Knee Osteoarthritis (KOA), the most prevalent joint disease, significantly impacts elderly mobility due to progressive cartilage degeneration. Early prediction is crucial for preventing disease progression and guiding effective treatment plans. This paper proposes an EnsembleTL-ACO, fully automated, computer-aided diagnosis (CAD) system for accurate and rapid KOA severity grading. The proposed CAD system leverages an ensemble transfer learning strategy to extract robust deep features by fusing multiple deep learning models. It combines features from two consecutive AI models: (1) AlexNet for implicit class-wise deep feature extraction from preprocessed data, and (2) a custom IsrNet for further feature depth. Unsupervised k-means clustering based on PCA dimensionality reduction decomposes each class into subgroups, further refining features. Finally, Ant Colony Optimization (ACO) selects the most informative features. Evaluated on the Osteoarthritis Initiative (OAI) dataset, the proposed system achieves high accuracy in classifying the five KOA severity grades. With 1000 optimized features, it reaches average overall accuracies of 89.89

Various diseases may affect the gastrointestinal (GI) tract, which can be transformed into fatal cancerous infections if not diagnosed in time. Early recognition and treatment of these diseases can reduce the mortality rate. The GI tract datasets have challenges due to their intra-class variabilities, such as texture, size, and color, and inter-class similarities. To address these challenges, this study proposed a framework based on the concatenation of deep features and Histogram of Oriented Gradients (HOG) features. First, a 112-layered Convolutional Neural Network (CNN) model is developed, named GIK-Net. A total of six residual blocks are utilized with inverted bottleneck layers of 1 & times; 1, 3 & times; 3, and 1 & times; 1 convolutions. In each residual block, the output of the previous 5 & times; 5 convolution is combined with the output of inverted bottleneck layers, and a 3 & times; 3 convolution using a skip connection. The GIK-Net is trained from scratch, and then activations are performed on the first Fully Connected (FC1) layer to extract deep features. Second, the classical HOG features are extracted, and the top-value 200 features are selected. Third, the extracted deep and HOG features are serially concatenated and given to Machine Learning (ML) classifiers. After extensive experimentation, the classification accuracy of 99.30%, 99.40%, 99.60%, and 99.10% is achieved on the Kvasir-V1, Kvasir-V2, Hyper-Kvasir, and GastroEndoNet datasets, respectively. The framework outperformed comparative methods with improved accuracies of 2.87% on Kvasir-V1, 0.6% on Kvasir-V2, 1.07% on Hyper-Kvasir, and 18.10% on GastroEndoNet. Overall classification performance is promising according to the results.

Surveillance systems are integral to ensuring public safety by detecting unusual incidents, yet existing methods often struggle with accuracy and robustness. This study introduces an advanced framework for anomaly recognition in surveillance, leveraging deep learning to address these challenges and achieve significant improvements over current techniques. The framework begins with preprocessing input images using histogram equalization to enhance feature visibility. It then employs two DCNNs for feature extraction: a novel 63-layer CNN, "Up-to-the-Minute-Net," and the established Inception-Resnet-v2. The features extracted by both models are fused and optimized through two sophisticated feature selection techniques: Dragonfly and Genetic Algorithm (GA). The optimization process involves rigorous experimentation with 5- and 10-fold cross-validation to evaluate performance across various feature sets. The proposed approach achieves an unprecedented 99.9% accuracy in 5-fold cross-validation using the GA optimizer with 2500 selected features, demonstrating a substantial leap in accuracy compared to existing methods. This study's contribution lies in its innovative combination of deep learning models and advanced feature optimization techniques, setting a new benchmark in the field of anomaly recognition for surveillance systems and showcasing the potential for practical real-world applications.

[This corrects the article DOI: 10.1371/journal.pone.0312177.].

VGG-16 and Inception are widely used CNN architectures for image classification, but they face challenges in target categorization. This study introduces B4-GraftingNet, a novel deep learning model that integrates VGG-16's hierarchical feature extraction with Inception's diversified receptive field strategy. The model is trained on the OCT-CXR dataset and evaluated on the NIH-CXR dataset to ensure robust generalization. Unlike conventional approaches, B4-GraftingNet incorporates binary particle swarm optimization (BPSO) for feature selection and grad-CAM for interpretability. Additionally, deep feature extraction is performed, and multiple machine learning classifiers (SVM, KNN, random forest, na & iuml;ve Bayes) are evaluated to determine the optimal feature representation. The model achieves 94.01% accuracy, 94.22% sensitivity, 93.36% specificity, and 95.18% F1-score on OCT-CXR and maintains 87.34% accuracy on NIH-CXR despite not being trained on it. These results highlight the model's superior classification performance, feature adaptability, and potential for real-world deployment in both medical and general image classification tasks.

Various diseases, such as colon cancer, gastric cancer, celiac, and bleeding, pose a significant risk to the gastrointestinal (GI) tract, which serves as a fundamental component of the human body. It is less invasive to observe the inner part for disease recognition by using endoscopy and colonoscopy devices. Gastroenterologists consider the increased frame rate in video endoscopy to be challenging when it comes to identifying pathological findings. The detailed examination requires an experienced gastroenterologist. The ordinary procedure takes much time in disease classification. A machine-learning-based computer-aided diagnostic system (CADx) is in high demand for helping Gastroenterologists diagnose GI tract diseases with high accuracy (Acc). CADx takes very little time in diagnosing diseases and supports the training of clinicians. With the assistance of a gastroenterologist, CADx has an impact on reducing the mortality rate by finding diseases in their early stages. In an extensive examination of CADx, the focus is placed on ailments affecting the GI tract, various imaging methods, as well as diverse forms of CADx and techniques. These encompass preprocessing, feature extraction (both handcrafted and deep learning features), feature selection, and classification. In addition, future research directions in the area of automatic disease identification and categorization employing endoscopic frames are being looked into based on the existing literature.

In human activity-recognition scenarios, including head and entire body pose and orientations, recognizing the pose and direction of a pedestrian is considered a complex problem. A person may be traveling in one sideway while focusing his attention on another side. It is occasionally desirable to analyze such orientation estimates using computer-vision tools for automated analysis of pedestrian behavior and intention. This article uses a deep-learning method to demonstrate the pedestrian full-body pose estimation approach. A deep-learning-based pre-trained supervised model multi-branched deep learning pose net (MBDLP-Net) is proposed for estimation and classification. For full-body pose and orientation estimation, three independent datasets, an extensive dataset for body orientation (BDBO), PKU-Reid, and TUD Multiview Pedestrians, are used. Independently, the proposed technique is trained on dataset CIFAR-100 with 100 classes. The proposed approach is meticulously tested using publicly accessible BDBO, PKU-Reid, and TUD datasets. The results show that the mean accuracy for full-body pose estimation with BDBO and PKU-Reid is 0.95%, and with TUD multiview pedestrians is 0.97%. The performance results show that the proposed technique efficiently distinguishes full-body poses and orientations in various configurations. The efficacy of the provided approach is compared with existing pretrained, robust, and state-of-the-art methodologies, providing a comprehensive understanding of its advantages.

Corn diseases significantly impact crop yields, posing a major challenge to agricultural productivity. Early and accurate detection of these diseases is crucial for effective management and mitigation. Existing methods, mostly relying on analyzing corn leaves, often lack the precision to identify and classify a wide range of diseases under varying conditions. This study introduces a novel approach to detecting corn diseases using image processing and deep learning techniques, aiming to enhance detection accuracy through pre-processing, improved feature extraction and selection, and classification algorithms. A new deep Convolutional Neural Network (CNN) model named TreeNet, with 35 layers and 38 connections, is proposed. TreeNet is pre-trained using the Plant Village imaging dataset. For image pre-processing, the YCbCr color space is utilized to improve color representation and contrast. Feature extraction is performed using TreeNet and two pre-trained models, Darknet-53, and DenseNet-201, with features fused using a serial-based fusion method. The Entropy-coded Sine Cosine Algorithm is applied for feature selection, optimizing the feature set for classification. The selected features are used to train Support Vector Machine (SVM) and K-Nearest Neighbor (KNN) classifiers, with extensive experiments conducted using both 5-fold and 10-fold cross-validation, and feature sizes ranging from 200 to 1150. The proposed method achieves classification accuracy, precision, recall, and F1-score of 99.8%, 99%, 100%, and 99%, respectively, surpassing existing benchmarks. The integration of TreeNet with Darknet-53 and DenseNet-201, along with robust pre-processing and feature selection, significantly improves corn disease detection, highlighting the potential of advanced CNN architectures in agriculture.

The lungs are a very important organ in a human. Any abnormality in the lungs ultimately affects the whole body. Pulmonary nodules-initiated lung cancer that is very small in size initially. Lung cancer spreads in the whole body if it's not detected/treated at the start; therefore, detection of lung cancer at the earliest is mandatory for the survival of patients. In medical imaging, machine/deep learning methods are very useful for accurate and efficient detection of diseases. Therefore, two models are proposed for the segmentation and classification based on transformer/deep/machine learning. The segmentation model named TransSegNet is proposed in which input images are converted into [44] patches and embedded into four transformer blocks. The model is trained from scratch by using 10 epochs, 4 batch-size, and an Adam optimizer as selected hyperparameters. The model is evaluated on the ACDC-LungHP dataset that provides 99.62 % detection accuracy having 0.0067 loss, Mean IOU of 49.81, validation accuracy of 99.02 %, validation-loss of 0.0219, and validation mean IOU of 49.52. The minimum and maximum improvement in the results of the proposed segmentation model is +04.62 and +11.62 respectively as compared to existing methods. The second classification model named MinClassNet is proposed which contains seven layers including input, conv, ReLU, batch normalization, softmax, fully connected, and output. The model is also trained from scratch based on selected hyperparameters after extensive experimentation such as Adam optimizer, 8 batch size, and 10 epochs. The model performance is tested using a publically available histopathological lung cancer dataset which provided 98.39 % accuracy in classifying the healthy/abnormal cells. The minimum and maximum improvement in the results of the proposed classification model are +1.29 and +20.06 respectively as compared to existing methods.

Accurate detection and classification of artifacts within the gastrointestinal (GI) tract frames remain a significant challenge in medical image processing. Medical science combined with artificial intelligence is advancing to automate the diagnosis and treatment of numerous diseases. Key to this is the development of robust algorithms for image classification and detection, crucial in designing sophisticated systems for diagnosis and treatment. This study makes a small contribution to endoscopic image classification. The proposed approach involves multiple operations, including extracting deep features from endoscopy images using pre -trained neural networks such as Darknet-53 and Xception. Additionally, feature optimization utilizes the binary dragonfly algorithm (BDA), with the fusion of the obtained feature vectors. The fused feature set is input into the ensemble subspace k nearest neighbors (ESKNN) classifier. The Kvasir-V2 benchmark dataset, and the COMSATS University Islamabad (CUI) Wah private dataset, featuring three classes of endoscopic stomach images were used. Performance assessments considered various feature selection techniques, including genetic algorithm (GA), particle swarm optimization (PSO), salp swarm algorithm (SSA), sine cosine algorithm (SCA), and grey wolf optimizer (GWO). The proposed model excels, achieving an overall classification accuracy of 98.25% on the Kvasir-V2 benchmark and 99.90% on the CUI Wah private dataset. This approach holds promise for developing an automated computer -aided system for classifying GI tract syndromes through endoscopy images.

[This corrects the article DOI: 10.1371/journal.pone.0292601.].

Colorectal cancer has become a significant global health challenge and causes millions of fatalities every year. A key factor contributing to its high fatality rate is the absence of timely screening and accurate diagnosis. Early diagnosis is based on the detection and analysis of polyps which are cancer precursors linked to aging and declining health. Artificial intelligence-based computer-aided polyp detection (CAPD) tools are vital for affordable and efficient screening while mitigating the global impact of colorectal cancer. Aimed at improving the polyp detection capabilities, this research presents a real-time framework for detecting polyps in endoscopy frames, employing contrast-limited adaptive histogram equalization (CLAHE) and the proposed polyp detector YOLOv8p. The proposed detector uses multiple convolutional, C2f (Faster Cross Stage Partial Bottleneck with 2 convolutions), and SPPF (Spatial Pyramid Pooling-Fast) blocks. Moreover, a generic suitability assessment metric named YOLO-Score is proposed for overall performance evaluation. The proposed framework demonstrates superior clinical relevance and comparable performance to existing SOTA methods by achieving a mAP50 score of 95.5% and an $F_{1}$ score of 92.6% on the Hyper-Kvasir-Seg dataset. The research underscores the essential role of CAPD tools in improving diagnostic precision and streamlining real-time clinical processes in colorectal cancer screening.

In this manuscript, imbalanced and small sample space (IB-SSS) dataset problems for pedestrian gender classification using fusion of selected deep and traditional features (PGC-FSDTF) are considered. In this regard, data preparation is first done through data augmentation and preprocessing steps to handle imbalanced classification problem and environmental effects, respectively. The proposed approach follows different types of feature extraction schemes, for instance, pyramid histogram of oriented gradients, hue saturation value histogram, deep visual features of DenseNet201 and InceptionResNetV2-based convolutional neural network architectures. The parallel fusion method computes the maximum and average values-based features from the learned features of both deep networks. Features are selected through features selection methods such as entropy and principal component analysis (PCA). The subsets of features are serially fused and provided to multiple classifiers to perform gender classification on IB-SSS datasets. Resultantly, the proposed PGC-FSDTF method shows better results in terms of different accuracies (overall, mean, and balanced), and area under curve on selected datasets. Further, improved results are achieved on applied datasets using PCA-based selected features and medium Gaussian support vector machine (M-SVM) classifier. These results on different datasets confirm that the selected feature combination provides a way to handle IB-SSS issues for PGC effectively.

Computer-aided classification of diseases of the gastrointestinal tract (GIT) has become a crucial area of research. Medical science and artificial intelligence have helped medical experts find GIT diseases through endoscopic procedures. Wired endoscopy is a controlled procedure that helps the medical expert in disease diagnosis. Manual screening of the endoscopic frames is a challenging and time taking task for medical experts that also increases the missed rate of the GIT disease. An early diagnosis of GIT disease can save human beings from fatal diseases. An automatic deep feature learning-based system is proposed for GIT disease classification. The adaptive gamma correction and weighting distribution (AGCWD) preprocessing procedure is the first stage of the proposed work that is used for enhancing the intensity of the frames. The deep features are extracted from the frames by deep learning models including InceptionNetV3 and GITNet. Ant Colony Optimization (ACO) procedure is employed for feature optimization. Optimized features are fused serially. The classification operation is performed by variants of support vector machine (SVM) classifiers, including the Cubic SVM (CSVM), Coarse Gaussian SVM (CGSVM), Quadratic SVM (QSVM), and Linear SVM (LSVM) classifiers. The intended model is assessed on two challenging datasets including KVASIR and NERTHUS that consist of eight and four classes respectively. The intended model outperforms as compared with existing methods by achieving an accuracy of 99.32% over the KVASIR dataset and 99.89% accuracy using the NERTHUS dataset.

Fruit is an essential element of human life and a significant gain for the agriculture sector. Guava is a common fruit found in different countries. It is considered the fourth primary fruit in Pakistan. Several bacterial and fungal diseases found in guava fruit decrease production daily. Leaf Blight is a common disease found in guava fruit that affects the growth and production of fruit. Automatic detection of leaf blight disease in guava fruit can help avoid decreases in its production. In this research, we proposed a CNN-based deep model named SidNet. The proposed model contains thirty-three layers. We used a guava dataset for early recognition of leaf blight, which consists of two classes. Initially, the YCbCr color space was employed as a preprocessing step in detecting leaf blight. As the original dataset was small, data augmentation was performed. DarkNet-53, AlexNet, and the proposed SidNet were used for feature acquisition. The features were fused to get the best-desired results. Binary Gray Wolf Optimization (BGWO) was used on the fused features for feature selection. The optimized features were given to the variants of SVM and KNN classifiers for classification. The experiments were performed on 5- and 10-fold cross validation. The highest achievable outcomes were 98.9% with 5-fold and 99.2% with 10-fold cross validation, confirming the evidence that the identification of Leaf Blight is accurate, successful, and efficient.

Breast cancer (BC) is a widely diagnosed deadly disease commonly present in middle-aged women around the globe. Ultrasound (U/S) imaging is widely used for the early prediction and segmentation of BC due to low radiation and cheapness. Manual BC segmentation from ultrasound imaging is a complex and laborious task due to inherited noise. Many deep learning-based breast cancer diagnostic methods are presented that can further be enhanced to improve the segmentation performance. This work proposed a U-shaped auto encoder-based multi-attention triple decoder convolution neural network for BC segmentation from U/S images. To capture multi-scale diverse spatial image features this work introduced a multi-scale convolution operation-based encoder network. To process the multi-scale learned diverse spatial features in the encoder path multi-scale triple decoder network is designed that was not found in earlier studies. To highlight the tumor region at different scales multi-attention mechanism is introduced in each decoder network. The multi-attention mechanism is designed to suppress the other region information and to highlight the tumor region features at different scales. The proposed deep network produced the segmentation dice of 90.45% on the UDIAT dataset and the segmentation dice of 89.13% on the BUSI dataset. The testing Jaccard index of 83.40% is recorded on the UDIAT dataset and a Jaccard index of 82.31% is recorded on the BUSI dataset. The result comparison with existing methods shows that our method achieved the highest results. The segmentation performance of the triple decoder-based BC segmentation model suggested that it can effectively be used to automate the manual breast cancer segmentation task from ultrasound images.

Video surveillance and activity monitoring are the practical real-time applications of Human Action Recognition (HAR). A fusion of several Convolutional Neural Network (CNN) architectures has been widely used for effective HAR and achieved impressive results. Feature fusion of multiple pre-trained models also extracts redundant features due to the combinations of identical layers in all CNN architectures. In this study, network-level fusion is proposed, which reduces the possibility of having identical layers throughout the fusion process and helps extract unique features. Three pre-trained models, i.e., NASNetLarge, DenseNet201, and DarkNet53 are selected and analyzed to select the most efficient combinations of layers among these networks. Selected combinations of these networks are fused using five proposed strategies, i.e., sum, max, concatenation, convolutional and bilinear fusion. In the end, a proposed minimized CNN architecture is utilized to extract descriptors, which are optimized using the proposed Elastic Net-based Genetic Algorithm (ENGA) approach. A two-phase hybrid ENGA technique is suggested to pick features using both GA and EN. GA is used in the initial stage to reduce the dimensionality of retrieved features. To eliminate the unnecessary features, EN regularization is put into place in the second phase. The proposed ENGA model is evaluated on four publicly available datasets including UTKinect-Action, MSR-Action3D dataset, Florence3D-Action dataset, and Youtube-8 m, and achieved 99.63%, 99.69%. 98.63% and 91.46% accuracies, respectively.

Anomalous situations in surveillance videos or images that may result in security issues, such as disasters, accidents, crime, violence, or terror-ism, can be identified through video anomaly detection. However, differentiat-ing anomalous situations from normal can be challenging due to variations in human activity in complex environments such as train stations, busy sporting fields, airports, shopping areas, military bases, care centers, etc. Deep learn-ing models' learning capability is leveraged to identify abnormal situations with improved accuracy. This work proposes a deep learning architecture called Anomalous Situation Recognition Network (ASRNet) for deep feature extraction to improve the detection accuracy of various anomalous image situations. The proposed framework has five steps. In the first step, pretraining of the proposed architecture is performed on the CIFAR-100 dataset. In the second step, the proposed pre-trained model and Inception V3 architecture are used for feature extraction by utilizing the suspicious activity recognition dataset. In the third step, serial feature fusion is performed, and then the Dragonfly algorithm is utilized for feature optimization in the fourth step. Finally, using optimized features, various Support Vector Machine (SVM) and K-Nearest Neighbor (KNN) based classification models are utilized to detect anomalous situations. The proposed framework is validated on the suspicious activity dataset by varying the number of optimized features from 100 to 1000. The results show that the proposed method is effective in detecting anomalous situations and achieves the highest accuracy of 99.24% using cubic SVM.