Search

COVID‐19 has been ravaging the world for a long time, and although its effects are currently the same as those of a cold or a fever, timely diagnosis of COVID‐19 in the elderly and in patients with related illnesses is still a matter of great urgency. To address this challenge, we propose a model that combines the strengths of the Swin Transformer and ResNet34 architectures to efficiently diagnose COVID‐19 in elderly and vulnerable patients. In this paper, we design a model that integrates Swin transformer and resnet34, which not only integrates the advantages of transformer and CNN but also achieves excellent performance in this image classification problem. Moreover, a pre‐processing method is also proposed to increase the accuracy of the model to 99.08%. In this paper, experiments were conducted on Kaggle's publicly available three‐classification and four‐classification datasets, respectively, and on the three main evaluation metrics of Accuracy, Precision, and Recall, the first dataset obtained 98.81%, 99.49%, and 97.99%, while the second dataset obtained 88.82%, 88.92%, and 86.38%. These findings highlight the validity and potential of our proposed model for diagnosing the presence or absence of COVID‐19 in elderly and vulnerable patients. [ABSTRACT FROM AUTHOR]

Colorectal cancer is a common gastrointestinal malignancy. Early screening and segmentation of colorectal polyps are of great clinical significance. Colonoscopy is the most effective method to detect polyps, but some polyps may be missed in the detection process. On this basis, the use of computer‐aided diagnosis technology is particularly important for colorectal polyp segmentation. To improve the detection rate of intestinal polyps under colonoscopy, a polyp segmentation network (MobileRaNet) based on a lightweight model and reverse attention (RA) mechanism was proposed to accurately segment polyps in colonoscopy images. The coordinated attention module is used to improve MobileNetV3 and make it the backbone network (CaNet). Second, a part of the output of the high‐level feature from the backbone network is passed into the parallel axial receptive field module (PA_RFB) to extract the global dependency representation without losing the details. Third, a global map is generated based on this combined feature as the initial boot area of the subsequent components. Finally, the RA module is used to mine the target region and boundary clues to improve the segmentation accuracy. To verify the effectiveness and lightweight performance of the algorithm, five challenging datasets, including CVC‐ColonDB, CVC‐300, and Kvasir, are used in this paper. In six indexes, including MeanDice, MeanIoU, and MAE, compared with seven typical models such as PraNet and TransUnet, accuracy, FLOPs, parameters, and FPS were compared. The experimental results show that the MobileRaNet proposed in this paper has improved the performance of the five datasets to varying degrees, especially the MeanDice and MeanIOU indexes of the Kvasir dataset reach 91.2% and 85.6%, which are, respectively, increased by 1.4% and 1.6% compared with PraNet. Compared with PraNet, FLOPs and parameters decreased by 83.3% and 76.7%, respectively. [ABSTRACT FROM AUTHOR]

Correct identification of tumor in brain images is critical for treatment. In the medical domain, class distributions of recorded data could differ with locations and require high levels of privacy while collaboratively training the deep learning (DL) models for classifications. The main aim of this paper is to propose a privacy‐preserving collaborative model for the classification of brain tumor in heterogeneously distributed magnetic resonance imaging (MRI) images. In this paper, initially, an open‐source dataset has been acquired and analyzed as per the required competencies. The acquired dataset has four types of MRI images: pituitary tumor, meningioma tumor, glioma tumor, and no tumor. First, the acquired dataset was analyzed using DL and transfer learning algorithms. By applying implementations of basic algorithms, better algorithms were identified for further implementations in a federated learning ecosystem. DenseNet201‐based transfer learning was identified as a better neural network and further utilized for collaborative transfer learning implementations. Here, the paper also focused on developing a suitable system for a heterogeneous distributed tumor database. Heterogeneous data were converted from the available data by applying nonidentical data distribution. The study discovered that the federated DL models, involving multiple clients, exhibited superior performance compared to conventional pretrained models. The proposed framework possesses distinctive characteristics that distinguish it from existing classification methods for brain tumor identification, particularly in terms of ensuring data privacy for edge devices with limited resources. Due to these additional features, the framework stands as the optimal alternative solution for early diagnosis of brain tumor. [ABSTRACT FROM AUTHOR]

Image registration is a fundamental task in medical image analysis, and large deformation registration poses significant challenges due to substantial appearance differences between the fixed image and the moving image. With the advancement of artificial intelligence, reinforcement learning has been gradually incorporated into image registration tasks. Notably, Ziwei Luo, Jing Hu, and other researchers introduced the first reinforcement learning‐based large deformation registration model called Stochastic Planner‐Actor‐Critic (SPAC), achieving superior performance in large deformation registration compared with advanced deep learning registration methods, which is the only large deformation registration reinforcement learning model. However, the existing reinforcement learning model for large deformation registration do not consider the impact of Q‐value prediction accuracy on the overall task results. This study aims to investigate brain large deformation registration under the influence of Alzheimer's disease. To further improve the accuracy of brain large deformation registration, this paper combines the proven advanced Q‐value distribution prediction in reinforcement learning with the large deformation registration model SPAC, designing a distributional SPAC model (DSPAC). Subsequent experiments on the ADNI dataset demonstrate the effectiveness of this model. [ABSTRACT FROM AUTHOR]

Foot pain, particularly caused by heel spurs and Sever's disease, significantly impacts mobility and daily activities for many people. These diseases are traditionally diagnosed by orthopedic specialists using X‐ray images of the lateral foot. In certain situations, the absence of specialists requires the adoption of AI‐based methods; however, the lack of a dataset hinders the use of AI for the preliminary diagnosis of these diseases. Therefore, this study first presents a novel dataset consisting of 3956 annotated lateral foot X‐ray images and uses the original capsule network (CapsNet) to automatically detect and classify heel bone diseases. The low accuracy of 73.99% of CapsNet due to the low extraction feature layers led us to search for a new model. For this reason, this paper also proposes a new enhanced capsule network (HeCapsNet) by adjusting the features extraction layers, adding extra convolutional layers, using "he normal" kernel initializer instead of "normal" and utilizing the "same" padding scheme to perform better with medical images. Evaluating the performance of the proposed model, higher accuracy rates are achieved, including 97.29% for balanced data, 94.19% for imbalanced data, area under the curve (AUC) of 98.69%, and a fivefold cross‐validation accuracy of 95.77%. We then compared our proposed model with state‐of‐the‐art modified CapsNet models using various datasets (MNIST, Fashion‐MNIST, CIFAR10, and brain tumor). HeCapsNet performed similarly to modified CapsNets on relatively simple non‐medical datasets such as MNIST and Fashion‐MNIST, but performed better on more complex medical datasets. [ABSTRACT FROM AUTHOR]

Non‐cancerous growths called uterine fibroids develop in the uterus. They can vary in size, location, and number, and can produce symptoms including excessive menstrual flow, pelvic discomfort, and reproductive problems. Early detection of uterine fibroids is important because it allows for timely intervention and appropriate management strategies. Extracting meaningful features from ultrasound (US) images requires robust and effective techniques. However, different feature extraction methods may yield varying results, and the choice of technique can influence the accuracy of fibroid detection. This paper presents an efficient approach for early detection of uterine fibroids in US images. Our proposed technique combines efficient feature extraction methods and a hybrid deep learning approach. Firstly, we employ the well‐known canny edge detection algorithm to accurately identify the edges of the fibroid region in the US images. This step helps in segmenting the target edge portion for further analysis. Additionally, we introduce an improved bird swarm optimization (IBSO) algorithm to extract a comprehensive set of features, utilizing both known features and newly obtained features. This approach enhances the accuracy of fibroid detection in uterine images. The proposed method uses a convolutional recurrent neural network (CRNN) to accurately detect and treat uterine fibroids, ensuring timely diagnosis and treatment. Our proposed IBSO‐CRNN model produces accuracy rates of 99.897% and 97.568% on the augmented and original datasets, respectively. [ABSTRACT FROM AUTHOR]

Accurate segmentation of brain tumors has a vital impact on clinical diagnosis and treatment, and good segmentation results are helpful for the treatment of this disease, which is a serious threat to human health. High‐precision segmentation of brain tumors remains a challenging task due to their diverse shapes, sizes, locations, and complex boundaries. Considering the special structure of medical brain tumor images, many researchers have proposed a brain tumor segmentation (BraTS) network based on 3D U‐Net. However, there are also problems such as insufficient receptive fields and excessive computing costs. In this paper, we propose an efficient BraTS model based on group normalization (GN) and 3D U‐Net (3D‐EffUNet). First, according to the characteristics of brain tumor images, the medical image of the whole case is input into the model, and 3D convolution layers are used to extract features and filter irrelevant information. Then, using 3D U‐Net as the main framework, an efficient convolutional module is designed for more precise processing of brain tumor features. Moreover, an efficient convolution module based on GN and an attention mechanism is introduced to reduce the complexity of the network without affecting the segmentation performance and to increase the awareness of voxels between adjacent dimensions and the local space. Finally, the decoder was used to reconstruct high‐precision BraTS information. The model is trained and tested on the BraTS2021 dataset, and the experimental results show that it can maintain good segmentation performance and greatly reduce the calculation cost. [ABSTRACT FROM AUTHOR]

The intravenous (IV) injection procedure can be a challenging task, especially for individuals with thin veins, obesity, or patients with damaged and pigmented skin. Therefore, the IV procedure necessitates a portable medical device that can be used for academic demonstrations to train medical students or by health care professionals to perform venipuncture. Vein visualization with a vein detector is principally based on the interaction of blood components with wavelengths of the electromagnetic spectrum (EMS). In this paper, we first present the process of image formation in the spectral band of the near‐infrared region (NIR) of EMS. Then, we introduce the image acquisition system with image processing to extract the veins as a noninvasive vein detection method. A Raspberry Pi (Model 4B), along with a night vision camera, serves as an image acquisition tool to capture skin area illuminated by NIR. Following this, the data is transferred to the laptop where it can be filtered and processed using Python image processing tools before being viewed on the monitor. The results achieved through the device are quite encouraging, as the image recognition between veins and adjacent tissues from the skin sample can be clearly marked. The functionality, accuracy, and simplicity associated with this vein detection system make it a potential device for IV placement and the morphological study of disease detection. [ABSTRACT FROM AUTHOR]

Sleep affects the functioning of all biological processes in the human body. Sleep disorders disrupt the well‐being of an individual and can potentially lead to other health complications. It affects a large part of the population, and their timely and efficient detection is crucial for improving the quality of health. Sleep disorders are generally analyzed using polysomnogram (PSG) signals, which must be captured in sophisticated laboratory settings. However, electrocardiogram (ECG) signals provide information about the electrical activity of the heart, which is strongly linked with disturbances in sleep and can be used as an indicator of disordered sleep. Along with that, the recording of ECG signals is much easier than other signals and can be done remotely without causing any significant disturbance to the subject. This brings practicality to the systems that are already being employed for sleep disorder detection. In this paper, we present a method for identifying types of sleep disorders using ECG signals. The objective of this paper is to investigate the use of ECG signals for the automated identification of insomnia, narcolepsy, periodic leg movement (PLM), rapid eye movement (REM) behavior disorder (RBD), and nocturnal frontal lobe epilepsy (NFLE) against healthy subjects. We aim to develop a machine learning‐based algorithm that automatically classifies ECG signals into these sleep disorder categories. The cyclic alternating pattern (CAP) sleep database was used in this study, which contains PSG recordings from individuals with and without sleep disorders, including insomnia, narcolepsy, PLM, RBD, and NFLE. A wavelet scattering network has been used to extract features from the ECG signals. Various classifiers were tested, and the ensemble bag of trees classifier provided the optimum performance. An overall accuracy of 98% was obtained for the identification of sleep disorders, along with a classification accuracy of 99.37%, 99.45%, 99.23%, 99.4%, and 99.65% for insomnia, narcolepsy, NFLE, PLM, and RBD, respectively, for binary classification against healthy subject data. Our results show that wavelet scattering network and ensemble of bagged tree (EbagT) classification can accurately identify various sleep disorders. They provided evidence that ECG signals can be used to identify and diagnose sleep disorders, which could lead to the development of accurate detection of these sleep disorders. Also, our study demonstrated the effectiveness of the proposed algorithm in identifying multiple sleep disorders, which could lead to more efficient and cost‐effective diagnosis and treatment of sleep disorders. [ABSTRACT FROM AUTHOR]

Parkinson's Disease (PD) is a brain‐related disease that eventually causes disability and disrupts a person's normal life. Most physicians and researchers try to diagnose it quickly and treat it on time. In recent years, computer science and the field of artificial intelligence (machine learning) have helped researchers find a way to detect the disease early. This article proposes a method that diagnoses Parkinson's disease by analyzing the hand drawing shaped by the individuals using Bi‐Directional Long Short‐Term Memory (Bi‐LSTM) neural network. In addition, this paper proposes a Fuzzy Inferential Classifier for the Dense layer, which classifies the output of LSTM Blocks to the associated classes by modifying the Soft‐max function. Our decision to propose this classifier was because, in some cases, hand‐drawing data related to people with Parkinson's disease have no significant difference with the healthy subjects for the distinguishing and are often very similar to each other. A standard dataset has been used in this paper, which includes spirals drawn test (including static spiral, dynamic spiral, and stability specific point) by a group of healthy people and people with Parkinson's disease. The proposed method has reached 97%, 98.5%, and 100% accuracy rates for the three mentioned spiral tests with a smoother training loss and accuracy plots. In addition, the results outperform state‐of‐the‐art research conducted on this dataset and show at least more than 2.5% improvement in the accuracy rate in comparison. [ABSTRACT FROM AUTHOR]

This paper presents a two‐stage deep learning framework, RR‐HCL‐SVM, designed to aid in the assessment of residual thyroid tissues following thyroidectomy, utilizing single‐photon emission computed tomography (SPECT) images. Leveraging the power of deep learning, our model offers a comprehensive solution for the detection and assessment of remaining thyroid tissues. To enhance accuracy, we introduce a unique combination of features, incorporating the Radio Scan Index (RSI) and radiomics features. These features not only improve accuracy but also provide valuable insights into tissue characterization. Moreover, we employ efficient clustering techniques for feature dimension reduction, preserving model performance while reducing computational complexity. Experimental results demonstrate the effectiveness of our approach, achieving an impressive F1 score of 0.97, sensitivity of 0.96, and specificity of 0.98. The RR‐HCL SVM framework holds great promise in the clinical setting for the precise evaluation of residual thyroid tissues post‐thyroidectomy, offering potential benefits for patient care and treatment planning. [ABSTRACT FROM AUTHOR]

In recent years, deep learning techniques have been increasingly utilized to automate and reduce human errors in leukocyte classification. However, reliability and accuracy issues remain a challenge. This paper proposes CAB‐Net, an end‐to‐end deep learning‐based model that integrates a convolutional neural network with an attention mechanism and BLSTM. The model uses a stack of convolutional and pooling layers to extract initial features from input images, which are then fed into the attention block to generate class‐specific discriminative features and BLSTM employed to model deterministic class correlations between highlighted discriminative features and labels. Experimental results show that CAB‐Net outperforms state‐of‐the‐art methods in term of accuracy, precision, recall, F1‐Score, and AUC‐ROC. In addition, the CAB‐Net model demonstrates good performance even with low‐resolution images. Consequently, CAB‐Net can be considered as the current state‐of‐the‐art prediction model for improving leukocyte classification in clinical environments. [ABSTRACT FROM AUTHOR]

Skin cancer is a common type of malignant tumor that poses a serious threat to patients' lives and health, especially melanoma. It may spread to other body parts, resulting in serious complications and death. In the medical field, accurate identification of skin lesion images is crucial for diagnosing different diseases. However, due to the similarity between different skin lesions, it brings some challenges to medical diagnosis. In this paper, a novel Ensemble Learning Model (EL‐DLOA) based on deep learning and optimization algorithms is proposed, which uses four different deep neural network architectures to generate confidence levels for classes, and optimization algorithms are used to integrate these confidence levels to make the final predictions. To ensure the model's accuracy and reliability, it is first trained using three different learning rates to find the best classification performance of the model. Then, a new search method based on the grey wolf optimization algorithm is proposed to enhance the grey wolf search efficiency. The method improves the search mechanism by changing the grey wolf's individual position through random perturbation or adaptive mutation, which solves the problem that the grey wolf algorithm is easy to fall into local optimum. Finally, four different ensemble strategies are used to reduce individual model bias in the classification process. The proposed model is trained and evaluated using the publicly available dataset HAM10000. The experimental results show that the improved grey wolf optimization algorithm effectively avoids the premature convergence problem and improves the search combination efficiency. Furthermore, in the ensemble methods, the adaptive weight average ensemble strategy effectively improves the classification performance, yielding accuracy, precision, recall, and F1 scores of 0.888, 0.837, 0.897, and 0.862, respectively. These metrics show varying degrees of improvement over the best performing single model. In general, the results indicate that the proposed method achieves high accuracy and practicality in skin lesion classification. Our model shows excellent performance in comparison with other existing models, which makes it significant for research and application in dermatology diagnosis. [ABSTRACT FROM AUTHOR]

Compressed sensing (CS) is a novel technique for MRI acceleration. The purpose of this paper was to assess the effects of CS on the radiomic features extracted from amide proton transfer‐weighted (APTw) images. Brain tumor MRI data of 40 scans were studied. Standard images using sensitivity encoding (SENSE) with an acceleration factor (AF) of 2 were used as the gold standard, and APTw images using SENSE with CS (CS‐SENSE) with an AF of 4 were assessed. Regions of interest (ROIs), including normal tissue, edema, liquefactive necrosis, and tumor, were manually drawn, and the effects of CS‐SENSE on radiomics were assessed for each ROI category. An intraclass correlation coefficient (ICC) was first calculated for each feature extracted from APTw images with SENSE and CS‐SENSE for all ROIs. Different filters were applied to the original images, and the effects of these filters on the ICCs were further compared between APTw images with SENSE and CS‐SENSE. Feature deviations were also provided for a more comprehensive evaluation of the effects of CS‐SENSE on radiomic features. The ROI‐based comparison showed that most radiomic features extracted from CS‐SENSE‐APTw images and SENSE‐APTw images had moderate or greater reliabilities (ICC ≥ 0.5) for all four ROIs and all eight image sets with different filters. Tumor showed significantly higher ICCs than normal tissue, edema, and liquefactive necrosis. Compared to the original images, filters (such as Exponential or Square) may improve the reliability of radiomic features extracted from CS‐SENSE‐APTw and SENSE‐APTw images. [ABSTRACT FROM AUTHOR]

The assessment of glomerular lesions is a fundamental step toward the diagnosis of glomerular diseases. This requires diagnosis and fusion of information from all the glomeruli at multiple levels and stainings. The lack of research on multi‐level multistained glomerular identification and matching has resulted in renal pathologists devoting much time and attention to this time‐consuming and labor‐intensive process. This limits the overall efficiency of the diagnosis of glomerular diseases. This paper constructed a dataset consisting of 600 cases, each containing 12 levels of whole slide images from H&E, PAS, Masson trichrome, and PASM staining. The glomeruli identifying and matching was proposed. First, a multistained transformer‐based Mask R‐CNN is proposed to extract the position and contours of the glomeruli. Second, coherent point drift‐based coarse matching and hybrid feature‐based fine matching achieve pairwise matching. Finally, the voting‐based cross‐matching realizes 12‐level multistained matching. This system constitutes a practical human‐computer interface. Intensive experiments were conducted to validate the ability to identify and match 12‐level multistained glomeruli. The mAP@50 for detection and segmentation reached 95.40% and 95.70%, respectively. The basic and comprehensive matching rates of the glomeruli matching reached 98.25% and 74.59%, respectively. Visualization results further demonstrate that the model achieved accurate identification and matching. The proposed system achieves accurate identification and matching of 12 levels of multistained glomeruli and can serve as a tool for pathological diagnosis of glomerular diseases by simplifying the diagnostic process. More importantly, this system can lay the foundation for the fully automated assisted diagnosis of glomerular diseases. [ABSTRACT FROM AUTHOR]

Abnormal changes in the retinal vessel width usually indicate the presence of certain diseases. This paper presents a novel model to measure the width of retinal vessels. This method firstly employs cubic spline fitting to determine the vessel cross‐sectional direction at the center point of a vessel profile, then uses the extended edge‐searching mode to detect vessel edge points along the vessel cross‐sectional direction, and finally obtains the width value by calculating the distance between the edge point and the center point. The method can accurately locate vessel edge points and successfully solve the problem of central light reflex in blood vessels. Experimental analyses on a publicly available database show that the proposed approach has excellent performance in accuracy with other existing measurement techniques. It provides an effective idea for analyzing morphological changes in blood vessels and can be extended to obtain the shape features of other objects. [ABSTRACT FROM AUTHOR]

One of the most difficult problems that develop when brain cells start to grow out of control is a brain tumor, which is regarded as the most lethal disease of the century. Finding and identifying malignant brain magnetic resonance imaging (MRI) images is the major challenge before therapy. Researchers have been putting a lot of effort into creating the best method for more accurate real‐world medical image recognition. For manual categorization, it is quite time‐consuming to segment large quantities of MRI data. To mitigate these issues, this paper suggests the information exchange gateway‐based residual UNet (IEGResUNet) model, which uses the ResUNet model as a base model. Additionally, including principal component analysis (PCA) data augmentation will increase the model's efficiency while also enhancing its speed. The IEGResUNet model shows an ablation investigation on three Brats datasets, with and without PCA augmentation. The results demonstrate that IEGResUNet will improve segmentation effectiveness and can also manage the imbalance in input data when PCA data augmentation models are included. The dice score on BraTS 2019 for whole tumor, region of core tumor, and region of enhancing tumor were 0.9083, 0.883, and 0.8106 respectively. Also, on BraTS 2020, the dice score for WT, CT, and ET 0.9083, 0.883, and 0.8106 was respectively. Similarly, on BraTS 2021, the dice score for WT, CT, and ET was 0.8737, 0.8866, and 0.7963 respectively. Comparing against baseline models, the IEGResUNet scored well in terms of dice score and intersection over union. [ABSTRACT FROM AUTHOR]

Medical diagnostics rely heavily on ultrasound imaging since it is non‐invasive, inexpensive, and able to provide images in real time. However, the inherent existence of a kind of signal‐dependent noise in these images diminishes the use of ultrasound imaging systems. Speckle noise is inherently present in ultrasound images. Its inherent presence occurs during the image acquisition phase. It decreases the diagnostic usefulness of this imaging modality since it interferes with and reduces the image resolution and contrast. Speckle noise is multiplicative in nature. This multiplicative nature of speckle noise causes adverse effects on ultrasound imaging because it distorts the image quality and causes a loss of the patient's informative content in ultrasound imaging. This causes difficulty in tissue characterization. A lot of research has been done in this field to remove speckle noise while preserving medical information in the image. To handle this issue, research has been grouped into multiple domains. The two main domains include homomorphic and non‐homomorphic filtering. The homomorphic filtering uses a logarithmic transform that converts this multiplicative nature into an additive nature and uses any additive image restoration model to do a despeckling task. Non‐homomorphic filtering encompasses all methods that do not employ a logarithmic transform. Since the presence of speckle noise in ultrasound imaging is an inherent property. Therefore, the despeckling of medical ultrasound images is a mandatory task that cannot be avoided or ignored. This paper proposes a two‐step hybrid and homomorphic despeckling technique, where modified total variation method is applied as step one for speckle reduction purposes. Step two implements method of noise thresholding in the non‐subsampled contourlet transform (NSCT) domain using the bivariate shrinkage rule for edge preservation. The effectiveness and reliability of the proposed method are tested by comparing it with some of the latest methods based on qualitative and quantitative analyses. This paper will help the new researchers understand various problems and their solutions related to ultrasound image despeckling. [ABSTRACT FROM AUTHOR]

Fetal abdominal subcutaneous soft tissue thickness (FASSTT) is a key indicator in evaluating fetal growth, development, and nutritional status. Currently, manual measurement in FASSTT faces the problems of difficulty in positioning, time consumption, and inaccurate measurement. Therefore, this article proposes an automatic measurement scheme for FASSTT. Firstly, a U‐shaped attention network VGG‐SeUnet is proposed to automatically segment the subcutaneous soft tissue area of the fetal abdomen. Secondly, based on the segmentation results, a morphological method is proposed to obtain FASSTT. Specifically, the segmentation network uses VGG16 as the encoder and connects the decoder through jump connections for multi‐scale fusion. We introduce channel attention to jump connections, which enables the model to select key channels for feature fusion. At the same time, the model uses the proposed DF_Loss function for training to solve the problem of sample imbalance. Based on the segmentation results, a morphological distance transformation algorithm is proposed to obtain FASSTT by drawing the maximum inscribed circle and calculating its diameter. The method is evaluated on a fetal abdominal circumference ultrasound dataset containing 135 samples. The DSC, mIOU, mPA, Precision and Recall of the segmentation experiments reache 88.9%, 81.02%, 90.44%, 86.38%, and 90.44% respectively. The difference between FASSTT's result from that provided by radiologist is 0.0615 cm in Average Error (AVE) and 0.081 cm in Root Mean Square Error (RMSE). The overall performance of this algorithm is superior to existing methods with excellent performance, and the measurement error is less than 1 millimeter. This result proves that the proposed scheme in this paper can achieve automated measurement of FASSTT and assist doctors in subsequent evaluation of fetal development and nutritional status. [ABSTRACT FROM AUTHOR]

With rapid developments in convolutional neural networks for image processing, deep learning methods based on pixel classification have been extensively applied in medical image segmentation. One popular strategy for such tasks is the encoder‐decoder‐based U‐Net architecture and its variants. Most segmentation methods based on fully convolutional networks will cause the loss of spatial and contextual information due to continuous pooling operations or strided convolution when decreasing image resolution, and make less use of contextual information and global information under different receptive fields. To overcome this shortcoming, this paper proposes a novel structure called RAAU‐Net. In our proposed RAAU‐Net structure, which is a modified U‐shaped architecture, we aim to capture high‐level information while preserving spatial information and focusing on the regions of interest. RAAU‐Net comprises three main components: a feature encoder module that utilizes a pre‐trained ResNet‐18 model as a fixed feature extractor, a multi‐receptive field extraction module that we developed, and a feature decoder module. We have tested our method on several 2D medical image segmentation tasks such as retinal nerve, breast tumor, skin lesion, lung, gland, and polyp segmentation. All the indexes of the model reached the best in the dataset of skin lesions, in which Accuracy, Precision, IoU, Recall, and Dice Score were 3.26%, 5.42%, 9.92%, 6.52%, and 5.95% higher than UNet. [ABSTRACT FROM AUTHOR]

To aid in detection of tuberculosis, researchers have concentrated on developing computer‐aided diagnostic technologies based on x‐ray imaging. Since it generates noninvasive standard‐of‐care data, a chest x‐ray image is one of the most often used diagnostic imaging modalities in computer‐aided solutions. Due to their significant interclass similarities and low intra‐class variation abnormalities, chest x‐ray pictures continue to pose difficulty for proper diagnosis. In this paper, a novel automated framework is proposed for the classification of tuberculosis, COVID‐19, and pneumonia from chest x‐ray images using deep learning and improved optimization technique. Two pre‐trained convolutional neural network models such as EfficientB0 and ResNet50 have been utilized and fine‐tuned based on the additional layers. Both models are trained with fixed hyperparameters on the selected datasets and obtained newly trained models. A novel feature selection technique has been proposed that selects the best features. In the novel version, distance and update position formulation has been modified. The selected features are further fused using a novel technique that is based on the serial and standard deviation threshold function. The experimental process of the proposed framework is conducted on three datasets and obtained an accuracy of 98.2%, 99.0%, and 98.7%, respectively. In addition, a detailed Wilcoxon signed‐rank analysis is conducted and shows the proposed method significance performance. Based on the results, it is concluded that the proposed method accuracy is improved after the fusion process. In addition, the comparison with recent techniques shows the proposed method as more significant in terms of accuracy and precision rate. [ABSTRACT FROM AUTHOR]

Automatic survival prediction of gliomas from brain magnetic resonance imaging (MRI) volumes is an essential step for a patient's prognosis analysis. Radiomics research delivers beneficial feature information from MRI imaging which is substantially required by clinicians and oncologists for predicting disease prognosis for precise surgical treatment and planning. In recent years, the success of deep learning has been vast in the field of medical imaging, and it shows state‐of‐the‐art performance in applications like segmentation, classification, regression, and detection. Therefore, in this paper, we proposed a collective method using deep learning and radiomics techniques for the survival prediction of brain tumor patients. We first propose a hierarchical channel attention (HAM) module and a multi‐scale‐aware feature enhancement (MSAFE) to efficiently fuse adjacent hierarchical features in the proposed segmentation model. After segmentation, deep/latent features (LCNN) are extracted from the bottom layer of the proposed segmentation model. Later, we extracted selected radiomics features (histogram, location, and shape) using input images and segmented masks from the proposed segmentation model. Further, the 3D deep learning regressor has been trained for 3D regressor‐based deep feature extraction. We proposed the method of overall survival prediction for the brain tumor patients by combining all the meaningful features including clinical features (age) that also favorably contribute to the survival days prediction for the glioma's patients. To predict the survival days for each patient, the selected features are trained to analyze the performance of various regression techniques like random forest (RF), decision tree (DT), and XGBoost. Our proposed combined feature‐based method achieved the highest performance for survival days prediction over the state‐of‐the‐art methods. We also perform extensive experiments to show the effectiveness of each feature extraction method. The experimental results infer that deep learning‐based features along with radiomic features and clinical features are truly vital paradigms to estimate survival days. [ABSTRACT FROM AUTHOR]

This paper introduces an efficient wavelet thresholding strategy and robust shrinkage approach (WTS) for de‐noising the ECG signals. The optimal mother wavelet for de‐noising the ECG signal is automatically selected based on cross‐correlation and energy‐to‐entropy indices. A dynamic threshold is applied to various levels of decomposition to eliminate different types of noise. A shrinkage approach is suggested to attain seamless on–off transitions. The achievement of the WTS is compared with the state of the art reported in the literature using 180 real ECG signals collected from 60 volunteers. The ECG signals were recorded using three standard ECG machines without applying the filtering option to provide a diverse range of noisy ECG signals. The de‐noising outcomes show that the presented WTS can automatically select the optimal mother wavelet for the particular ECG signal, and it is superior to other techniques in terms of the average predicted signal‐to‐noise ratio (SNRp), the average predicted root‐mean‐square error (RMSEp), and the average execution time (AET). [ABSTRACT FROM AUTHOR]

In the field of histopathology, it has been observed that most images are of poor quality as a result of inadequate luminance and contrast. The first step in identifying and analyzing them is to enhance their quality while maintaining their inherent characteristics. This paper presents a novel method for enhancing histopathology images' important features, brightness preservation, and contrast by utilizing the adaptive intensity transformation gamma function in the wavelet domain. The RGB histopathological color image is first converted to the HSV color model. The V component is then stretched in order to compensate for any color distortion. In the proposed scheme, the stretched V component is split into four sub‐bands using discrete wavelet transform (DWT). Next, it determines the optimal value of the gamma correction function in the DWT lower sub‐band. A correction factor is applied using the singular value decomposition (SVD) technique to determine the optimal gamma function value using the nature‐inspired particle swarm optimization method (NIPSO). The corrected LL sub‐band is then applied to the inverse DWT (IDWT) together with the other unprocessed sub‐bands. In addition, the contrast of each feature is enhanced by employing an adaptive histogram equalization model that is contrast‐limited. The proposed method was evaluated through experiments and validations on a variety of histopathology images retrieved from multiple databases. Several quantitative studies have demonstrated that the proposed method outperforms existing enhancement techniques in terms of entropy, edge preservation index, contrast ratio, and Universal Image Quality Index (UIQI). This proposed technique increases contrast, while preserving the inherent characteristics of the original image, resulting in better quality histopathology images that can be used in disease diagnosis and inspection. [ABSTRACT FROM AUTHOR]

Urine sediment examination (USE) is an important issue in the detection of urinary system diseases and is a prerequisite, especially in the diagnosis of kidney diseases. The low‐contrast urine sediment images contain many particles, and these particles are superimposed or adherent, making it difficult to detect and interpret the particles. In this paper, a YOLOv5‐based urine analysis system, which is one of the fastest and most successful architectures used for object detection, is presented to identify particles in urine. The artificial intelligence‐based system, which can recognize bacteria, leukocytes, erythrocytes, crystals, yeast, cylinders, epithelium, and other particles, provides counting and reporting of components in the images obtained from the centrifuged urine sample through a microscope. The system consists of taking images from the microscope and recording them on SBC (Single Board Computer), artificial intelligence‐based software running on SBC, and database systems where urine analysis results are recorded. Different YOLOv5x architectures were used to evaluate the performance of the system and Precision, Recall, mAP, and F1_score values were obtained as metrics. According to the results obtained with the YOLOv5n, YOLOv5s, and YOLOv5m architectures, the highest mAP value of the system, which can recognize eight different particles, was 95.8% with YOLOv5m. This artificial intelligence‐based system, which will help laboratory workers, will not only save time but also eliminate the standardization differences in manual microscopy and will provide benefits as full‐time educational material. [ABSTRACT FROM AUTHOR]

Precise and unambiguous segmentation of pulmonary nodules from the CT images is imperative for a CAD framework implementation delineated for the prognosis of lung cancer. Lung nodule segmentation is an appealing research discipline for accurate dismemberment of lung cancer but the irregularity in shades, contours, and compositions, and the affinity between the tumors and the neighboring regions makes it an arduous task. This paper proffers a series atrous convolution enhanced U‐Net which uses a series of concatenated dilated convolution blocks after every stage in the encoder and decoder path. Our approach helps in obtaining the quintessential components from the feature maps, in addition to the absolute convergence of the model. It is largely assessed on the publicly accessible LIDC‐IDRI dataset. The average Dice Similarity Coefficient (DSC) obtained is 81.10% with an Intersection over Union (IoU/ Jaccard Index) of 72.24%. Exploratory outcomes prove that our architecture achieves ameliorate performance. [ABSTRACT FROM AUTHOR]

Monkeypox virus has become a popular disease as many people have been recently infected with this virus around the world. This virus causes fears of another global pandemic similar to COVID‐19 pandemic. Therefore, it is important to determine infected people at an early stage to reduce the spread of this virus. Creating an automatic approach to detect monkeypox virus is a challenging problem due to the similar appearance between skin infections. Currently, deep convolutional neural network approaches have been used in detecting and classifying skin infections. However, these approaches sever from certain challenges such as being unable to capture the global context and low accuracy rate. In this paper, we propose a novel approach based on vision transformer to automatically detect monkeypox virus in skin images. We fine‐tune vision transformer for the human monkeypox classification problems. This fine‐tuned vision transformer splits an image into image patches and then these patches are used as input to the transformer in a sequence structure, followed by added layers to detect monkeypox virus. The experimental results show that the proposed approach achieves excellent detection accuracy using a public dataset. [ABSTRACT FROM AUTHOR]

MRI is a popular imaging method for examining brain tumours. The ability to precisely segment tumours from MRI is absolutely essential for medical diagnostics and surgical planning. Manual tumour segmentation might be unrealistic for more comprehensive studies. Deep learning is the most widely used technique in medical diagnosis. For effective tumour dissection from brain MRI, this paper proposed a novel combination of FLAME and EHO Algorithm. FLAME is a type of clustering method that groups the most similar pixels in to a single cluster. EHO algorithm is one of the nature‐inspired metaheuristic optimization algorithms based on the social herding behaviour of elephants and swimming search methods. The proposed methodology's efficiency is validated through testing on various BraTS challenge datasets. The average computational time, mean squared error, peak signal to noise ratio, tanimoto coefficient, and dice score ‐ obtained are 23.3775 s, 0.213, 54.9669 dB, 54.6148%, and 84.053%, respectively. [ABSTRACT FROM AUTHOR]

This paper's primary goal is to diagnose COVID‐19 contamination based on the artificial intelligence approach automatically. We used convolutional neural network deep learning algorithm for analyzing the ECG images to detect cardiac abnormalities, consequent of the contamination by the SARS‐CoV‐2 virus, responsible for the COVID‐19 epidemic. We designed, trained, and evaluated the performance of two deep learning models (MobileNetV2 and VGG16) in detecting and distinguishing between two different classes (healthy subjects and COVID‐19 positive cases). Indeed, this virus attacks the human respiratory system, which could affect the heart system. Thus, developing a deep learning model could help for a quick and efficient diagnosis, prediction, and physician decision‐making. The performed deep learning model will be used for predicting abnormal cardiac activities consequent to the contamination by the virus. The overall classification rate achieved by the models was 99.34% and 99.67% for MobileNetV2 and VGG16, respectively. Therefore, this approach can efficiently contribute to the diagnosis of COVID‐19 contamination. [ABSTRACT FROM AUTHOR]

Skin cancer is the most common deadly disease caused due to abnormal and uncontrolled growth of cells in the human body. According to a report, annually nearly one million people are affected by skin cancer worldwide. To protect human lives from such life‐threatening diseases, early identification of skin cancer is the only precautionary measure. In recent times, there already exist numerous automated techniques to detect and classify skin lesion malignancies using dermoscopic images. However, analyzing the dermoscopic images becomes an arduous task due to the presence of troublesome features such as light reflections, illumination variations, and uneven shape and dimension. To address the challenges faced during skin cancer recognition process, in this paper, we proposed an efficient intelligent automated system to detect and discriminate the dermoscopic images into malignant or benign. The proposed skin cancer detection model utilizes the HAM10000 dataset for evaluation. The dermoscopic images acquired from the HAM10000 dataset are initially preprocessed to enhance the quality of image and thus making it fit to train the classifier. Afterward, the most significant image patterns are extracted by the AlexNet architecture without any loss of detailed information. Later on, the extracted features are inputted to the proposed Improved Adaboost‐based Aphid–Ant Mutualism (IAB‐AAM) classification model to discriminate the images into malignant and benign categories. The proposed IAB‐AAM approach witnessed extensive enhancement in classification accuracy. The enhanced performance is attributed by integrating the AAM optimization concept with the IAB model. By comparing the performance of the proposed IAB‐AAM with other modern methods in terms of different evaluation indicators namely accuracy, precision, specificity, sensitivity, and f‐measure, the efficiency of the proposed IAB‐AAM technique is analyzed. From the experimental results, it is known that the proposed IAB‐AAM technique attains a greater accuracy rate of 95.7% in detecting skin cancer classes than other compared approaches. [ABSTRACT FROM AUTHOR]

There is a lack of a safe and effective technology to monitor intracranial pressure accurately and non‐destructively. In this paper, the relationship between reduced scattering coefficient and intracranial pressure in clinical patients under different brain edema states was studied. First, the brain edema MC simulation model was constructed based on different types of brain edema. The results laid a foundation for clinical ICP detection. Second, the changes of ICP and reduced scattering coefficient caused by brain edema were studied in clinical application. It was found that when the patient developed brain edema, the reduced scattering coefficient and ICP showed corresponding changes. Finally, the mathematical model of invasive ICP and reduced scattering coefficient was established. It was found that the sensitivity of 834 nm light to brain tissue changes was higher than that of 692 nm light, which could be used to reflect the changes of brain tissue physiological state. [ABSTRACT FROM AUTHOR]

The worldwide spread of the coronavirus (COVID‐19) outbreak has proven devastating to public health. The severity of pneumonia relies on a rapid and accurate diagnosis of COVID‐19 in CT images. Accordingly, a computed tomography (CT) scan is an excellent screening tool for detecting COVID‐19. This paper proposes a deep learning‐based strategy for recognizing and segmenting a COVID‐19 lesion from chest CT images, which would introduce an accurate computer aided decision criteria for the physicians about the severity rate of the patients. Two main stages have been proposed for detecting COVID‐19; first, a convolutional neural network (CNN) deep structure recognizes and classifies COVID‐19 from CT images. Second, a U‐Net deep structure segments the COVID‐19 regions in a semantic manner. The proposed system is trained and evaluated on three different CT datasets for COVID‐19, two of which are used to illustrate the system's segmentation performance and the other is to demonstrate the system's classification ability. Experiment results reveal that the proposed CNN can achieve classification accuracy greater than 0.99, and the proposed U‐Net model outperforms the state‐of‐the‐art in segmentation with an IOU greater than 0.92. [ABSTRACT FROM AUTHOR]

For the effective diagnosis of cardio vascular disease (CVD), anatomical characteristics of the heart must be examined, which depends on segmenting the cardiac tissues of interest and then classifying them into appropriate pathological groups. In recent years, deep learning (DL)‐based computer aided design (CAD) segmentation has been employed to automate the segmentation process. Despite the evolution of several DL methods, they still fail due to the shape variation of the heart in patients and the availability of a limited amount of data. This paper proposes an effective Saliency and Active Contour‐based Attention UNet3+ algorithm to segment the ventricles of the heart, which is a challenging task for most researchers, especially with an irregularly shaped right ventricle (RV) that varies over cardiac phases. The algorithm outperforms other state‐of‐the‐art methods in DC metrics, which proves its efficiency in automating the segmentation process. [ABSTRACT FROM AUTHOR]

To achieve accurate segmentation of thyroid nodule ultrasound images, obtain information on the physiological parameters of the lesion area and guide the clinical formulation of individualized treatment plan, an improved network based on the U2‐Net model is proposed in this paper. Thyroid images of 264 patients and 215 healthy volunteers at the First Hospital of Shanxi Medical University from February 2016 to June 2022 are studied, and the digital database thyroid image (DDTI) data set is proposed for data expansion. The experimental results show that the Dice coefficient on the test set was 80.58%, and the mean intersection over union was 81.21%. The improved U2‐Net model has the best segmentation accuracy compared with similar models, realizes the automatic segmentation of thyroid nodules, provides help for manual segmentation, and has good application prospects and clinical value. [ABSTRACT FROM AUTHOR]

Breast cancer (BC) is the second leading cause of death in women worldwide, caused by the uncontrolled proliferation of malignant cells in the breast. It is very tough to differentiate cancer from non‐cancer cells because of their shape and structure. Therefore, early diagnosis and classification of breast tumors as benign is a hot research topic. However, existing methods for early diagnosis have shortcomings like high computational cost and low classification accuracy. To trounce those limitations, this paper presented a computer‐aided diagnosis (CAD) system employing metaheuristic algorithms and machine learning classifiers. The proposed system uses fused features of ultrasound (US) and elastography images to enhance classification accuracy. Firstly, both US and elastography images are separately preprocessed using an enhanced wavelet thresholding method to diminish speckle noise. Secondly, a new segmentation algorithm is proposed to extract the lesion region from the preprocessed US image. Then the contour of the segmented lesion is mapped on an elastography image to obtain the corresponding region of interest (ROI), and texture and morphology features are separately extracted and concatenated. Subsequently, the enhanced grasshopper optimization algorithm (EGOA) is adopted to select salient features and reduce the dimension. Finally, based on the selected features, two machine learning classifiers, linear discriminant analysis (LDA) and support vector machine (SVM), are employed to classify the lesions into benign and malignant. The effectiveness of the proposed method is validated on 520 tumor images, including both benign and malignant samples. The suggested performance has been assessed by computing the confusion matrix and receiver operating characteristic curve. The developed method, EGOA‐SVM, outperforms the other classifiers by reaching a higher classification rate of 99.4%, specificity of 100%, the sensitivity of 99.2%, and Matthew's coefficient of 0.979. The introduced method can improve the radiologist's performance in categorizing breast lesions with a good classification rate. It could be a promising tool for clinical use. [ABSTRACT FROM AUTHOR]

Recently, automatic diagnostic approaches widely use various retinal images to classify ocular diseases. And retinal vein occlusion (RVO) is the second most common retinal vascular disease after diabetic retinopathy. In clinical practice, ophthalmologists are usually accustomed to resorting to images of one modality. But single‐modality images often ignore other modality‐specific information. To solve this problem, this paper uses a novel retinal imaging, the multicolor (MC) imaging, for RVO recognition. It can obtain four multiple modal images with different wavelengths to provide much richer information about retinal features. Since the MC images contain local and global pathologies at multiple scales, a multiscale attention structure is proposed to recognize RVO. In simple terms, this structure uses Resnet as the backbone network for feature extraction, with simultaneous input of images in four modalities. Then, the feature maps at different scales are fed into an attention module to fuse the global and local features, which combines two attention mechanisms, the channel attention mechanism and the spatial attention mechanism. The extensive experimental results demonstrate that our proposed framework achieves quite promising classification performance on the fundus diseases and normal images. [ABSTRACT FROM AUTHOR]

Image‐Guided Radiation Therapy (IGRT) is a cancer treatment method. IGRT usually follows a fragmented treatment track. The weight loss of a patient causes the first treatment plan to be flawed. Medical experts can avert retaking Fan Beam Computed Tomography (FBCT) images before each session of the therapy if the Cone Beam Computed Tomography (CBCT), which is mounted on a Medical Linear Accelerator (LINAC), gives high‐quality images. Rather than improvising the CBCT machine, deep‐learning techniques can be used to enhance the CBCT images. This paper proposes a Feature‐Embedded Variational AutoEncoder (FE‐VAE) for CBCT image enhancement. A feature embedding is incorporated using a MultiResUNet‐based feature extractor to preserve structural information in the input image. A perceptual loss function, which is the product of Mean Absolute Error (MAE) and inverse Structural Similarity Index Measure (SSIM), is proposed in this work. The model is trained using Head‐and‐Neck (H&N) CBCT‐FBCT pairs of 59 cancer patients. Our proposed architecture successfully generates images much closer to the ground truth FBCT images. The model gives an average peak signal‐to‐noise ratio of 32.89 dB, mean squared error of 297.36, MAE of 32.37 HU, and SSIM of 0.99. The trained model gave the least deviated SSIM and MAE, implying that the model is well‐optimised. The experimentation with different losses proves the prominence of the proposed loss function. The visual quality of the results obtained is also comparable with the ground truth. The results indicate that the model produces enhanced CBCT images. Hence, the proposed FE‐VAE is useful in fractionated radiotherapy to minimise the number of scanning sessions for replanning the dosimetry. [ABSTRACT FROM AUTHOR]

Fundus disorders like Glaucoma are a serious health risk that affects people all over the world, lowering their quality of life. Optic disc (OD) segmentation is a crucial step for automatic detection of glaucoma with fundus images. This paper presents an efficient and accurate OD segmentation methodology from eye fundus images. In the proposed methodology, fundus images are transformed from RGB (Red, Green, and Blue) color space to LAB (Luminance, A‐axis, B‐axis) color space. A new image is generated using the luminance, A‐axis component, and image information content. As a pre‐segmentation stage, the unsupervised approach extracts the region of interest containing OD. The adaptive threshold method, followed by morphological operations, extracts the contours of the OD region and removes the spur. Segmentation of OD is a challenging task due to similar intensity levels with neighborhood, vascular occlusion, and retinal atrophy. The proposed work overcomes the challenges by segmentation of the OD at three levels, which enhances the functionality at different illumination conditions. The features are extracted at various levels to determine whether the segmented image contains the OD or not. The efficiency of the proposed methodology demonstrates by applying it to a variety of fundus images from datasets, namely ISBI IDRiD, DRISHTI‐GS, and SYSU. Key performance metrics like Intersection over Union and Dice Coefficient are used to evaluate the performance of the proposed methodology. The proposed methodology achieves Intersection over Union of 0.996, 0.997, and 0.994 on ISBI IDRiD, DRISHTI‐GS, and SYSU datasets, respectively. Similarly, the proposed methodology obtains 0.96, 0.97, and 0.98 Dice Coefficient on the three datasets. [ABSTRACT FROM AUTHOR]

Glioma represents one of the most aggressive cancers, which can develop in the brain. The automatic tumor segmentation and its sub‐regions represent a challenging task owing to their considerable structural variation. It can appear in different ways and with several shapes, which makes tissue identification a crucial task. Reliable and accurate segmentation presents an important component in tumor treatment and diagnosis planning. To overcome these drawbacks, various Deep Learning (DL) schemes are proposed to aid doctors. In this paper, a novel Convolutional Neural Network (CNN) scheme for glioma segmentation was proposed. Our suggested technique consists of three phases. First, we use intensity normalization to ameliorate the image quality as a preprocessing step. Second, an automatic segmentation technique based on CNN has been proposed. The new model has several layers. Finally, and with the goal of refining the segmentation results, we employ a post‐processing approach. We use the public benchmark BraTS datasets from 2018 and 2020 with low‐grade and high‐grade glioma tumors to test the suggested framework. These datasets contain about 285 and 369 patients, respectively. The four modalities are exploited. Each patient has about 155 2D images from every modality. All the images have the same size (240 × 240 pixels). Our technique performs well compared to new methods, with Dice scores of 0.88 for the Whole Tumor, 0.84 for Tumor Core, and 0.71 for Enhancing Tumor based on the first dataset. According to the second dataset, the three regions had an average of 0.88, 0.9, and 0.75, respectively. The Jaccard indexes for the first data set are 0.8, 0.73, and 0.56 for the three regions, respectively. The second data set attains 0.8, 0.82, and 0.6 for the three regions. The results show that the suggested framework is an excellent way to segment data, especially compared to other methods. [ABSTRACT FROM AUTHOR]

Medical image segmentation is a key step in medical image analysis. The small differences in the background and foreground of medical images and the small size of most medical data sets make medical segmentation difficult. This paper uses a global‐local training strategy to train the network. In the global structure, ResNest is used as the backbone of the network, and parallel decoders are added to aggregate features, as well as gated axial attention to adapt to small datasets. In the local structure, the extraction of image details is accomplished by dividing the images into equal patches of the same size. To evaluate the performance of the model, qualitative and quantitative comparisons were performed on five datasets, Kvasir‐SEG, CVC‐ColonDB, CVC‐ClinicDB, CVC‐300, and ETIS‐LaribPolypDB, and the segmentation results were significantly better than the current mainstream polyp segmentation methods. The results show that the model has better segmentation performance and generalization ability. [ABSTRACT FROM AUTHOR]

The current paper proposes a new hierarchical procedure for efficient diagnosis of lung cancer computed tomography (CT) images. Here, after noise removal based on median filtering, a contrast enhancement based on general histogram equalization (GHE) has been utilized. Then, a modified version of K‐means clustering has been used for the area of interest segmentation in the CT images. The major characteristics of the segmented images have been selected during an optimization technique and the outputs are injected into an optimized radial basis function (RBF) network for the final classification. Optimization in the classification stage and feature selection is by an improved metaheuristic technique, called Amended Whale Optimization Algorithm was proposed. The designed method is then applied to "The RIDER Lung CT" database and its achievements are validated by several latest techniques to show its higher efficacy. [ABSTRACT FROM AUTHOR]

Mitral valve (MV) diseases constitute one of the etiologies of cardiovascular mortality and morbidity. MV pathologies need evaluating and classifying via echocardiographic videos. Transformers have significantly advanced video analytics. MV motion is divided by Carpentier functional classification into four types: normal, increased, restricted, and restricted only during systole. This paper introduces CarpNet, a deep transformer network that incorporates video transformers capable of direct MV pathology Carpentier's classification from the parasternal long‐axis (PLA) echocardiographic videos. The network, instead of processing frames independently, analyzes stacks of temporally consecutive frames using multi‐head attention modules to incorporate MV temporal dynamics into the learned model. To that end, different convolutional neural networks (CNNs) are evaluated as the backbone, and the best model is selected using the information of the PLA view. The use of information obtained by our proposed deep transformer network from consecutive echocardiographic frames yielded better results concerning the Carpentier functional classification than information obtained by CNN‐based (single‐frame) models. Using the Inception_Resnet_V2 architecture as the backbone, CarpNet achieved 71% accuracy in the test dataset. Deep learning and transformers in echocardiographic videos can render quick, precise, and stable evaluations of various MV pathologies. [ABSTRACT FROM AUTHOR]

Chronic pain is a common problem among stroke patients, resulting from neurological damage to the central nervous system. This discomfort is primarily caused by the improper use of unaffected limbs or musculoskeletal issues. It can be challenging to differentiate neuropathic pain resulting from central nervous system damage. To address these challenges, researchers have developed a cutting‐edge technology called a Brain‐Computer Interface (BCI) based on electroencephalogram (EEG) data. In this paper, a novel BCI classifier has been developed using the Weighted Incremental‐Decremental Support Vector Machine (WIDSVM) classification method. The classifier has been trained using EEG‐based motor images from patients with central nervous system damage. The Quantum Chaos Butterfly Optimization Algorithm (QCBOA) has been used to enhance the performance of the WIDSVM classifier by creating a new dataset. The efficiency of the proposed model has been evaluated by comparing the results obtained from normal participants and those who developed chronic pain. The classification accuracy has been calculated for different regions, including the left hand, right hand, and feet, among the different participant groups. A total of 28 participants have been separated into three groups with pain in different regions, such as the lower abdomen and legs. The classifier has been tested using both 3‐channel bipolar montages and Common Spatial Patterns (CSPs). The results have shown that the proposed model offers higher classification accuracy and statistical significance in identifying the patient's risk of developing central neuropathic pain. However, it is important to note that further studies with larger sample sizes and different types of chronic pain are needed to validate the efficacy of the proposed model. [ABSTRACT FROM AUTHOR]

Image segmentation requires large amounts of annotated data. However, collecting massive datasets with annotations is difficult since they are expensive and labor‐intensive. The unsupervised domain adaptation (UDA) for image segmentation is a promising approach to address the label‐scare problem on the target domain, which enables the trained model on the source labeled domain to be adaptive to the target domain. The adversarial‐based methods encourage extracting the domain‐invariant features by training a domain discriminator to mitigate the domain gap. Existing UDA segmentation methods fail to obtain satisfied segmentation results as they only consider the global knowledge of output space while neglecting the local information of feature space. In this paper, a fusing feature and output (FFO) space method is proposed for UDA, which in the context of medical image segmentation. The proposed model is learned by training a more powerful domain discriminator, which considers features extracted from both feature space and output space. Extensive experiments carried out on several medical image datasets show the adaptation effectiveness of our approach in improving the segmentation performance. [ABSTRACT FROM AUTHOR]

This paper proposes a novel three‐dimensional convolution neural network‐based modified bidirectional long short‐term memory with pelican optimization (3D CNN based MBiLSTM with PO) algorithm for multiclass ovarian tumor detection. Initially, the International Collaboration on Cancer Reporting endometrial cancer dataset images are provided in pre‐processing phase, which uses a pre‐emphasis filter to process the input image. In the segmentation phase, pre‐processed data is then partitioned into diverse subgroups (i.e., pixels), which minimizes the complexity of images. In this paper, a factorization‐based active contour technique is employed in the effective segmentation of images. The segmented features are then extracted and classified using the 3D CNN‐MBiLSTM with PO algorithm. Finally, the experimental results are conducted and compared with various other approaches for various performance metrics. Each metric is evaluated with respect to the different number of iterations. The accuracy, sensitivity, and specificity have reached a higher value of 98.5%, 96%, and 98.25%, respectively. [ABSTRACT FROM AUTHOR]

Diabetic macular edema (DME) is a typical fundus disease that can cause blindness in severe cases. The morphology of the inner limiting membrane (ILM) to the retinal pigment epithelium (RPE) layer in the retina and the macular edema (ME) area are important features for the diagnosis of DME. Doctors usually use non‐invasive and high‐resolution optical coherence tomography (OCT) to examine the fundus of the patient. However, manual diagnosis has low efficiency and strong subjectivity. Realizing the automatic segmentation of the ILM‐RPE layer and ME is extremely important for the early diagnosis of DME. In this paper, the attention mechanism based on residual convolution module U‐Net (RCU‐Net) is proposed for the automatic segmentation of the retinal layer and cystoid edema lesions. Through the fusion of the residual structure and CBAM for feature extraction, the useful features in the channel and space are effectively strengthened, and the network can better learn different levels of information. The proposed network is combined with the Lovász‐softmax loss, which can better target the correlation between targets to obtain the optimal segmentation model during training. Finally, this paper compares the proposed method with several other segmentation methods. The experimental results show that the MIoU of the method in this model reaches 88.595%, and the Accuracy reaches 99.171%. The RCU‐Net proposed in this paper is used to segment the ILM‐RPE layer and ME region in the retina OCT B‐scan images, and its overall performance is better than other networks. [ABSTRACT FROM AUTHOR]

Deep learning has recently been widely used in medical image analysis due to its ability to learn complex features in images. An increasing number of deep learning methods has been devoted to the classification of chest X‐ray (CXR) images. Most of the existing deep learning methods are based on classical pretraining models trained by global chest X‐ray images. In this paper, we classify 14 different thorax diseases based on the RE‐DSENet model, which applies the image registration algorithm to the multi‐atlas segmentation model. The experimental results on the ChestX‐ray14 dataset show that the RE‐DSENet's classify‐cation precision of 14 thorax diseases is significantly improved compared with three existing classic algorithms, and the average AUC (area under the ROC curve) score is up to 0.857. RE‐DSENet can also reduce the computations in the classification network and improve the training efficiency. In addition, the focus areas diagnosed by the RE‐DSENet model are visualized in this paper, and the results further prove the effectiveness of the model. [ABSTRACT FROM AUTHOR]

Pulmonic nodules are unusual growing of tissues; originate on one lung or both lungs. They are the round, trifling mass of soft tissues in the lung area. Habitually, pulmonic nodules are indications of lung tumors, but they may be nonthreatening. When identified earlier and treated in time, the patient's life expectancy increases. The anatomy of the lung is highly interconnected in nature, which makes it difficult to diagnose pulmonic nodules by diverse clinical imaging practices. A network model is presented in this paper for accurate classification of pulmonic nodules from computed tomography scans images. The lung images are subjected to semantic segmentation using Attention U‐Net to isolate the pulmonary nodules. The proposed Directional Hexagonal Mixed Pattern is applied to generate a new texture pattern. Then, the nodules are classified by combining the proposed multilevel network model with the self‐attention network. This paper also demonstrates an experimental arrangement called tenfold cross‐validation without a segmentation mask, in which the nodules that had been marked as less than 3 mm by radiologists are discarded. This has obtained an improved result. The experimental results show that with and without segmentation masks the proposed classifier scores an accuracy of 90.48% and 91.83%. In addition, it has efficiently produced the measure of area under curve as 98.08%. [ABSTRACT FROM AUTHOR]

Lung cancer is the leading cause of death for men and women worldwide and the second most frequent cancer. Therefore, early detection of the disease increases the cure rate. This paper presents a new approach to evaluate the ability of positron emission tomography/computed tomography (PET/CT) images to classify and detect lung cancer using deep learning techniques. Our approach aims to fully automate lung cancer's anatomical localization from PET/CT images. It also searches to classify the tumor, which is essential as it makes it possible to determine the disease's speed of progression and the best treatments to adopt. We have built, in this work, an approach based on transformers by implementing the DETR model as a tool to detect the tumor and assist physicians in staging patients with lung cancer. The TNM staging system and histologic subtype classification were both taken as a standard for classification. Experimental results demonstrated that our approach achieves sound results on tumor localization, T staging, and histology classification. Our proposed approach detects tumors with an intersection over union (IOU) of 0.8 when tested on the Lung‐PET‐CT‐Dx dataset. It also has yielded better accuracy than state‐of‐the‐art T‐staging and histologic classification methods. It classified T‐stage and histologic subtypes with an accuracy of 0.97 and 0.94, respectively. [ABSTRACT FROM AUTHOR]

The 2019 coronavirus (COVID‐19), started in China, spreads rapidly around the entire world. In automated medical imagery diagnostic technique, due to presence of noise in medical images and use of single pre‐trained model on low quality images, the existing deep network models cannot provide the optimal results with better accuracy. Hence, hybrid deep learning model of Xception model & Resnet50V2 model is proposed in this paper. This study suggests classifying X‐ray images into three categories namely, normal, bacterial/viral infections and Covid positive. It utilizes CLAHE & BM3D techniques to improve visual clarity and reduce noise. Transfer learning method with variety of pre‐trained models such as ResNet‐50, Inception V3, VGG‐16, VGG‐19, ResNet50V2, and Xception are used for better feature extraction and Chest X‐ray image classification. The overfitting issue were resolved using K‐fold cross validation. The proposed hybrid deep learning model results high accuracy of 97.8% which is better than existing techniques. [ABSTRACT FROM AUTHOR]