Search

Convolutional neural networks (CNNs) for automatic classification and medical image diagnosis have recently displayed a remarkable performance. However, the CNNs fail to recognize original images rotated and oriented differently, limiting their performance. This paper presents a new capsule network (CapsNet) based framework known as the multi-lane atrous feature fusion capsule network (MLAF-CapsNet) for brain tumor type classification. The MLAF-CapsNet consists of atrous and CLAHE, where the atrous increases receptive fields and maintains spatial representation, whereas the CLAHE is used as a base layer that uses an improved adaptive histogram equalization (AHE) to enhance the input images. The proposed method is evaluated using whole-brain tumor and segmented tumor datasets. The efficiency performance of the two datasets is explored and compared. The experimental results of the MLAF-CapsNet show better accuracies (93.40% and 96.60%) and precisions (94.21% and 96.55%) in feature extraction based on the original images from the two datasets than the traditional CapsNet (78.93% and 97.30%). Based on the two datasets’ augmentation, the proposed method achieved the best accuracy (98.48% and 98.82%) and precisions (98.88% and 98.58%) in extracting features compared to the traditional CapsNet. Our results indicate that the proposed method can successfully improve brain tumor classification problems and support radiologists in medical diagnostics.

The hippocampus is closely related to many brain diseases, such as Alzheimer's disease. Accurate measurement of the hippocampus is helpful for clinicians in identifying lesions and then diagnosing and treating the related brain diseases. Therefore, accurate segmentation of the hippocampus is of vital significance for the in-depth study of many brain diseases. However, the accurate measurement of the hippocampus depends on its accurate segmentation, and hippocampal segmentation has always been a challenging problem due to the small size, irregular shape, and fuzzy boundaries with surrounding tissues of the hippocampus. With the development of machine learning, many innovative methods have been proposed to segment the hippocampus. The purpose of this survey is to provide a comprehensive overview of hippocampal segmentation in brain MRI images using machine learning methods. First, a brief introduction to hippocampal segmentation in brain MRI images is given. Then, common evaluation metrics of hippocampal segmentation are introduced. Next, brain hippocampal segmentation methods based on traditional machine learning and deep learning are described. Subsequently, some common open datasets and toolkits applied to brain hippocampal segmentation are presented. Finally, objective conclusions regarding hippocampal segmentation in brain MRI images using machine learning methods are drawn, and future developments and trends are identified for brain hippocampal segmentation.

Magnetic resonance imaging is a standard modality used in medicine for bone diagnosis and treatment. It offers the advantage to be a non-invasive technique that enables the analysis of bone tissues. The early detection of tumor in the bone leads on saving the patients' life through proper care. The accurate detection of tumor in the MRI scans are very easy to perform. Furthermore, the tumor detection in an image is useful not only for medical experts, but also for other purposes like segmentation and 3D reconstruction. The manual delineation and visual inspection will be limited to avoid time consumption by medical doctors. The bone tumor tissue detection allows localizing a mass of abnormal cells in a slice of magnetic resonance (MR).

Manual tumor diagnosis from magnetic resonance images (MRIs) is a time-consuming procedure that may lead to human errors and may lead to false detection and classification of the tumor type. Therefore, to automatize the complex medical processes, a deep learning framework is proposed for brain tumor classification to ease the task of doctors for medical diagnosis. Publicly available datasets such as Kaggle and Brats are used for the analysis of brain images. The proposed model is implemented on three pre-trained Deep Convolution Neural Network architectures (DCNN) such as AlexNet, VGG16, and ResNet50. These architectures are the transfer learning methods used to extract the features from the pre-trained DCNN architecture, and the extracted features are classified by using the Support Vector Machine (SVM) classifier. Data augmentation methods are applied on Magnetic Resonance images (MRI) to avoid the network from overfitting. The proposed methodology achieves an overall accuracy of 98.28% and 97.87% without data augmentation and 99.0% and 98.86% with data augmentation for Kaggle and Brat's datasets, respectively. The Area Under Curve (AUC) for Receiver Operator Characteristic (ROC) is 0.9978 and 0.9850 for the same datasets. The result shows that ResNet50 performs best in the classification of brain tumors when compared with the other two networks.

Ewing’s sarcoma (ES) is a tumor that often occurs in the long bones and rarely arises from visceral organs primarily. Here, we report a case of primary hepatic ES, discuss its computed tomography (CT) and gadobenate dimeglumine-enhanced magnetic resonance (MRI) features. This is the first Chinese and fifth primary hepatic ES case reported, based on a literature review. Imaging examinations showed that the tumor was solid, with necrosis and hemorrhage. Contrast-enhanced images showed that the tumor was hypervascular and especially had heterogeneous signal intensity on hepatobiliary phase MRI images. Intratumoral vessels and vascular invasion were also present.

The magnetic resonance imaging (MRI) image processing capabilities were investigated based on the improved particle swarm optimization (IPSO) algorithm, and the clinical application analysis of MRI images in the diagnosis of placenta accreta (PA) was evaluated in this study. The MRI uterine images were detected on the basis of IPSO. Besides, the clinical data of 89 patients with PA were selected and collected, who were diagnosed by clinical cesarean section surgery and pathological comprehensive diagnosis in hospital from January 2018 to July 2020. Then, all of them underwent the ultrasound (US) and MRI examinations, and the differences of sensitivity, specificity, and accuracy between MRI and US under IPSO in the diagnosis of PA were compared, as well as the differences in the diagnosis of adhesive, implantable, and penetrated PA. The results showed that the difference in detection between IPSO-based MRI images and US images was not statistically substantial ( p > 0.05 ), but the number of initial detections was higher than the number of US examination. MRI examination had higher sensitivity and specificity in the diagnosis of PA during pregnancy, especially for implantable PA, compared with US examination ( p < 0.05 ). In conclusion, MRI images based on the improved particle swarm optimization algorithm showed a good application effect in the diagnosis of placental implantation diseases, which was worthy of further promotion in clinical practice.

One of the most prevalent forms of tumor found in males all over the world is prostate cancer. The main risk factors are age and family history. Magnetic Resonance Imaging (MRI) is highly recommended for detecting and localizing prostate cancer. It is very important for precise segmentation of the prostate region in MRI scans to improve the treatment possibilities and the chance of patient survival with prostate cancer. Manually segmenting the prostate region is a daunting task and often time-consuming because of the variation in shapes of prostates among patients, poor delineation of the boundary, and the use of different MRI modes. In this paper, we propose an automatic segmentation model for the prostate regions in MRI scans based on Unet and Xception net. To boost the performance of model, local residual connections are added in the decoder stage of the Unet. The empirical results are compared to different Unet based models with different preprocessing methods to assess the effectiveness of the proposed model. The experimentations are performed to support the fact that the proposed model performs better than other methods taken under study.

This work aimed to explore the relationship between spine surgery lumbar facet joint (LFJ) and lumbar disc herniation (LDH) via compressed sensing algorithm-based MRI images to analyze the clinical symptoms of patients with residual neurological symptoms after LDH. Under weighted BM3D denoising, Epigraph method was introduced to establish the novel CSMRI reconstruction algorithm (BEMRI). 127 patients with LDH were taken as the research objects. The BEMRI algorithm was compared with others regarding peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM). Patients’ bilateral LFJ angles were compared. The relationships between LFJ angles, lumbar disc degeneration, and LFJ degeneration were analyzed. It turned out that the PSNR and SSIM of BEMRI algorithm were evidently superior to those of other algorithms. The proportion of patients with grade IV degeneration was at most 31.76%. Lumbar disc grading was positively correlated with change grading of LFJ degeneration ( P < 0.001 ). LFJ asymmetry was positively correlated with LFJ degeneration grade and LDH ( P < 0.001 ). Incidence of residual neurological symptoms in patients aged 61–70 years was as high as 63.77%. The proportion of patients with severe urinary excretion disorders was 71.96%. Therefore, the BEMRI algorithm improved the quality of MRI images. Degeneration of LDH was positively correlated with degeneration of LFJ. Asymmetry of LFJ was notably positively correlated with the degeneration of LFJ and LDH. Patients aged 61–70 years had a high incidence of residual neurological symptoms after surgery, most of which were manifested as urinary excretion disorders.

This paper combines optimized CSMRI algorithm (CS) and magnetic resonance imaging (MRI) to shorten the scanning time of MRI image data and improve the imaging quality. At the same time, the paper applies functional magnetic resonance imaging (BOLD-fMRI) based on the principle of blood oxygen level dependence to explore the application value of the nerve function reconstruction therapy system for the rehabilitation of active and passive motor functions in patients with acute cerebral infarction. Methods. In this paper, 20 patients with acute cerebral infarction were included. The random drawing method was used to divide them into active group and passive group, each with 10 cases. Both groups were treated with conventional medication and acupuncture. The active group used the active mode of the nerve function reconstruction treatment system to guide the patients’ limb active exercise; all training in the passive group is provided by the nerve function reconstruction treatment system to passively exercise the patients’ limbs; both groups undergo BOLD-fMRI examination before treatment and after 2 weeks of treatment and observe the activated parts of the brain functional area and corresponding parts of the two groups before and after treatment. We observe the activation volume and, at the same time, the ADL score. Results. After treatment, the activation volume and ADL scores of brain functional areas in the two groups were significantly improved compared with those before treatment, and the difference was statistically significant ( P < 0.05 ). Conclusion. The combination of optimized CSMRI algorithm (CS) and magnetic resonance imaging (MRI) can be used to evaluate the early rehabilitation efficacy of patients with acute cerebral infarction and has certain guiding value for clinical treatment.

To explore the performance of the improved DenseNet network in diagnosing pulmonary nodules (PNs) and differentiating benign and malignant PNs, improved DenseNet network was applied to segment MRI images of 60 PN patients, which were defined as the test group, while those segmented by the traditional one were undertaken as the control group. The MRI results were compared with the pathological diagnostic results, and the segmentation effects were evaluated factoring in precision, recall, Dice similarity coefficient, and intersection-over-union (IoU). The results showed that the improved DenseNet network algorithm showed higher accuracy, recall rate, Dice coefficient, and IoU versus the traditional one, and the difference was notable ( P < 0.05 ). The improved DenseNet network algorithm had higher diagnostic accuracy in terms of the PN volume, lobes, burrs, edges, and adhesion to surrounding tissue, with notable differences noted ( P < 0.05 ). The accuracy in differentiating benign and malignant PNs in the test group was higher (92.38 ± 8.74% vs. 75.56 ± 7.56%) versus the control group, and the difference was notable ( P < 0.05 ). In short, the MRI image segmentation algorithm based on the improved DenseNet network shows high accuracy in diagnosing PNs and differentiating benign and malignant PNs, and it is worthy of further promotion in the clinic.

This study focused on the application value of MRI images processed by a Support Vector Machine (SVM) algorithm-based model in diagnosis of benign and malignant solitary pulmonary nodule (SPN). The SVM algorithm was constrained by a self-paced regularization item and gradient value to establish the MRI image segmentation model (SVM-L) for lung. Its performance was compared factoring into the Dice index (DI), sensitivity (SE), specificity (SP), and Mean Square Error (MSE). 28 SPN patients who underwent the parallel MRI examination were selected as research subjects and were divided into the benign group (11 patients) and malignant group (17 patients) according to different plans for diagnosis and treatment. The apparent diffusion coefficient (ADC) at different b values was analyzed, and the steepest slope (SS) and washout ratio (WR) values in the two groups were calculated. The result showed that the MSE, DI, SE, SP values, and operation time of the SVM-L model were (0.41 ± 0.02), (0.84 ± 0.13), (0.89 ± 0.04), (0.993 ± 0.004), and (30.69 ± 2.60)s, respectively, apparently superior to those of the other algorithms, but there were no statistic differences ( P > 0.05 ) in the WR value between the two groups of patients. The SS values of the time-signal curve in the benign and malignant groups were (2.52 ± 0.69) %/s and (3.34 ± 00.41) %/s, respectively. Obviously, the SS value of the benign group was significantly lower than that of the malignant group ( P < 0.01 ). The ADC value with different b values in the benign group was significantly lower than that of the malignant group ( P < 0.01 ). It suggested that the SVM-L model significantly improved the quality of lung MRI images and increased the accuracy to differentiate benign and malignant SPN, providing reference for the diagnosis and treatment of SPN patients.

Brain tumor detection successfully in early-stage plays important role in improving patient treatment and survival. Evaluating magnetic resonance imaging (MRI) images manually is a very difficult task due to the numerous numbers of images produced in the clinic routinely. So, there is a need for using a computer-aided diagnosis (CAD) system for early detection and classification of brain tumors as normal and abnormal. The paper aims to design and evaluate the convolution neural network (CNN) Transfer Learning state-of-the-art performance proposed for image classification over the recent years. Five different modifications have been applied to five different famous CNN to know the most effective modification. Five-layer modifications with parameter tuning are applied for each architecture providing a new CNN architecture for brain tumor detection. Most brain tumor datasets have a small number of images to train the deep learning structure. Therefore, two datasets are used in the evaluation to ensure the effectiveness of the proposed structures. Firstly, a standard dataset from the RIDER Neuro MRI database including 349 brain MRI images with 109 normal images and 240 abnormal images. Secondly, a collection of 120 brain MRI images including 60 abnormal images and 60 normal images. The results show that the proposed CNN Transfer Learning with MRI’s can learn significant biomarkers of brain tumor, however, the best accuracy, specificity, and sensitivity gained is 100% for all of them.

Tumor is an unusual growth of the tissues in any part or organ of the human body. A brain tumor is an abnormal growth of cells in the brain or central spine canal. Brain tumors can be malignant (cancerous) or benign (non-cancerous) and have different treatments. Any extra growth of cells in the brain can result in pressure inside the skull as it is rigid and hence damages the brain. In the recent years, a progressive research has been done in this medical imaging field to more accurately detect the brain tumors. This is very important for early treatment. The proposed model detects the location and dimensions of the tumor present in the brain of a patient accurately using deep convolutional neural networks. In this paper we have presented the two models, Inception V3 and VGG-Net. The main aim of this paper is to find the best technique to detect the brain tumor from the MRI scanned images. The proposed models have given good results with good accuracy.

Test the hypotheses that (1) cardinal ligament (CL) straightening and lengthening occur with parity and prolapse, (2) CL straightening occurs before lengthening, and (3) CL length is correlated with level III measures. We performed a secondary analysis of MRIs from women in three groups: (1) nulliparous with normal support, (2) parous with normal support, and (3) uterine prolapse (POP-Q point C > − 4 and Ba > 1 cm). The 3D stress MRI images at rest and maximal Valsalva were analyzed. CLs were traced from their origin to cervico-vaginal insertions. Curvature ratio was calculated as curved length/straight length. Level III measures included urogenital hiatus (UGH), levator hiatus (LH), and levator bowl volume (LBV), and their correlations with CL length were calculated. Ten women were included in each group. Compared to the nulliparous group, CL length was 18% longer in parous controls (p = .04) and 59% longer with prolapse (p < .01) at rest, while at Valsalva, CL length was 10% longer in parous controls (p = .21) and 49% longer with prolapse (p < .01). Curvature ratios showed 18% more straightening in women with prolapse compared to parous controls (p < .01). Curved CL length and level III measures were moderately to strongly correlated: UGH (rest: R = 0.68, p < .01; Valsalva: R =0.80, p

To explore an accurate radiologic–pathologic control method in soft tissue tumor. First, we performed a preliminary experimental study of the radiologic–pathologic control method in soft tissue mass simulation model. Targeted localization markers were performed during MRI, surgery, and pathology, to ensure accurate correspondence between MRI slices and pathological sections. And the MRI image after printing was matched with the pathological material surface to obtain the pathological material corresponding to the region of interest (ROI). Second, we performed the same method on soft tissue tumor clinical research population. The long and short diameters of the tumor MRI slices and pathological sections were measured. A paired t test was used to assess differences between them. A paired t test showed that there were no signifacant differences of the length and diameter of the simulation tumor measured between MRI slices and pathological sections in soft tissue tumor simulation model and clinical research population (p > 0.05). The “Body position line-MRI slice-Pathological section Five-step Control Method” created by this study could carry out a more accurate radiologic–pathologic comparison study in soft tissue tumors.

Stroke is a kind of cerebrovascular disease that heavily damages people’s life and health. The quantitative analysis of brain MRI images plays an important role in the diagnosis and treatment of stroke. Deep neural networks with massive data learning ability supply a powerful tool for lesion detection. In order to study the property of the stroke lesions and complete intelligent automatic detection, we collaborated with two authoritative hospitals and collected 5,668 brain MRI images of 300 ischemic stroke patients. All the lesion regions in the images were accurately labeled by professional doctors to ensure the authority and effectiveness of the data. Three categories of deep learning object detection networks including Faster R-CNN, YOLOV3, and SSD are applied to implement automatic lesion detection with the best precision of 89.77%. Meanwhile, statistical analysis of the locations, shapes of the lesions, and possible related diseases is conducted with valid conclusions. The research contributes to the intelligent assisted diagnosis and prevention and treatment of ischemic stroke.

MRI was used to measure the changes in the angle of the facet joints of the lumbar spine and analyze the relationship between it and the herniated lumbar intervertebral disc. Analysis of the causes of lumbar disc herniation from the anatomy and morphology of the spine provides a basis for the early diagnosis and prevention of lumbar disc herniation. There is a certain correlation between the changes shown in MRI imaging of lumbar disc herniation and the TCM syndromes of lumbar intervertebral disc herniation. There is a correlation between the syndromes of lumbar disc herniation and the direct signs of MRI: pathological type, herniated position, and degree of herniation. Indirect signs with MR, nerve root compression and dural sac compression, are related. The MRI examination results can help syndrome differentiation to improve its accuracy to a certain extent. MRI has high sensitivity for the measurement of the angle of the facet joints of the lumbar spine and can be used to study the correlation between the changes of the facet joint angles and the herniated disc. Facet joint asymmetry is closely related to lateral lumbar disc herniation, which may be one of its pathogenesis factors. The herniated intervertebral disc is mostly on the sagittal side of the facet joint, and the facet joint angle on the side of the herniated disc is more sagittal. The asymmetry of the facet joints is not related to the central lumbar disc herniation, and the angle of the facet joints on both sides of the central lumbar disc herniation is partial sagittal.

Objectives Brain neoplasms or intracranial tumors, which are more common in older adults, can affect individuals of any age including pediatric and children. Exposure to carcinogenic agents including ionizing radiation and family history is one of the main causes of the disease. Early diagnosis is crucial to avoid prolonged. patients' suffering. The aim of the study was to efficiently recognize the brain tumors from the other brain tissues which include grey and white matter as well as cerebrospinal fluid (CSF). Materials and methods This study was performed using axial, sagittal and coronal views for fifty brain tumor patients randomly selected from a set of 200 patients, with a “control” set consisting of images showing no sign of disease; and the “test” brain MRI images for patients diagnosed with brain tumor. The study includes both genders with age ranging from 18 years to 83 years old, (56.5 ± 17.2). The brain images were acquired using a standard head coil Philips Intera 1.5 Tesla machine (USA). The thickness of each section in the entire sequence was 8 mm. Acquisition of T2-weighted and T1-weighted were performed. Interactive Data Language software was used to analyze the data. Results The results of this study showed that: the overall accuracy of classification process was 94.8%, and for the tumor; the sensitivity was 97.3%. White matter and grey matter showed a classification accuracy of 95.7% and 89.7% and for CSF the accuracy was 94.3%. Conclusion The results showed that brain tumor can be classified successfully and delineated using texture analysis with minimum efforts and with high accuracy for brain tumors.

Numerous methods have been published to segment the infarct tissue in the left ventricle, most of them either need manual work, post-processing, or suffer from poor reproducibility. We proposed an automatic segmentation method for segmenting the infarct tissue in left ventricle with myocardial infarction. Cardiac images of a total of 60 diseased hearts (55 human hearts and 5 porcine hearts) were used in this study. The epicardial and endocardial boundaries of the ventricles in every 2D slice of the cardiac magnetic resonance with late gadolinium enhancement images were manually segmented. The subsequent pipeline of infarct tissue segmentation is fully automatic. The segmentation results with the automatic algorithm proposed in this paper were compared to the consensus ground truth. The median of Dice overlap between our automatic method and the consensus ground truth is 0.79. We also compared the automatic method with the consensus ground truth using different image sources from different centers with different scan parameters and different scan machines. The results showed that the Dice overlap with the public dataset was 0.83, and the overall Dice overlap was 0.79. The results show that our method is robust with respect to different MRI image sources, which were scanned by different centers with different image collection parameters. The segmentation accuracy we obtained is comparable to or better than that of the conventional semi-automatic methods. Our segmentation method may be useful for processing large amount of dataset in clinic.

Alzheimer’s disease (AD) is an irreversible, progressive brain disorder that slowly destroys memory and thinking skills. It is one of the leading types of dementia for persons aged above 65 worldwide. In order to achieve accurate and timely diagnosis, and for detection of AD in its early stages, numerous Artificial Intelligence (AI) based Computer-aided Diagnostic (CAD) approaches have been proposed using data from brain imaging. In this paper, we review the recent application of AI based CAD systems on AD and its stages, with a particular focus on the use of structural MRI due to its cost effectiveness and lack of ionizing radiation. We will review important factors of different AI techniques pertinent to AD, summarize contributions from different research groups, critically discuss challenges involved and propose directions for future research. Ultimately, it would be ideal for development of a diagnostic framework that could be applicable to not only AD, but to different types of dementia as well in the future.

MRI technique is widely used in the field of medicine because of its high spatial resolution, non-invasive characteristics, and soft tissue contrast. In this review article, a systematic study has been conducted to analyze the performance and issues of various techniques for brain tumor segmentation. Latest research on BTS in MRI with the higher resolution is utilized for the systematic review. The high-resolution images increase execution time of the classification, and accuracy is the other problem in BTS. Still, there is some research lacking in accuracy on the brain segmentation. Few researchers carried out the classification of different kinds of tissues in the brain images and also on the prediction on growth of tumor. Each method has specific technique to improve the performance of the BTS, and these methods are compared with one another in terms of result. Research comparison helps to understand the proposed method with their achieved results. Clustering algorithms such as K-means and FCM are generally used for segmentation, and GA, ANN, ANFIS, FCNN, SVM are commonly used as classifiers.

Tumor is basically a most common disease of brain and the Brain Tumor (BT) treatment has crucial significance. A diagnostic procedure called MRI image that is employed for detecting BT. It is the utmost important and intricate tasks in numerous medical-image applications since it typically involves a huge quantity of data. A lot of methods were applied in BT detection ranging as of image processing to examine the BT; however, the prevailing BT technique is tedious and less effective. So, this paper proposed the detection of the BT in MRI images utilizing optimized ANFIS classifier. Originally, the input MR image is preprocessed utilizing Gaussian Filter (GF) that removes the noise from the inputted image, additionally, the non-brain tissues (NBT) are removed using the technique of skull stripping (SS). After that, segmentation is performed wherein the tumor part is segmented utilizing CBAC technique and edema part is segmented utilizing HLSS segmentation technique. Then, GLCM in addition to GLRLM features are extracted afterward that extorted features is chosen by BFO algorithm. Finally, the selected features inputted to the optimized ANFIS classifier that classifies the tumor class types as Meningioma, Glioma, along with Pituitary. In ANFIS, the optimization procedure is achieved utilizing the PSO. The proposed system’s performance is contrasted to the prevailing systems regarding precision, recall, specificity, sensitivity, accuracy, together with F-Measure.

Medical imaging plays a crucial role in correct extraction of the significant information for monitoring the patient’s health and providing the quality treatment. A deluge of medical images requires initial interpretation for the presence of any abnormality, however, the correct diagnosis requires the images to be of good quality. To cope with the problem of poor contrast in medical images, this paper presents a method based on morphological transforms to improve the quality of the images. The proposed method incorporates Particle Swarm Optimization to find an optimum value of a parameter which controls the enhancement of the resulting image. The proposed algorithm is executed on a set of MRI images for testing its efficacy. The experimental results are compared in terms of both qualitative and quantitative parameters. The mean opinion score is obtained with the help of experts, which clearly shows the better performance of the proposed method. Furthermore, the parameters like Contrast Improvement Ratio, signal-to-noise ratio, peak signal-to-noise ratio, PL, and Structural Similarity Index are evident of better performance of proposed method when compared with the state-of-the-art methods and few recent methods. The comparison shows that the performance of the proposed method based on morphological transforms incorporating Particle Swarm Optimization is better not only visually but also in terms of other evaluation parameters.

Functional and anatomical information extraction from Magnetic Resonance Images (MRI) is important in medical image applications. The information extraction is highly influenced by the artifacts in the MRI images. The feature extraction involves the segmentation of MRI images. We present a MRI image segmentation using Bat Optimization Algorithm (BOA) with Fuzzy C Means (FCM) clustering. Echolocation of bats is utilized in Bat Optimization Algorithm. The proposed segmentation technique is evaluated with existing segmentation techniques. Results of experimentation shows that proposed segmentation technique outperforms existing methods and produces 98.5% better results.

Magnetic Resonance Imaging is an important tool for characterizing volumetric changes of the piglet brain during development. These analyses have been aided by the development of piglet brain atlases which are based on averages drawn from multiple piglets. Because these atlases typically contain only brain tissue, their use is limited to "brain extracted" images from which the surrounding tissues have been removed. Brain extractions, or segmentations, are typically performed manually. This approach is time-intensive and can lead to variation between segmentations when multiple raters are used. Automated segmentations processes are important for reducing the time required for analyses and improving the uniformity of the segmentations. Here we demonstrate the use of region-based recurrent convolutional neural networks (RCNNs) on a dataset consisting of 32 piglet brains. The RCNNs are trained from manual segmentations of sets of 27 piglets and then applied to sets of the remaining 5 piglets. The volumes of the machine-generated brain masks are highly correlated with those of the manually generated masks, and visual inspection of the segmentations show acceptable accuracy. These results demonstrate that neural networks provide a viable tool for the segmentation of piglet brains.

The anterolateral ligament is a rotational stabilizer of the knee joint. It is not always clear what we actually see on MRI in the area of anterolateral ligament (ALL).The aim of the study was to evaluate the ALL variants on MRI images to summarize their common features and differences, and to try to find an explanation for the phenomenon of the ALL variability.200 series of MRI images of knee joints were analyzed. The presence of the ALL, the number of its layers, the relation to the joint capsule, and other anatomical features were assessed.The ALL was visualized on MRI at least partially in 88 % of cases. At least partially two-layer structure was detected in 68 % of all 200 MRI series. The wavy appearance of the certain portions of the anterolateral ligament was observed in some normal knee joints without a history of injuries.Determined that the ALL is a separate anatomical element of the knee joint that has a variable, but in most cases two-layered, anatomical structure and can be detected on MRI in at least 88 % of cases. Axial sections help to identify ALL in complex cases and allow analyzing its anatomy, but adding little in the diagnosis of ALL injury.

In the analysis of medical images, one of the challenging tasks is the recognition of brain tumours via medical resonance images (MRIs). The diagnosis process is still tedious due to its complexity and considerable variety in tissues of tumor perception. Therefore, the necessities of tumor identification techniques are improving nowadays for medical applications. In the past decades, different approaches in the segmentation of various precisions and complexity degree have been accomplished, which depends on the simplicity and the benchmark of the technique. An overview of this analysis is to give out the summary of the semi-automatic techniques for brain tumor segmentation and classification utilizing MRI. An enormous amount of MRI based image data is accomplished using deep learning approaches. There are several works, dealing on the conventional approaches for MRI-based segmentation of brain tumor. Alternatively, in this review, we revealed the latest trends in the methods of deep learning. Initially, we explain the several threads in MRI pre-processing, including registration of image, rectification of bias field, and non-brain tissue dismissal. And terminally, the present state evaluation of algorithm is offered and forecasting the growths to systematise the MRI-based brain tumor into a regular cyclic routine in the clinical field are focussed.

Quantification of tau load using 11C-PBB3-PET has the potential to improve diagnosis of neurodegenerative diseases. Although MRI-based pre-processing is used as a reference method, not all patients have MRI. The feasibility of a PET-based pre-processing for the quantification of 11C-PBB3 tracer was evaluated and compared with the MRI-based method.Fourteen patients with decreased recent memory were examined with 11C-PBB3-PET and MRI. The PET scans were visually assessed and rated as either PBB3(+) or PBB3(-). The image processing based on the PET-based method was validated against the MRI-based approach. The regional uptakes were quantified using the Mesial-temporal/Temporoparietal/Rest of neocortex (MeTeR) regions. SUVR values were calculated by normalizing to the cerebellar reference region to compare both methods within the patient groups.Significant correlations were observed between the SUVRs of the MRI-based and the PET-based methods in the MeTeR regions (rMe=0.91; rTe=0.98; rR=0.96; p

This paper proposes a new unsupervised fuzzy feature mapping method based on fMRI data and combines it with multi-view support vector machine to construct a classification model for computer-aided diagnosis of autism. Firstly, a multi-output TSK fuzzy system is adopted to map the original feature data to the linear separable high-dimensional space. Then a manifold regularization learning framework is introduced, and a new method of unsupervised fuzzy feature learning is proposed. Finally, a multi-view SVM algorithm is used for classification tasks. The experimental results show that the method in this paper can effectively extract important features from the fMRI data in the resting state and improve the model's interpretability on the premise of ensuring the superior and stable classification performance of the model.

Orthogonal polar image moments which are defined over a unit disk, are used often in watermarking applications. One such image moment is polar harmonic Fourier transformation (PHFT). The quality of the extracted watermark depends on the accuracy of the computed image moments. Various methods have already been used for the computation of the moments with varying degrees of accuracy. This study proposes application of a novel image moments computation method which is a synergy of analytic, and numerical techniques, to image watermarking. The computed moment results in a robust watermark extraction. The PHFT have rotation, and scaling invariance property, which allows the watermark to resist various geometric attacks. In addition to the geometric attacks the watermark has the capability to resist various signal processing attacks. The performance of the proposed watermarking scheme is performed with the methods used in recently published articles. The experimental results show that the proposed method has the better performance compared to other techniques, and has better capability to resist various attacks.

Many patients are dying day after day due to the hypothetical treatment of brain tumors. Brain tumor detection and location identification are very ticklish and time-consuming task in biomedical image processing due to the complex structure of the brain. It is an important factor for detecting the tumor at the early stage that prevents the death from this epidemic disease. One of the most used and popular for the brain tumor diagnosis system is Magnetic Resonance Imaging (MRI) but it is a manual and time-consuming approach. In this research, an automated system has proposed to detect, locate and observe the tumor area in the patient brain. In the tumor extraction stage, the morphological operation based and marker control watershed segmentation methods used to extract the affected region from brain hemisphere through some additional step like filtrations and noise removal. In the tumor location identification stage, the brain hemisphere divided into four parts and compared among them by counting the pixels value that gave the proper location of the tumor in the hemisphere of the brain. Finally, this research study proposed a patient observation technique that compared the tumor growth of patient in timely.

Background: Image segmentation plays a vital role in clinical study and in diagnosing the stages of diseases. The hospitals are marching towards filmless imaging process in par with digital film technology. Materials and Methods: The paper proposes a novel method based on morphometric segmentation algorithm with Region of Interest (ROI). The core objective is to detect better segmentation of a brain tumor from an MRI image for a clinical study. Results and Conclusion: A set of experiment has been performed for the analysis of the proposed work on several MRI images. The proposed morphometric segmentation method provides better accuracy on segmentation of tumor image from the input MRI image.

Medical image fusion has been carried out to obtain information benefits from multi-modalities of medical images. The purpose of this study is to improve the image contrast of fusion image with adaptive method. The median filter was implemented to the images before registration to remove noise for obtaining good image fusion. Geometric transformation-based image registration was used to automatically align two images of computed tomography (CT) scanner and magnetic resonance imaging (MRI) to a common coordinate system. After that, the image contrast was improved with adaptive method. Finally, the fused image was assessed using the signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR). From this study, it was found that the average SNR value in the image fusion before contrast improvement is 0.09 and after that is 0.73. While the average CNR value in image fusion before contrast improvement is 1.54 and after that is 1.79. It means that the CNR increases 14.02%.

This article was to explore the application of deep learning algorithms in the classification and diagnosis of deep pelvic endometriosis (DPE) by ultrasound and magnetic resonance imaging (MRI). Vaginal ultrasound (VU) and MRI images of 118 patients with DPE and 206 patients with other gynaecological diseases who were treated in our hospital from January 2015 to January 2018 were analysed. Firstly, the global average pooling (GAP) was introduced based on the visual geometry group (VGG) network to design the VGG-GAP model for VU image recognition. Next, the C3D model was improved as the IC3D model for MRI image recognition. The diagnostic values of patients with DPE based on classified VU and MRI images were compared and analysed. The results revealed that the classification accuracy of the VGG-GAP model in the VU image was 96.5%, and the classification accuracy of the IC3D model in the MRI image was 99.2%. The diagnosis accuracy using VU images and MRI images was 90.68% and 92.37%, respectively. Based on this, the classification of VU or MRI images based on deep learning algorithms could provide a basis for improving the diagnosis efficiency of DPE. The value of MRI in the diagnosis of DPE was significantly higher than that of VU.

Objective: To analyze the accuracy of DWI-MRI combined with conventional MRI sequence in the diagnosis and classification of anal fistula, and to explore its application value. Methods: The clinical and imaging data of 56 patients with anal fistula confirmed by surgery were collected. All patients underwent conventional MR scans (axial T1WI, T2WI and axial, coronary, sagittal T2WI fat suppression) and axial DWI. With the surgical pathology results as the gold standard, the diagnostic compliance rate of anal fistula inside and outside mouth, main fistula, branch fistula, and abscess were observed and relied on both conventional sequence and conventional sequence combined with DWI. At the same time, the patients in this group were classified according to Parks classification of anal fistula, and the results were compared with the surgical results. Results: Fifty-six patients with anal fistula were confirmed to have 7 simple anal fistulas and 49 complicated anal fistulas. According to Parks anal fistula classification, MRI diagnosed 7 cases of intersphincteric fistula, 24 cases of trans sphincteric type fistula, 16 cases of superior sphincteric type fistula, and included 9 cases of rectal vaginal/scrotal fistula into the scope of extra-sphincteric fistula. The overall coincidence rate of scores was 87.5% (49/56). Conclusion: In preoperative MRI examination of anal fistula, compared with relying only on conventional sequences, the combined use of DWI sequences has higher accuracy in displaying the inner, outer and branch fistulas, and provides more detailed images before clinical operation. Anatomical information should be applied as a routine examination sequence for preoperative diagnosis of anal fistula.

In this manuscript, the authors have studied the Disc Displacement (DD) status of patients with acute Closed Lock (CL) to determine differences between DD with Reduction (DDwR) and DD without Reduction (DDwoR). Among the acute CL patients who visited our hospital within 2 weeks of the onset of CL, we studied 10 patients whose CL was released (DDwR) and 13 patients whose CL was not released (DDwoR). The DDwoR group was significantly older than the DDwR group. Although the mouth opening distance was significantly greater in the DDwoR group than in the DDwR group, the two groups were identical in the duration of CL. Sagittal MRI images showed no significant differences between the two groups in disc length and disc thickness (anterior band, intermediate zone, and posterior band). Multisection sagittal and coronal images identified lateral DD in 7 of the 10 patients in the DDwR group, although no specific direction of the DD was observed in the DDwoR group. Furthermore, deformation of the medial disc was common in the DDwoR group but uncommon in the DDwR group.

Objective: To discuss the diagnostic value of X-ray film and MRI of knee osteoarthritis by using support vector machine algorithm. Materials and Methods: The data of patients with osteoarthritis of the knee joint confirmed by open or arthroscopy surgery were collected in this paper. A total of 23 knee joints were performed in 23 cases, and plain films and MRI were performed before surgery. The support vector machine algorithm was used to retrospectively analyze its image performance and compare the positive rates of the two inspection methods. Results: The plain bone film showed better bone changes, and it could not show soft tissue lesions. MRI showed less changes in bone quality than plain films, and showed better changes in cartilage, meniscus degeneration, joint effusion, intra-articular free body, synovial thickening, bone marrow edema, and soft tissue swelling. Conclusion: MRI provides more positive signs than plain film, and the diagnosis accuracy of knee osteoarthritis is higher than plain film.

Objective: To study the effect of bone marrow platelet-rich plasma (PRP) injection on osteoarthritis using a level set-based MRI image segmentation algorithm. Methods: 180 patients with knee osteoarthritis (206 knees) were randomly divided into observation group (92 knees) and control group (88 knees). The observation group was injected with 2 mL of enriched stem cells and PRP suspension in the joint cavity, and then injected once again after 6 weeks. The control group was injected with 2 mL of sodium hyaluronate in the joint cavity, once a week for a total of 5 times. After 3 months and 6 months of treatment, the patients were followed up, and the clinical efficacy of the two groups was evaluated by the WOMAC score. The MRI image segmentation of the patients was analyzed using the level set-based MRI image segmentation method. Results: The WOMAC score observation group was 45.88 ± 9.54 points before treatment, 25.26 ± 6.67 points 3 months after treatment, 22.44 ± 5.19 points 6 months after treatment; the control group was 46.76 ± 8.06 points before treatment and 32.12 ± 5.35 3 months after treatment. Points, 33.34 ± 6.32 points 6 months after treatment. The WOMAC score of the observation group and the control group was improved after treatment, and the difference was statistically significant (P < 0.05) compared with that before treatment. Conclusion: The observation group has a clear effect on the treatment of knee osteoarthritis and is worthy of clinical application.

Objective: The paper applies spatial fuzzy clustering algorithm to explore the role and value of neuroendoscopic assisted technology in the operation of tumors in the saddle region, and analyze the MRI image characteristics of tumors in the saddle region. Methods: The clinical data of 63 patients from our hospital who underwent neuroendoscopic assisted microscopy to remove tumors in the saddle area from 2017 to 2019 (neuroendoscopy-assisted group) were collected. Seventy six patients who occupied the saddle area by microscopic resection only in the same period (Simple microscope group) clinical data. By comparing the patient’s tumor resection rate, postoperative complication rate and postoperative recurrence rate, the surgical effect was evaluated. Results: The total resection rates of the tumors in the neuroendoscopy-assisted group and the microscope-only group were 95.24% (60/63) and 80.26% (61/76). The incidence of postoperative vasospasm was 3.17% (2/63) and 13.16% (10/76), the incidence of nerve injury was 0 (0/63) and 6.58% (5/76), the difference was statistically significant (P 0.05). Conclusion: Neuroendoscopy-assisted microscopy-based removal of the saddle area occupying space based on spatial fuzzy clustering algorithm can increase the total tumor resection rate and reduce the incidence of complications. doi: https://doi.org/10.12669/pjms.37.6-WIT.4850 How to cite this:Zhang P, Zhang L, Zhao R. Application of MRI images based on Spatial Fuzzy Clustering Algorithm guided by Neuroendoscopy in the treatment of Tumors in the Saddle Region. Pak J Med Sci. 2021;37(6):1600-1604. doi: https://doi.org/10.12669/pjms.37.6-WIT.4850 This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Introduction: During deep brain stimulation (DBS) surgery, computed tomography (CT) and magnetic resonance imaging (MRI) scans need to be co-registered or fused. Image fusion is associated with the error that can distort the location of anatomical structures. Co-registration in DBS surgery is usually performed automatically by proprietary software; the amount of error during this process is not well understood. Here, our goal is to quantify the error during automated image co-registration with FrameLink™, a commonly used software for DBS planning and clinical research. Methods: This is a single-center retrospective study at a quaternary care referral center, comparing CT and MR imaging co-registration for a consecutive series of patients over a 12-month period. We collected CT images and MRI scans for 22 patients with Parkinson’s disease requiring placement of DBS. Anatomical landmarks were located on CT images and MRI scans using a novel image analysis algorithm that included a method for capturing the potential error inherent in the image standardization step of the analysis. The distance between the anatomical landmarks was measured, and the error was found by averaging the distances across all patients. Results: The average error during co-registration was 1.25 mm. This error was significantly larger than the error resulting from image standardization (0.19 mm) and was worse in the anterior-posterior direction. Conclusions: The image fusion errors found in this analysis were nontrivial. Although the estimated error may be inflated, it is sig­nificant enough that users must be aware of this potential inaccuracy, and developers of proprietary software should provide details about the magnitude and direction of co-registration errors.