Magnetic Resonance Imaging Classification Research Articles

MRI Image-Based Automatic Segmentation and Classification of Brain Tumor and Swelling Using Novel Methodologies

In the medical image analysis field, brain tumors (BTs) classification is a complicated process. For effortlessly detecting the tumor devoid of any surgical interference, the radiologists are aided with automated along with computerized technology. Currently, in the field of medical image processing along with analysis, admirable progress has been made by deep learning (DL) methodologies. In medical fields, for resolving several issues, huge attention was paid to DL techniques. For automation of several performed by radiologists like (1) lesion detection, (2) segmentation, (3) classification, (4) monitoring, along with (5) also prediction of treatment response that is not achievable without software, DL might be wielded. Nevertheless, classifying BTs by utilizing magnetic resonance imaging (MRI) has various complications like the difficulty of brain structure along with the intertwining of tissues in it; additionally, the brain’s higher density nature also makes the BT Classification (BTC) process quite complex. Therefore, by utilizing novel systems, MRI-centric Automatic segmentation together with classifications of BT and swelling have been proposed to overcome the aforementioned issues. The proposed methodology underwent various operations to detect BTs effectively. Initially, by utilizing the Range-centric Otsu’s Thresholding (ROT) algorithm, the skull stripping (SS) is conducted. After that, by performing contrast enhancement (CE) along with noise removal, the skull-stripped images are pre-processed. Next, by employing the Rectilinear Watershed Segmentation (RWS) algorithm, the tumor or swelling areas are segmented. Afterward, to obtain the precise tumor or swelling region, the morphological operations are executed on the segmented areas; subsequently, the desired along with relevant features are extracted. Lastly, the features being extracted are inputted to the classifier termed Uniform Convolution neural network (UCNN). The tumor tissues along with the swelling tissues are classified precisely in the classification phase. Here, the openly accessible BT Image Segmentation Benchmark (BRATS) datasets are utilized. Then, the outcomes obtained are analogized with prevailing methodologies. The experiential outcomes displayed that the BTC is performed by the proposed model with a higher accuracy rate; thus, outshined the other prevailing models.

An MRI image automatic diagnosis model for lumbar disc herniation using semi-supervised learning

Lumbar disc herniation is a common disease that causes low back pain. Due to the high cost of medical diagnosis, as well as a shortage and uneven distribution of medical resources, a system that can automatically analyze and diagnose lumbar spine Magnetic Resonance Imaging (MRI) is becoming an urgent need. This study uses deep learning methods to establish a classifier to diagnose lumbar disc herniation. An MRI classification dataset of lumbar disc herniation consisting of public MRI images is presented and is used to train the proposed classifier. Because a common difficulty in applying computer vision technology to medical images is labeling training data, we use a semi-supervised model training method, while multilayer transverse axial MRI images are used as the model input. In this method, we first use unlabelled MRI images for random self-supervised pre-training and the pre-trained model as a feature extractor for MRI images. Then, all marked cross-sections of each intervertebral disc are used to calculate the feature vector through the feature extractor. The information of all feature vectors is integrated, while a multilayer perceptron is used for classification training. After training, the model achieved 87.11% accuracy, 87.50% sensitivity, 86.72% specificity and 0.9487 AUC (Area Under the ROC Curve) index on the test set. To analyze the rationality of the diagnostic results more quickly, we output the severity of degenerative changes in each region using a heatmap.

Expression of GADD45G and CAPRIN1 in Human Nucleus Pulposus: Implications for Intervertebral Disc Degeneration.

Marked cellular changes occur in human intervertebral disc (IVD) degeneration during disc degeneration with biochemical changes. Genome-wide analysis of the DNA methylation profile has identified 220 differentially methylated loci associated with human IVD degeneration. Among these, two cell-cycle-associated genes, growth arrest and DNA damage 45 gamma (GADD45G) and cytoplasmic activation/proliferation-associated protein-1 (CAPRIN1), were focused on. The expression of GADD45G and CAPRIN1 in human IVDs remains unknown. We aimed to examine the expression of GADD45G and CAPRIN1 in human nucleus pulposus (NP) cells and evaluate those in human NP tissues in the early and advanced stages of degeneration according to Pfirrmann magnetic resonance imaging (MRI) and histological classifications. Human NP cells were cultured as monolayers after isolation from NP tissues by sequential enzyme digestion. Total RNA was isolated, and the mRNA expression of GADD45G and CAPRIN1 was quantified using real-time polymerase chain reaction. To examine the effects of pro-inflammatory cytokines on mRNA expression, human NP cells were cultured in the presence of IL-1β. Protein expression was evaluated using Western blotting and immunohistochemistry. GADD45G and CAPRIN1 expression was identified in human NP cells at both mRNA and protein levels. The percentage of cells immunopositive for GADD45G and CAPRIN1 significantly increased according to the Pfirrmann grade. A significant correlation between the histological degeneration score and the percentage of GADD45G-immunopositive cells was identified, but not with that of CAPRIN1-immunopositive cells. The expression of cell-cycle-associated proteins (GADD45G and CAPRIN1) was enhanced in human NP cells at an advanced stage of degeneration, suggesting that it may be regulated during the progression of IVD degeneration to maintain the integrity of human NP tissues by controlling cell proliferation and apoptosis under epigenetic alteration.

Paper 04: Differences in the Severity and Location of Patellofemoral Cartilage Damage Between Instability Patients With and Without Patella Alta

Objectives: Patellar instability is estimated to affect approximately 150 out of 100,000 adolescents between the age of 14-18 years old, a rate which is expected to increase over time. Several risk factors have been associated with patellar instability such as younger age, trochlear dysplasia and patella alta. Patella alta is defined as a superiorly-positioned patella, commonly determined using the Caton-Deschamps Index (CDI), with a ratio of 1.3 or greater indicative of patella alta. Patients with patella alta are approximately 2.5-4.0 times more likely to suffer a recurrent patellar dislocation, which is thought to be due to the changes in biomechanics seen in patella alta. Differences in chondral injury in patients with patellar instability between patella alta and patella norma has not been evaluated. Our objective was to retrospectively analyze whether preoperative cartilage damage differs in severity and location between patellar instability patients with and without patella alta. Methods: Patients with patellar instability who underwent patellar realignment surgery at a single institution with a preoperative MRI within 3 months were included. After measurement of CDI on MRI, patients were divided into patella alta (CDI>1.3) and patella norma groups. To quantify cartilage defect severity, the Magnetic Resonance Imaging Score and Classification system (AMADEUS) was utilized by two sports medicine clinical fellows. The AMADEUS score characterizes cartilage defect size, defect depth, presence of subchondral bone defects, and bone marrow edema. The scale is scored 0-100 with 100 indicating no cartilage damage. In patients with multiple cartilage defects on MRI, each defect was characterized and analyzed independently. Continuous variable data were reported as means with standard deviations and categorical data reported as frequencies with percentages. The distribution of data was assessed with the Shapiro-Wilk test. Continuous data were assessed with the Student’s t test and Mann-Whitney U tests as appropriate. Categorical data were assessed with chi-squared analysis. The threshold for significance was set to p < 0.05. Multivariate analysis was conducted with defect depth, bone edema, and AMADEUS score as the dependent variable. All statistical analyses were performed with R Studio (Version 3.6.3, Vienna, Austria). Results: One hundred-twenty-one patients were divided into patella alta (n=50) and patella norma (n=71). The groups did not significantly differ in sex ratio, age at MRI, body mass index (BMI), mean reported number of prior dislocations, or mean interval between first reported dislocation and date of MRI. Thirty-four (68%) of the patella alta group and 44 (62%) of the patella norma group had chondral defects ( P=0.625) with no statistically significant difference in the size of the chondral defects between the groups ( P=0.419). In both groups, chondral injuries most affected the medial patellar facet (55% in patella alta vs. 52% in patella norma), followed by the lateral facet (25% vs 18%), and lateral femoral condyle (10% vs 14%)( P=0.859). The patella alta group had a smaller proportion of patients who had full thickness defects in comparison to the patella norma group (60% vs 82%; P=0.030). The overall AMADEUS score for the patella alta group was higher (68.9 vs. 62.1; P=0.023) compared to the patella norma group, indicating superior articular cartilage status. When looking at a multivariate regression using AMADEUS score as the dependent outcome, female sex shows an increase in AMADEUS score with a beta estimate of 11.96 ( P<0.001). Female sex also shows a higher probability of having no defect associated bone marrow edema (Odds ratio:6.7; P=0.010), and a higher probability of partial-thickness defect rather than full-thickness (Odds ratio:3.9; P=0.004). Age, BMI, interval between dislocation and MRI, and reported number of dislocations were not found to be associated with total AMADEUS score, probability of having defect-associated bone marrow edema, or defect depth. Conclusions: There is a high percentage of cartilage injury in patients with a history of patellar instability with over 60% of patients showing cartilage damage on MRI. Patients with normal patellar height have a higher frequency of full thickness cartilage defects and more severe injury compared to those with patella alta. The location of cartilage injury between alta and norma is similar, with most defects affecting the medial facet, lateral facet, and lateral femoral condyle in descending frequency.

Tumor Diagnosis against Other Brain Diseases Using T2 MRI Brain Images and CNN Binary Classifier and DWT.

Brain tumors are diagnosed and classified manually and noninvasively by radiologists using Magnetic Resonance Imaging (MRI) data. The risk of misdiagnosis may exist due to human factors such as lack of time, fatigue, and relatively low experience. Deep learning methods have become increasingly important in MRI classification. To improve diagnostic accuracy, researchers emphasize the need to develop Computer-Aided Diagnosis (CAD) computational diagnostics based on artificial intelligence (AI) systems by using deep learning methods such as convolutional neural networks (CNN) and improving the performance of CNN by combining it with other data analysis tools such as wavelet transform. In this study, a novel diagnostic framework based on CNN and DWT data analysis is developed for the diagnosis of glioma tumors in the brain, among other tumors and other diseases, with T2-SWI MRI scans. It is a binary CNN classifier that treats the disease "glioma tumor" as positive and the other pathologies as negative, resulting in a very unbalanced binary problem. The study includes a comparative analysis of a CNN trained with wavelet transform data of MRIs instead of their pixel intensity values in order to demonstrate the increased performance of the CNN and DWT analysis in diagnosing brain gliomas. The results of the proposed CNN architecture are also compared with a deep CNN pre-trained on VGG16 transfer learning network and with the SVM machine learning method using DWT knowledge. To improve the accuracy of the CNN classifier, the proposed CNN model uses as knowledge the spatial and temporal features extracted by converting the original MRI images to the frequency domain by performing Discrete Wavelet Transformation (DWT), instead of the traditionally used original scans in the form of pixel intensities. Moreover, no pre-processing was applied to the original images. The images used are MRIs of type T2-SWI sequences parallel to the axial plane. Firstly, a compression step is applied for each MRI scan applying DWT up to three levels of decomposition. These data are used to train a 2D CNN in order to classify the scans as showing glioma or not. The proposed CNN model is trained on MRI slices originated from 382 various male and female adult patients, showing healthy and pathological images from a selection of diseases (showing glioma, meningioma, pituitary, necrosis, edema, non-enchasing tumor, hemorrhagic foci, edema, ischemic changes, cystic areas, etc.). The images are provided by the database of the Medical Image Computing and Computer-Assisted Intervention (MICCAI) and the Ischemic Stroke Lesion Segmentation (ISLES) challenges on Brain Tumor Segmentation (BraTS) challenges 2016 and 2017, as well as by the numerous records kept in the public general hospital of Chania, Crete, "Saint George". The proposed frameworks are experimentally evaluated by examining MRI slices originating from 190 different patients (not included in the training set), of which 56% are showing gliomas by the longest two axes less than 2 cm and 44% are showing other pathological effects or healthy cases. Results show convincing performance when using as information the spatial and temporal features extracted by the original scans. With the proposed CNN model and with data in DWT format, we achieved the following statistic percentages: accuracy 0.97, sensitivity (recall) 1, specificity 0.93, precision 0.95, FNR 0, and FPR 0.07. These numbers are higher for this data format (respectively: accuracy by 6% higher, recall by 11%, specificity by 7%, precision by 5%, FNR by 0.1%, and FPR is the same) than it would be, had we used as input data the intensity values of the MRIs (instead of the DWT analysis of the MRIs). Additionally, our study showed that when our CNN takes into account the TL of the existing network VGG, the performance values are lower, as follows: accuracy 0.87, sensitivity (recall) 0.91, specificity 0.84, precision 0.86, FNR of 0.08, and FPR 0.14. The experimental results show the outperformance of the CNN, which is not based on transfer learning, but is using as information the MRI brain scans decomposed into DWT information instead of the pixel intensity of the original scans. The results are promising for the proposed CNN based on DWT knowledge to serve for binary diagnosis of glioma tumors among other tumors and diseases. Moreover, the SVM learning model using DWT data analysis performs with higher accuracy and sensitivity than using pixel values.

Research on classification algorithm of cerebral small vessel disease based on convolutional neural network

The investigation on brain magnetic resonance imaging (MRI) of cerebral small vessel disease (CSVD) classification algorithm based on deep learning is particularly important in medical image analyses and has not been reported. This paper proposes an MRI classification algorithm based on convolutional neural network (MRINet), for accurately classifying CSVD and improving the classification performance. The working method includes five main stages: fabricating dataset, designing network model, configuring the training options, training model and testing performance. The actual training and testing datasets of MRI of CSVD are fabricated, the MRINet model is designed for extracting more detailedly features, a smooth categorical-cross-entropy loss function and Adam optimization algorithm are adopted, and the appropriate training parameters are set. The network model is trained and tested in the fabricated datasets, and the classification performance of CSVD is fully investigated. Experimental results show that the loss and accuracy curves demonstrate the better classification performance in the training process. The confusion matrices confirm that the designed network model demonstrates the better classification results, especially for luminal infarction. The average classification accuracy of MRINet is up to 80.95% when classifying MRI of CSVD, which demonstrates the superior classification performance over others. This work provides a sound experimental foundation for further improving the classification accuracy and enhancing the actual application in medical image analyses.

APPLICATIONS OF THE EXTENSION INNOVATION METHOD IN TREATMENT REGIMENS GENERATION WITH DEEP LEARNING

Traditional treatment research lays much emphasis on therapeutic effects of a single drug or combination drugs, failing to explore appropriate treatment regimens that can be formed based on the state of an illness, medical history, financial capacity, and adverse drug reactions. For this reason, the extension innovation method is introduced to implement doctors’ thinking process in formulating treatment regimens in deep learning. First, an extension model is established for diseases according to the basic-element theory and the extension set theory based on the magnetic resonance image classification result under deep learning. Subsequently, extension analysis is made to analyze pathogenesis and corresponding treatment procedures; and multiple feasible treatment regimens are generated through extension transformation. At last, priority-degree evaluation is carried out to quantitatively assess the proposed treatment regimens and select a better regime from them. Here, ankylosing spondylitis (AS) is taken for example to validate the feasibility of applying the extension innovation method in treatment regime generation. Main contributions of this research are that both the extension model and the priority-degree evaluation method are introduced in treatment regime generation. In this way, after the deep learning method was used to automatically extract and classify medical image data features and determine the disease activity stage of ankylosing spondylitis, a formal and quantitative effective method can be provided for establishing the framework model of diseases and evaluating corresponding therapeutic regimens, then a foundation can be laid for arithmetic research on intelligent treatment regime generation via computer simulation of human thinking.

Heterogeneity of clinical symptoms and therapeutic strategies for different subtypes of adenomyosis: An initial single-center study in China.

To investigate the relationship between the magnetic resonance imaging (MRI) classification of different clinical symptoms and corresponding therapeutic efficacy in adenomyosis patients. From January 2015 to October 2020, a total of 468 patients diagnosed with adenomyosis through MRI examination at Peking University Third Hospital were included in this retrospective cohort study. Totals of 184 (39.3%), 208 (44.4%), 17 (3.6%), and 59 (12.6%) patients were categorized into Subtypes I (intrinsic), II (extrinsic), III (intramural), and IV (penetrating), respectively. Clinical information such as age, dysmenorrhea, menorrhagia, infertility, assisted reproduction, and drug treatment and its efficacy were analyzed. By comparing the clinical information of different adenomyosis subtypes, we intend to provide better fertility guidance and find better treatment strategies for these patients. The proportion of dysmenorrhea increased in intrinsic, extrinsic, intramural, and penetrating subtypes (74.5% vs 82.7% vs 94.1% vs 94.9%, respectively, P= 0.002). The proportion of menorrhagia in the intrinsic subtype (53.3%) was significantly higher than that in the extrinsic (28.4%) and intramural (29.4%) subtypes (P < 0.001). The effective rate of progesterone in the intrinsic subtype was significantly lower than that in the extrinsic subtype (52.0% vs 86.5%, P < 0.001). The infertility rates of adenomyosis patients with different subtypes were relatively high (17.6%-41.3%), and that of the extrinsic subtype was the highest among all the subtype groups (41.3%, P < 0.001). Significant differences in age, dysmenorrhea, menorrhagia, and infertility were found among patients with different subtypes of adenomyosis. A novel classification of adenomyosis was proposed to provide a theoretical basis for the treatment of adenomyosis patients with infertility.

Clinical Applicability of the Sellar Barrier Concept in Patients with Pituitary Apoplexy: Is It Possible?

There is evidence of association between sellar barrier thickness and intraoperative cerebrospinal fluid (CSF) leakage, impacting the postoperative prognosis of the patients. The aim of this study is to analyze the clinical applicability of the sellar barrier concept in a series of operated patients with pituitary apoplexy (PA). A retrospective study was conducted including 47 patients diagnosed with PA who underwent surgical treatment through a transsphenoidal approach. Brain magnetic resonance imaging (MRI) of the patients were evaluated and classified utilizing the following criteria: strong barrier (greater than 1 mm), weak barrier (less than 1 mm), and mixed barrier (less than 1 mm in one area and greater than 1 mm in another). The association between sellar barrier types and CSF leakage was analyzed, both pre- and intraoperatively. The preoperative MRI classification identified 10 (21.28%) patients presenting a weak sellar barrier, 20 patients (42.55%) with a mixed sellar barrier, and 17 patients (36.17%) exhibiting a strong sellar barrier. Preoperative weak and strong sellar barrier subtypes were associated with weak (p ≤ 0.001) and strong (p = 0.009) intraoperative sellar barriers, respectively. Strong intraoperative sellar barrier subtypes reduced the odds of CSF leakage by 86% (p = 0.01). A correlation between preoperative imaging and intraoperative findings in the setting of pituitary apoplexy has been observed.

Does MR Signal Intensity Change have Prognostic Value in Multilevel Cervical Myelopathy? A Single-Center, Prospective Study.

Neurological recovery in cervical myelopathy remains unpredictable. There is contradictory literature regarding the prognostic value of magnetic resonance imaging (MRI) in such cases. The objective of the present study is to evaluate the morphological changes in the spinal cord in cervical spondylotic myelopathy and compare them with clinical outcome. This is a single-center, prospective, observational study. All patients with multilevel (two or more levels) cervical spondylotic myelopathy undergoing anterior spine surgery were included in the study. Patient demographics and radiological findings were recorded. MRI was repeated immediately post-op and at 1-year follow-up. MRI classification system based on axial images was used to evaluate presurgery and postsurgery changes and correlate them with clinical information. The study comprised 50 patients (40 males and 10 females) with a mean age of 59.5 years. Average duration of symptoms before surgery was 6.29 months. Thirty-four patients underwent two-level decompression, while 16 patients underwent more than two-level decompression. Average duration of follow-up was 26.82 months. Mean pre-op Nurick grade was 2.84, and mean recovery rate was 56.73. Most common pre-op MRI type was type 1. Analysis of data by logistic regression showed better recovery rate with lower age, lower pre-op Nurick grade, and lower pre-op MRI type. MR classification based on signal intensity changes in axial images have been found to correlate with recovery rate.

SSCLNet: A Self-Supervised Contrastive Loss-Based Pre-Trained Network for Brain MRI Classification

Brain magnetic resonance images (MRI) convey vital information for making diagnostic decisions and are widely used to detect brain tumors. This research proposes a self-supervised pre-training method based on feature representation learning through contrastive loss applied to unlabeled data. Self-supervised learning aims to understand vital features using the raw input, which is helpful since labeled data is scarce and expensive. For the contrastive loss-based pre-training, data augmentation is applied to the dataset, and positive and negative instance pairs are fed into a deep learning model for feature learning. Subsequently, the features are passed through a neural network model to maximize similarity and contrastive learning of the instances. This pre-trained model serves as an encoder for supervised training and then the classification of MRI images. Our results show that self-supervised pre-training with contrastive loss performs better than random or ImageNet initialization. We also show that contrastive learning performs better when the diversity of images in the pre-training dataset is more. We have taken three differently sized ResNet models as the base models. Further, experiments were also conducted to study the effect of changing the augmentation types for generating positive and negative samples for self-supervised training.

Secure Multi-Party Computation for Magnetic Resonance Imaging Classification

Accessing private data always requires a complex negotiation process. This process becomes even more challenging under privacy regulations such as the General Data Protection Regulation and the California Consumer Privacy Act. However, in most cases, the availability of greater amounts of data leads to significant technological breakthroughs. The perfect example is the ImageNet classification challenge, which leads to significant improvements in the image recognition area. The combination of these concerns raises the question of performing computations on data that cannot be seen, and a whole new research field that consolidates privacy-preserving concepts and modern data mining techniques comes into place. This is also the purpose of the work presented in this paper, the discovery of Secure Multi-Party Computation (SMPC) capabilities on protecting privacy during the machine learning process. The main attention is paid towards the combination of SMPC and image classification approach based on the convolutional neural network, especially the secure inference process on the encrypted magnetic resonance images.

Analysis of Deep Learning Techniques for Brain Tumour Classification from CT &amp; MRI Images

Brain tumour detection in an initialpoint is a critical step to saving human life. Computed Tomography (CT) and Magnetic Resonance Image (MRI) provide very detailed information about brain tumour tissues. So the segmentation of tumour region is possible from pancreatic CT and brain MRI. CT and MRI is a non-invasive technique and it does not produce any harmful radiation to the patient. The patient suspected of tumour undergoes radiological evaluation such that the area, location and grade of the tumour can be predicted from the CT and MRI analysis. This critical information helps the doctors to decide about further treatment like chemotherapy, surgery, or radiation. The diagnosis requires an accurate and very fast segmentation and classification of CT and MRI images. But nowadays radiologists are doing this task manually and it is a tedious and time-consuming procedure. Also, there is a chance of variation in the result from one expert to another. Here comes the significance of automatic segmentation and classification of tumour types with the help of computers.The proposed work aims to develop an efficient system that can detect pancreatic and brain tumour and can classify the pancreatic CT and brain MRI into normal, benign or malignant. This work can be categorized into two approaches. Thus the dataset prepared for this research work contains CT and MRI images.The first approach proposes traditional machine learning technologies to achieve the goal. Image pre-processing, feature extraction, segmentation and classification are the various steps of the traditional machine learning method. A detailed investigation is performed through various feature extraction techniques and classification techniques for pancreatic (CT) and brain MRI. Discrete Wavelet Transform (DWT) feature, Grey Level Co-occurrence Matrix (GLCM) feature, Gabor feature, Tamura features and Edge Orientation Histogram (EOH) features and their combinations are used for the extraction of CT and MRI features. Benign tumours are non-cancerous, but malignant tumours are cancerous. In the first approach, the Support Vector Machine (SVM) is the main classifier used for pancreatic CT and brain MRI classification as normal, benign or malignant.In this technology, a huge amount of data and machines with high computational capabilities like Graphic Processing Unit (GPU) are available. Thus the second approach of this paper is to exploit all these available resources to produce accurate results. In this part, deep learning, the latest fast growing technology introduced in 2015 is used for the classification of brain MRI. A Deep Convolutional Neural Networks (DCNN) model is proposed to perform the classification task efficiently. The CNN results are compared with the results of a simple neural network classifier. This method provides accurate and it shows that deep learning based classification outperforms traditional machine learning techniques which produce an accurate result only. This research work again concentrates on the Transfer Learning (TL) methods to classify pancreatic CT and brain MRI.

Diagnosis and treatment of 10 cases of idiopathic chondrolysis of the hip.

To analyze the clinical characteristics, X-ray and magnetic resonance imaging manifestations, and treatment efficacy of idiopathic chondrolysis of the hip. Ten patients with idiopathic chondrolysis of the hip treated at our hospital from September 2013 to April 2022 were collected, and their clinical features, X-ray and magnetic resonance imaging manifestations, and treatment outcomes were analyzed. Their main clinical features included single hip pain, claudication gait, and pelvic tilt, without specific clinical symptoms and signs. Laboratory tests such as blood analysis, erythrocyte sedimentation rate, C-reactive protein, rheumatoid factor, autoimmune test, coagulation function, interferon release test for tuberculosis infection, and tuberculosis antibody test were normal. In the late stage, the imaging shows degeneration or ankylosis of the joint. The disease progresses rapidly, and joint degeneration can occur in adolescence without effective treatment. Herein, seven patients were treated with recombinant human tumor necrosis factor receptor-antibody type II fusion protein. Among them, four (according to Camarnat magnetic resonance imaging classification, three were Stage I and one was Stage II) showed improved symptoms and function, while two (Stage III) had progressed. One patient (Stage I) who received recombinant human tumor necrosis factor receptor had no significant relief of symptoms, for whom anterior capsular release surgery significantly improved the hip joint motion. Of the four patients who underwent hip release surgery, three showed progression. Idiopathic chondrolysis of the hip has no specific clinical and laboratory tests and a high misdiagnosis rate. Thus, early magnetic resonance imaging is a critical reference for diagnosis. We recommend that patients with magnetic resonance imaging Stage II or earlier be actively treated with recombinant human tumor necrosis factor receptor and start treatment as early as possible. In the third stage, the treatment effect is mediocre, and the narrowed hip space is difficult to change.

Magnetic resonance imaging findings and the clinical characteristics of children with cerebral palsy at a public sector hospital in Gauteng Province, South Africa

Background. Cerebral palsy (CP) is a common cause of physical impairment in children. Brain magnetic resonance imaging (MRI) can define different neuropathological patterns of brain injury in CP. There are limited data available on MRI findings of children with CP in Africa. &#x0D; Objective. To describe the clinical characteristics, risk factors and MRI findings of children with CP attending a developmental clinic at a tertiary hospital in South Africa; and to assess possible associations between the clinical characteristics and pathogenic neuro-imaging patterns. &#x0D; Methods. This was a retrospective cross-sectional study. The cohort of 112 children was identified from the clinic’s REDcap database. Clinical information was obtained from existing medical records of the patients. Findings from brain MRI reports were classified according to the MRI classification system (MRICS) for CP. The MRI reports were rated independently by two study investigators. A descriptive analysis was conducted. &#x0D; Results. A total of 112 patient files and MRI brain reports were reviewed. Spastic CP was the most common type of CP (n=75%). The most common perinatal risk factors included prematurity (31%) and low birthweight (28%). Nineteen (17%) children acquired CP after the neonatal period. CP sub-type showed a significant association with functional motor impairment classified as per the gross motor function classification system (GMFCS), p&lt;0.001. Predominant grey matter injury (PGMI) was the most common pathogenic MRI pattern identified (30%). The radiological findings (per MRICS) had a significant association with both the CP sub-type (p&lt;0.005) and functional impairment according to the GMFCS (p&lt;0.001). &#x0D; Conclusion. Standardised classification of neuro-imaging findings can assist in defining the pathogenesis and clinical manifestations of CP.

Prediction of prostate cancer recurrence after radiation therapy using multiparametric magnetic resonance imaging and spectroscopy: assessment of prognostic factors on pretreatment imaging.

To assess whether data from pre-therapeutic multiparametric magnetic resonance imaging (mpMRI) combined with three-dimensional magnetic resonance spectroscopy (3D MRS) provide prognostic factors of biochemical relapse in patients with localized prostate cancer treated by external radiotherapy or brachytherapy. In our single institution observational retrospective study we included a cohort of 230 patients treated by external radiotherapy or brachytherapy who had an initial mpMRI with 3D MRS from January 2008 to December 2015 for newly diagnosed localized prostatic cancer, proven histologically. Three trained radiologists recorded tumor characteristics, MRI T-stage and metabolic abnormalities from 3D MRS data. Univariate and multivariate Cox analyzes explored the relationship between clinical and imaging variables to highlight prognostic factors for recurrence, using biochemical relapse as the primary endpoint. mpMRI data analysis allowed to reclassify 21.7% of the patients in a MRI National Comprehensive Cancer Network (NCCN) group higher than their initial clinical T-stage, but also to detect a lesion in 78% of the patients considered as clinically T1c. After a median of follow-up of 8.7 years (IQR, 6.6-10.1) following cancer diagnosis, 36 (16%) patients developed a biochemical relapse. The multivariate Cox analysis demonstrated the existence of 3 independent factors for prediction of biochemical recurrence: extracapsular extension (ECE) (HR =3.33; 95% CI: 1.93-5.73; P<0.01), choline/citrate ratio in healthy tissue in the transition zone (TZ) (HR =2.96; 95% CI: 1.06-8.28; P=0.04) and the NCCN Magnetic Resonance Imaging classification (intermediate versus low-risk, HR =3.06; 95% CI: 1.13-8.30; P<0.01). Combination of mpMRI and 3DMRS could aid in the prognostic stratification of localized prostate cancer treated by radiotherapy or brachytherapy, by combining accurate evaluation of tumor extension, and quantification of prostate metabolism.

Fetal heart rate patterns complicated by chorioamnionitis and subsequent cerebral palsy in Japan.

This retrospective study was performed to investigate whether certain fetal heart rate patterns were associated with subsequent cerebral palsy (CP) in infants with chorioamnionitis at or near term. We used cases registered by the Japan Obstetric Compensation System for CP, which is a nationwide population-based database. Among them, 133 infants with chorioamnionitis who were born at ≥34 weeks of gestation were enrolled. All infants underwent magnetic resonance imaging (MRI), and all fetal heart rate charts had been interpreted according to the National Institute of Child Health and Human Development criteria, focusing on antepartum and immediately before delivery. The incidence of CP after chorioamnionitis at ≥34 weeks of gestation was 0.3 per 10 000 in Japan. Between the clinical (24%) and subclinical groups (76%), the incidence of abnormal fetal heart rate patterns did not differ. According to the MRI classification, 88% of the infants with CP showed hypoxic-ischemic encephalopathy. Half of the infants with CP experienced terminal bradycardia, leading to severe acidosis and exclusively to hypoxic-ischemic encephalopathy. In another half, who did not experience bradycardia, 80% had moderate acidosis (pH7.00-7.20) resulting in hypoxic-ischemic encephalopathy, and the remaining 20% showed non-acidosis resulting in brain damage other than hypoxic-ischemic encephalopathy. The fetal heart rate patterns before the terminal bradycardia showed that the incidence rates of late deceleration or decreased variability were high (>60%). Fifty percent of pregnant women with chorioamnionitis-related CP had terminal bradycardia that exclusively resulted in hypoxic-ischemic encephalopathy.

Clinical Decision Support Framework for Segmentation and Classification of Brain Tumor MRIs Using a U-Net and DCNN Cascaded Learning Algorithm.

Brain tumors (BTs) are an uncommon but fatal kind of cancer. Therefore, the development of computer-aided diagnosis (CAD) systems for classifying brain tumors in magnetic resonance imaging (MRI) has been the subject of many research papers so far. However, research in this sector is still in its early stage. The ultimate goal of this research is to develop a lightweight effective implementation of the U-Net deep network for use in performing exact real-time segmentation. Moreover, a simplified deep convolutional neural network (DCNN) architecture for the BT classification is presented for automatic feature extraction and classification of the segmented regions of interest (ROIs). Five convolutional layers, rectified linear unit, normalization, and max-pooling layers make up the DCNN's proposed simplified architecture. The introduced method was verified on multimodal brain tumor segmentation (BRATS 2015) datasets. Our experimental results on BRATS 2015 acquired Dice similarity coefficient (DSC) scores, sensitivity, and classification accuracy of 88.8%, 89.4%, and 88.6% for high-grade gliomas. When it comes to segmenting BRATS 2015 BT images, the performance of our proposed CAD framework is on par with existing state-of-the-art methods. However, the accuracy achieved in this study for the classification of BT images has improved upon the accuracy reported in prior studies. Image classification accuracy for BRATS 2015 BT has been improved from 88% to 88.6%.

An Enhanced Machine Learning Approach for Brain MRI Classification.

Magnetic Resonance Imaging (MRI) is a noninvasive technique used in medical imaging to diagnose a variety of disorders. The majority of previous systems performed well on MRI datasets with a small number of images, but their performance deteriorated when applied to large MRI datasets. Therefore, the objective is to develop a quick and trustworthy classification system that can sustain the best performance over a comprehensive MRI dataset. This paper presents a robust approach that has the ability to analyze and classify different types of brain diseases using MRI images. In this paper, global histogram equalization is utilized to remove unwanted details from the MRI images. After the picture has been enhanced, a symlet wavelet transform-based technique has been suggested that can extract the best features from the MRI images for feature extraction. On gray scale images, the suggested feature extraction approach is a compactly supported wavelet with the lowest asymmetry and the most vanishing moments for a given support width. Because the symlet wavelet can accommodate the orthogonal, biorthogonal, and reverse biorthogonal features of gray scale images, it delivers higher classification results. Following the extraction of the best feature, the linear discriminant analysis (LDA) is employed to minimize the feature space's dimensions. The model was trained and evaluated using logistic regression, and it correctly classified several types of brain illnesses based on MRI pictures. To illustrate the importance of the proposed strategy, a standard dataset from Harvard Medical School and the Open Access Series of Imaging Studies (OASIS), which encompasses 24 different brain disorders (including normal), is used. The proposed technique achieved the best classification accuracy of 96.6% when measured against current cutting-edge systems.

Integrating convolutional neural networks, kNN, and Bayesian optimization for efficient diagnosis of Alzheimer's disease in magnetic resonance images

Deep learning is attracting growing interest from biomedical engineering community. Researchers and clinicians are also increasingly interested in development of machine learning and pattern recognition systems used to diagnose Alzheimer's disease (AD). To enhance diagnostic power for AD, we propose an automatic system integrating convolutional neural networks (CNN) to extract deep traits from magnetic resonance image (MRI) with no prior assumption, a filtering technique to reduce number of features, and k nearest neighbors (kNN) algorithm to discriminate AD subjects from healthy control (HC) ones. The kNN is tuned by Bayesian optimization (BO) algorithm. The experimental outcomes support the hypothesis that our proposed integrative system can be effective at performing MRI classification: 94.96% ± 0.0486 accuracy, 92.05% ± 0.0746 sensitivity, and 96.62% ± 0.0350 specificity. The obtained result underscore the utility of the proposed system for screening AD as it improves accuracy compared to existing models validated on the same data set.