides critical insights into advancing the efficiency and sustainability of solar–hydrogen systems, supporting the growth of AI-driven renewable energy solutions in university settings.

Comparison of Faster R-CNN, YOLO, and SSD for Third Molar Angle Detection in Dental Panoramic X-rays.

The use of artificial intelligence algorithms (AI) has gained importance for dental applications in recent years. Analyzing AI information from different sensor data such as images or panoramic radiographs (panoramic X-rays) can help to improve medical decisions and achieve early diagnosis of different dental pathologies. In particular, the use of deep learning (DL) techniques based on convolutional neural networks (CNNs) has obtained promising results in dental applications based on images, in which approaches based on classification, detection, and segmentation are being studied with growing interest. However, there are still several challenges to be tackled, such as the data quality and quantity, the variability among categories, and the analysis of the possible bias and variance associated with each dataset distribution. This study aims to compare the performance of three deep learning object detection models-Faster R-CNN, YOLO V2, and SSD-using different ResNet architectures (ResNet-18, ResNet-50, and ResNet-101) as feature extractors for detecting and classifying third molar angles in panoramic X-rays according to Winter's classification criterion. Each object detection architecture was trained, calibrated, validated, and tested with three different feature extraction CNNs which are ResNet-18, ResNet-50, and ResNet-101, which were the networks that best fit our dataset distribution. Based on such detection networks, we detect four different categories of angles in third molars using panoramic X-rays by using Winter's classification criterion. This criterion characterizes the third molar's position relative to the second molar's longitudinal axis. The detected categories for the third molars are distoangular, vertical, mesioangular, and horizontal. For training, we used a total of 644 panoramic X-rays. The results obtained in the testing dataset reached up to 99% mean average accuracy performance, demonstrating the YOLOV2 obtained higher effectiveness in solving the third molar angle detection problem. These results demonstrate that the use of CNNs for object detection in panoramic radiographs represents a promising solution in dental applications.