MRI In Diagnosis Research Articles

Two-Branch network for brain tumor segmentation using attention mechanism and super-resolution reconstruction

Accurate segmentation of brain tumor plays an important role in MRI diagnosis and treatment monitoring of brain tumor. However, the degree of lesions in each patient’s brain tumor region is usually inconsistent, with large structural differences, and brain tumor MR images are characterized by low contrast and blur, current deep learning algorithms often cannot achieve accurate segmentation. To address this problem, we propose a novel end-to-end brain tumor segmentation algorithm by integrating the improved 3D U-Net network and super-resolution image reconstruction into one framework. In addition, the coordinate attention module is embedded before the upsampling operation of the backbone network, which enhances the capture ability of local texture feature information and global location feature information. To demonstrate the segmentation results of the proposed algorithm in different brain tumor MR images, we have trained and evaluated the proposed algorithm on BraTS datasets, and compared with other deep learning algorithms by dice similarity scores. On the BraTS2021 dataset, the proposed algorithm achieves the dice similarity score of 89.61%, 88.30%, 91.05%, and the Hausdorff distance (95%) of 1.414 mm, 7.810 mm, 4.583 mm for the enhancing tumors, tumor cores and whole tumors, respectively. The experimental results illuminate that our method outperforms the baseline 3D U-Net method and yields good performance on different datasets. It indicated that it is robust to segmentation of brain tumor MR images with structures vary considerably.

Perinatal stroke and overlapping clinical presentation with HIE; studying practical diagnostic and management challenges through observational analysis of a single tertiary centre stroke patients

Background: While therapeutic hypothermia is accepted as a standard of care for improving neurological outcome for HIE patients, evidence of its benefits for neonatal stroke is lacking. Neurological presentations of HIE and neonatal stroke can be difficult to distinguish and indeed neonatal stroke can coexist with HIE. Our observational analysis of a single centre cohort of stroke infants with MRI diagnosis highlights the practical challenges for distinguishing between the two groups and reaching an early definitive diagnosis to inform appropriate treatment. Methods: A retrospective cohort study of term neonates >37 weeks gestation born between May 2011 and April 2020 born at a specialised NICU network in the East of England with a principal diagnosis of neonatal stroke were obtained from Badgernet data. Results: 81% of infants (13/16) had sentinel events around delivery. 50% (8/16) of infants were therapeutically cooled. Of the 8 cooled infants only 3 (37%) of cooled infants fulfilled both Toby A and B criteria. Infants who received therapeutic hypothermia were more likely to present with early onset seizures (5/8) than infants who were not cooled (1/8) (p=0.019) . 6 of the 8 non-cooled infants compared to 2of the 8 cooled infants (p=0.03) had a higher seizure burden requiring escalation of antiepileptic medication to second line or need for maintenance treatment. Conclusion: Infants with neonatal stroke who present with early seizures are more likely to receive therapeutic hypothermia despite failure to fulfil both Toby A and B criteria. Early suspicion to inform management can be obtained more effectively from CFM while cranial ultrasound findings are generally non-specific. A normal CFM background, unilateral abnormal background and unilateral seizure activity are highly suggestive of neonatal stroke.

Hierarchical design of intelligent α-MnO2-based theranostics nanoplatform for TME-activated drug delivery and T1-weighted MRI

Intelligent “theranostic” nanoplatforms were synthesized based on porous α-MnO2 nanostructures with various concentrations of Gd3+ doping agent (H-GdxMn1-xO2; x = 0.00–0.12 in steps 0.03). After optimization of the doping agent at x = 0.09, the surface of the nanostructures was covered with the chitosan cross-linked tripolyphosphate (Chi-TPP) to reduce the inherent toxicity of the Gd3+ ion. Then a significant amount of doxorubicin (DOX) was loaded on these porous nanoplatforms. The characteristic features of the tumor microenvironment (TME) such as low pH, high temperature, high level of glutathione (GSH), and overproduced endogenous H2O2 act as the driving force to enhance catalytic destruction of the as-synthesized nanoplatforms. Catalytic demolition of H-GdxMn1-xO2@[email protected] “all in one” nanoplatforms not only cause the targeted delivery of pharmaceutical cargo at TME of cancer tissue but also was used for MRI bioimaging and diagnosis of TME condition. Apart from serving as a TME-activated MRI contrast agent, as-synthesized nanoplatforms also serve as a TME-modulating agent to form molecular O2 and simultaneously attenuated tumor hypoxia. Furthermore, the future directions and challenges in the development of the H-GdxMn1-xO2@[email protected] “all in one” theranostic nanosystems were discussed.

Comparison of diagnostic accuracy of magnetic resonance spectroscopy with conventional magnetic resonance imaging for brain tumors and correlation with histopathology as gold standard

Background; Despite a large number of revolutionized advancements in Imaging technology clinicians are unable to get hundred percent accurate results. Multiple limitations have been observed in Imaging modalities. Multiple published trials are available online on the evaluations of these imaging modalities individually and in combinations to find out the level of accuracy of these technologies. Objective; This study was design to compare the diagnostic accuracy of MR Spectroscopy with MRI which is a conventional diagnostic tool for brain tumor diagnosis while keeping the Histopathology evaluation as Gold Standard. Methodology; The patients who fulfilled the inclusion criteria were included in study. All scans were carried out on GE 1.5 Tesla MR Scanner using head coil. Axial T1W, T2W, Fluid Attenuated Inversion Recovery (FLAIR), Sagittal T2W, Coronal Fluid Attenuation. Inversion Recovery (FLAIR) and T1W post- contrast Axial, Coronal and Sagittal images were acquired. MR spectroscopy was performed through single voxel technique. Final diagnosis was made on histopathology results. The reports of biopsy were co related with radiological diagnosis of MRI and MRS. Results; The study was conducted with 150 suspected cases. There were 85(56.7%) males and 65(43.3%) females. Histopathology 100(66.7%) cases were confirmed positive and 50(33.3%) negative.

The role of using the kaiser score as a clinical decision rule for diagnosis in multi-parametric breast MRI

The role of using the kaiser score as a clinical decision rule for diagnosis in multi-parametric breast MRI

Fracture of the Costal Cartilage A Misdiagnosed Pathology: about A Case and Literature Review

Costal cartilage fracture is a rare traumatic injury but the frequency is probably underestimated. This lesion, not visible on X-rays, is nevertheless simple diagnosis in ultrasound, computed tomography and MRI. We report here a case of low anterior anterior post-traumatic chest pain following a brawl in a young patient in general condition.

Study of T2 mapping in quantifying and discriminating uterine lesions under different magnetic field strengths: 1.5 T vs. 3.0 T

BackgroundMRI is the best imaging tool for the evaluation of uterine tumors, but conventional MRI diagnosis results rely on radiologists and contrast agents (if needed). As a new objective, reproducible and contrast-agent free quantification technique, T2 mapping has been applied to a number of diseases, but studies on the evaluation of uterine lesions and the influence of magnetic field strength are few. Therefore, the aim of this study was to systematically investigate and compare the performance of T2 mapping as a nonenhanced imaging tool in discriminating common uterine lesions between 1.5 T and 3.0 T MRI systems.MethodsA total of 50 healthy subjects and 126 patients with suspected uterine lesions were enrolled in our study, and routine uterine MRI sequences with additional T2 mapping sequences were performed. T2 maps were calculated by monoexponential fitting using a custom code in MATLAB. T2 values of normal uterine structures in the healthy group and lesions (benign: adenomyosis, myoma, endometrial polyps; malignant: cervical cancer, endometrial carcinoma) in the patient group were collected. The differences in T2 values between 1.5 T MRI and 3.0 T MRI in any normal structure or lesion were compared. The comparison of T2 values between benign and malignant lesions was also performed under each magnetic field strength, and the diagnostic efficacies of the T2 value obtained through receiver operating characteristic (ROC) analysis were compared between 1.5 T and 3.0 T.ResultsThe mean T2 value of any normal uterine structure or uterine lesion under 3.0 T MRI was significantly lower than that under 1.5 T MRI (p < 0.05). There were significant differences in T2 values between each lesion subgroup under both 1.5 T and 3.0 T MRI. Moreover, the T2 values of benign lesions (71.1 ± 22.0 ms at 1.5 T and 63.4 ± 19.1 ms at 3.0 T) were also significantly lower than those of malignant lesions (101.1 ± 4.5 ms at 1.5 T and 93.5 ± 5.1 ms at 3.0 T) under both field strengths. In the aspect of differentiating benign from malignant lesions, the area under the curve of the T2 value under 3.0 T (0.94) was significantly higher than that under 1.5 T MRI (0.90) (p = 0.02).ConclusionT2 mapping can be a potential tool for quantifying common uterine lesions, and it has better performance in distinguishing benign from malignant lesions under 3.0 T MRI.

脊髓梗死的MRI诊断以及临床表现分析

脊髓梗死的MRI诊断以及临床表现分析

CAT-Net: A Cross-Slice Attention Transformer Model for Prostate Zonal Segmentation in MRI.

Prostate cancer is the second leading cause of cancer death among men in the United States. The diagnosis of prostate MRI often relies on accurate prostate zonal segmentation. However, state-of-the-art automatic segmentation methods often fail to produce well-contained volumetric segmentation of the prostate zones since certain slices of prostate MRI, such as base and apex slices, are harder to segment than other slices. This difficulty can be overcome by leveraging important multi-scale image-based information from adjacent slices, but current methods do not fully learn and exploit such cross-slice information. In this paper, we propose a novel cross-slice attention mechanism, which we use in a Transformer module to systematically learn cross-slice information at multiple scales. The module can be utilized in any existing deep-learning-based segmentation framework with skip connections. Experiments show that our cross-slice attention is able to capture cross-slice information significant for prostate zonal segmentation in order to improve the performance of current state-of-the-art methods. Cross-slice attention improves segmentation accuracy in the peripheral zones, such that segmentation results are consistent across all the prostate slices (apex, mid-gland, and base). The code for the proposed model is available at https://bit.ly/CAT-Net.

An MRI brain tumor segmentation method based on improved U-Net.

In order to improve the segmentation effect of brain tumor images and address the issue of feature information loss during convolutional neural network (CNN) training, we present an MRI brain tumor segmentation method that leverages an enhanced U-Net architecture. First, the ResNet50 network was used as the backbone network of the improved U-Net, the deeper CNN can improve the feature extraction effect. Next, the Residual Module was enhanced by incorporating the Convolutional Block Attention Module (CBAM). To increase characterization capabilities, focus on important features and suppress unnecessary features. Finally, the cross-entropy loss function and the Dice similarity coefficient are mixed to compose the loss function of the network. To solve the class unbalance problem of the data and enhance the tumor area segmentation outcome. The method's segmentation performance was evaluated using the test set. In this test set, the enhanced U-Net achieved an average Intersection over Union (IoU) of 86.64% and a Dice evaluation score of 87.47%. These values were 3.13% and 2.06% higher, respectively, compared to the original U-Net and R-Unet models. Consequently, the proposed enhanced U-Net in this study significantly improves the brain tumor segmentation efficacy, offering valuable technical support for MRI diagnosis and treatment.

PATTERN RECOGNITION ,APPROACH AND CLINICO-RADIOLOGIC EVALUATION OF MYELOPATHIES

INTRODUCTION Myelopathy, the rapid or insidious onset of motor and sensory abnormalities referable to the spinal cord, occurs as a result of a variety of causes that may be classied on the basis of their location of origin (intramedullary, intradural-extramedullary, and extradural). Osteophytic spurring and disc herniation may also produce myelopathy localised to the thoracic spine, though less commonly. Degenerative cervical myelopathy (DCM) is a progressive spine disease and the most common cause of spinal cord dysfunction in adults worldwide. AIM OF THE STUDY: 1. To study the clinical prole of cases of Myelopathy 2. To study the pattern of spinal cord involvement 3. To compare the clinical diagnosis with MRI diagnosis MATERIALS AND METHODS Ÿ Study setting: The study was conducted on patients admitted in the departments of neurology and neurosurgery with clinical suspicion of spinal cord disease Ÿ Study design: The current study was a prospective stud Ÿ Study period: The study was conducted between august 2021 to July 2022 Ÿ Sample size: The present prospective study included 30 patients admitted in the departments of neurology and neurosurgery with clinical suspicion of spinal cord disease. INCLUSION CRITERIA: Inclusion criteria include patients with clinical suspicion of spinal cord disease of age group 20-80 years were included as study participants in the present study. EXCLUSION CRITERIA: The current study excluded patients with seizures, abnormal higher mental functions, and children. RESULTS Ÿ In the present series, the patient's age ranged from 20 years to 80 years Ÿ The majority of 53.33% of participants were aged between 51 to 60 years, followed by 61 to 70 years was 13.33%, 21 to 30 years and 41 to 50 was 10% for each Ÿ Distribution of cases according to etiology: Among the study population, 17 (57%) participants had cervical spondylosis, 4 (13%) participants had caries spine, 3 (10%) participants had a vertebral fracture and others for each, 2 (7%) participants had uorosis and only 1 (3%) participant had spinal tumor Ÿ Clinico-radiological patterns of spinal cord syndromes: The majority of 50% of participants had Anterior + Posterior Cord involvement. The proportion of posterior cord, pan cord, and the central cord was 20%, 13.33%, and 10% respectively CONCLUSION: Ÿ The present study results have analyzed the prole of myelopathies and found more male preponderance Ÿ The commonest cause of compressive myelopathy was Cervical spondylosis Ÿ Anterior plus posterior cord syndrome was the commonest pattern seen which was followed by posterior cord syndrome and the least observed pattern was anterior cord syndrome. Hence, it can be concluded that MRI could suggest the location of the lesion and correlated well with clinical diagnosis.

Mental Health and Sleep Disorders in Obese Patients, Their Relationship with Clinical and Pathogenetic Changes in the Organism and Impact on Quality of Life

In the world the problem of obesity is an important aspect in the non-infectious diseases progression. It is known, that 95% of the pathogenic factor for the obesity development depends on the changes caused by neurochemical, hormonal and metabolic mechanisms that occur by depression, anxiety, eating disorders, circadian rhythms and sleepiness. The objective: to determine the psychosocial features and their relationship with clinical and pathogenetic changes in obese patients of working age. Materials and methods. 75 patients with obesity of the 1st and 2d degrees (39.03±0.93 years old) and 75 practically healthy persons of the corresponding age (36.84±0.96 years old) were examined. Waist circumference (WC), hip circumference (HC), body surface area (BSA), waist/hip ratio (WHR), conicity index (ConI), a body shape index (ABSI), abdominal volume index (AVI), blood pressure, blood levels of glucose, insulin, index HOMA, cholesterol, lipidogram indicators, serotonin, and leptin were determined in the patients. The psychological status was assessed by Hospital Anxiety and Depression Scale (HADS), Hamilton Anxiety Scale (HAM-A), the Dutch Eating Behavior Questionnaire (DEBQ), Epworth Sleepiness Scale (ESS), Pittsburgh Sleep Quality Index (PSQI), the patient’s quality of life – by 36-Item Short Form Survey (SF-36). The statistical analysis was conducted by IBM SPSS Statistics, Statistica 12, descriptive statistics Excel 2010. Results. The patients of the studied group, unlike the control group, had significantly higher indices of WC, HC, BMI, WHR, ConI, ABSI, AVI, HOMA index, blood pressure, glucose, insulin, total cholesterol, lipidogram, leptin and the lower level of serotonin. This was accompanied with clinically expressed anxiety of depression which led to eating disorders (a tendency “to eat emotions”, to overeat when food is available, a habit to eat without restrictions), sleep disorders (excessive day time sleepiness, low quality of sleep) and reduced quality of life. The results of correlation analysis indicate a strong or moderate positive correlation between obesity indices, glucose level, lipidogram, atherogenicity index, HOMA index, scores of depression and anxiety scales, as well as strong negative correlation with ABSI index, high density lipoproteins, serotonin. In addition, a positive correlation was found between leptin level and scores of depression and anxiety scales, scores of eating behavior, sleepiness, and a negative correlation was determined between these indices and serotonin level. Conclusions. The patients of the studied group (100.0 %) had abdominal obesity. Among the anthropometric indices ABSI and AVI ones were the most significant and informative for determination of abdominal obesity in gender aspect, this can be an alternative for MRI diagnosis of visceral obesity at the level of primary medical care. A close relationship between abdominal obesity with psycho-emotional disorders, disorders of sleep and eat behavior, metabolic disorders, leptin and serotonin levels was determined. Taking into account these interrelationships in a patient-centered management in persons with obesity will improve the quality of medical care.

MRI diagnosis and radical prostatectomy outcomes.

MRI diagnosis and radical prostatectomy outcomes.

Difference in the fibular attachment structure between the superior and inferior fascicles of the anterior talofibular ligament using ultrasonography and histological examinations.

The anterior talofibular ligament (ATFL) is divided into superior (SB) and inferior bands (IB). Although the differences in length and width are known, the structure of the fibular attachment had not been elucidated. The present study aimed to clarify the differences in the fibular attachment structure between ATFL's SB and IB using cross-sectional images along the ligament. An anatomical study using 15 formalin-fixed ankles was performed. The lateral ankle ligament complex was collected after a longitudinal image of SB/IB was visualized by ultrasonography. The specimens were decalcified and sectioned longitudinally at the center of SB/IB using a microtome. Histological evaluation of the enthesis structure at the fibular attachment of SB/IB was performed using hematoxylin-eosin and Masson's trichrome stains. A fibrillar pattern could not be observed in the longitudinal image at the IB level by ultrasonography. The lengths of ATFL's SB and IB were 20.6 ± 1.6 and 15.3 ± 1.3mm, respectively, with thicknesses of 1.8 ± 0.4 and 1.0 ± 0.4mm, respectively. The ATFL's IB was significantly shorter and thinner than the ATFL's SB. The fibular attachment of ATFL's SB had distinct enthesis structure, whereas in the attachment structure of the ATFL's IB, there were several variations including a type with a narrower enthesis structure than the ATFL's SB and a type that merged with or wrapped around the calcaneofibular ligament. The fibular attachment structure between ATFL's SB and IB differs. Our results could be useful information when performing ultrasonography and MRI diagnosis.

Influence of COVID-19 pandemic on the decision making of patients in undergoing gamma knife radiosurgery

PurposeGamma knife radiosurgery (GK) is a commonly used approach for the treatment of intracranial lesions. Its radiation response is typically not immediate, but delayed. In this study, we analyzed cases from a prospectively collected database to assess the influence of COVID-19 pandemic on the decision making in patients treated by gamma knife radiosurgery.MethodsFrom January 2019 to August 2021, 540 cases of intracranial lesions were treated by GK with 207 cases before COVID-19 pandemic as a control. During the COVID-19 pandemic, 333 cases were similarly treated on patients with or without the COVID-19 vaccination. All the GK treated parameters as well as time profile in the decision making were analyzed. The parameters included age, sex, characteristic of lesion, targeted volume, peripheral radiation dose, neurological status, Karnofsky Performance Status (KPS), time interval from MRI diagnosis to consultation, time interval from the approval to treatment, frequency of outpatient department (OPD) visit, and frequency of imaging follow-up.ResultsLonger time intervals from diagnosis to GK consultation and treatment were found in the pandemic group (36.8 ± 25.5/54.5 ± 27.6 days) compared with the pre-COVID control (17.1 ± 22.4/45.0 ± 28.0 days) or vaccination group (12.2 ± 7.1/29.6 ± 10.9 days) (p < 0.001, and p < 0.001, respectively). The fewer OPD visits and MRI examinations also showed the same trends. High proportion of neurological deficits were found in the pandemic group (65.4%) compared with the control (45.4%) or vaccination group (58.1%) (p < 0.001). The Charlson comorbidity in the pandemic group was 3.9 ± 3.3, the control group was 4.6 ± 3.2, and the vaccination group was 3.1 ± 3.1. There were similar inter-group difference (p < 0.001). In multiple variant analyses, longer time intervals from the diagnosis to consultation or treatment, OPD frequency and MRI examination were likely influenced by the status of the COVID-19 pandemic as they were alleviated by the vaccination.ConclusionsThe decision making in patients requiring gamma knife treatment was most likely influenced by the status of the COVID-19 pandemic, while vaccination appeared to attenuate their hesitant behaviors. Patients with pre-treatment neurological deficits and high co-morbidity undergoing the gamma knife treatment were less affected by the COVID-19 pandemic.

PH-Responsive chitosan-modified gadolinium oxide nanoparticles delivering 5-aminolevulinic acid: A dual cellular and metabolic T1-T2* contrast agent for glioblastoma brain tumors detection

Diagnosis of glioblastoma multiforme (GBM) with high sensitivity and specificity requires cellular and metabolic imaging. In this study, we aimed to endow gadolinium oxide nanoparticles (Gd2O3 NPs) as an interesting T1 contrast agent with T2* contrast ability through intracellular iron assistance induced by 5-aminolevulinic acid (5-ALA) with ferric ammonium citrate (FAC). Chitosan (CS)-loaded Gd2O3 NPs were loaded with 5-aminolevulinic acid (Gd2O3/CS-TPP@5-ALA NPs), and then evaluated as a pH-responsive dual cellular and metabolic T1 and T2* contrast agent for GBM diagnosis. The target NPs with an average size of 138 nm revealed concentration-dependent toxicity against glioblastoma C6 cells. The relaxation rate (R1) of the C6 cells treated with Gd2O3/CS-TPP@5-ALA NPs plus FAC was remarkably higher than the controls including Gd2O3/CS-TPP NPs, Gd2O3/CS-TPP@5-ALA NPs, and FAC (17.2 ± 2.4 vs. 8 ± 1.5, 8.4 ± 1.2, and 2.3 ± 0.5, respectively). Interestingly, the transverse relaxation rate (R2*) of the C6 cells treated with Gd2O3/CS-TPP@5-ALA NPs + FAC was significantly higher than the controls including Gd2O3/CS-TPP NPs, Gd2O3/CS-TPP@5-ALA NPs, and FAC (96 ± 8 vs. 25 ± 4, 27 ± 5, and 19 ± 2.6, respectively). We believe that Gd2O3/CS-TPP@5-ALA NPs + FAC could be introduced as promising dual cellular and metabolic T1-T2* contrast agents for GBM diagnosis in MRI.

Quantitative MRI indicators and features for partial subscapularis tendon tears on conventional shoulder MRI

BackgroundDiagnosing partial subscapularis (SSC) tendon tears still faces challenges. A failure rate of massive posterosuperior rotator cuff tear repair will be highly increased when extending more than one-third of SSC tendon. This study aims to investigate the quantitative indicators and features of partial SSC tears on conventional shoulder MRI and improve the preoperative diagnostic accuracy.Materials and methodsFour hundred and thirty-seven patients underwent MRI and arthroscopy were retrospectively reviewed; 89 patients with partial SSC tears in case group and 50 patients with normal SSC in control group were included. Six MRI features with the explicit definition of some quantitative indicators were evaluated.ResultsFissure sign showed the highest diagnostic efficiency for the partial SSC tears, with a specificity of 92%, sensitivity of 75.3%, and accuracy of 81.3%. Thinning of SSC, fluid collection under the coraco-glenoid arch (CGA), and combined SSP complete tear also showed high specificity of 86%, 80%, and 80%, respectively, while the sensitivity and accuracy were moderate, with a sensitivity of 38.2%, 50.6%, and 48.3%, respectively, an accuracy of 55.4%, 61.2%, and 59.7%, respectively. The specificity, sensitivity, and accuracy of lesser tuberosity cysts were all moderate with values of 68%, 56.2%, and 60%, respectively. However, fat accumulation under the CGA showed no significant difference between the partial SSC tears group and the control group.ConclusionSeveral specific MRI features with quantitative indicators defined in this study can be used to improve the accuracy of preoperative MRI diagnosis of partial SSC tears.

Super-Resolution Reconstruction Algorithm-Based MRI Diagnosis of Prostate Cancer and Evaluation of Treatment Effect of Prostate Specific Antigen

MRI of prostate cancer (PCa) was performed using a projection onto convex sets (POCS) super-resolution reconstruction algorithm to evaluate and analyze the treatment of prostate-specific antigen (PSA) and provide a theoretical reference for clinical practice. A total of 110 patients with PCa were selected as the study subjects. First, the modified POCS algorithm was used to reconstruct the MRI images, and the gradient interpolation algorithm was used instead of the traditional bilinear algorithm to preserve the edge information. The diagnostic and therapeutic effects of MRI examination, PSA examination, and MRI combined with PSA based on a super-resolution reconstruction algorithm were then compared. The simulation results showed that the POCS algorithm was superior to the bilinear interpolation results and was superior to the common POCS algorithm. After adding noise artificially, the restoration algorithm was effective and could preserve the details in the image. The performance indexes of PSA in the diagnosis of PCa were 75.4%, 60.1%, 70.08%, 72.2%, and 60.3%, respectively; the performance indexes of MRI in the diagnosis of PCa were 84.6%, 61.4%, 71.11%, 73.08%, and 61.9%, respectively; and the performance indexes of MRI combined with PSA based on the super-resolution reconstruction algorithm in the diagnosis of PCa were 96.05%, 88.3%, 95.1%, 93.6%, and 92.7%, respectively. The indicators of MRI combined with PSA based on the super-resolution reconstruction algorithm were significantly higher than those of the other two methods ( P &lt; 0.05). The signal-to-noise ratio of MRI of PCa based on the super-resolution reconstruction algorithm has been greatly improved, with good clarity, which can improve the diagnostic accuracy of PCa patients and has certain advantages in the examination. MRI based on the super-resolution reconstruction algorithm has a high value in the diagnosis and treatment of PCa.

Accuracy of Magnetic Resonance Imaging in Prenatal Diagnosis of Choledochal Cysts: A Single-Center Retrospective Analysis.

Background The purpose of this study is to evaluate the accuracy of prenatal MRI in diagnosing choledochal cysts (CDC), evaluate the sensitivity and specificity of MRI signs in the diagnosis of fetal CDC, and first compare the trend of size of CC between prenatal and postpartum. Methods A total of 18 fetal who were diagnosed with CDCs through prenatal MRI were enrolled in the study. We summarized and analyzed the prenatal clinical data and prognosis information of prenatal and postpartum surgery, then compared the sensitivity, specificity, and diagnostic accuracy of various signs of MRI and postpartum MRCP diagnosis of CC. Finally, we tried to compare the earliest prenatal detection of common bile duct cysts with the size of surgery, and calculated the growth rate of common bile duct cysts for the first time. Results All 18 patients were delivered in our institution. Among these patients, 14 were confirmed with CDCs after postpartum surgery, two patients had CDCs that disappeared, and two patients were confirmed with cystic biliary atresia (CBA) through the Kasai operation. Furthermore, 13 patients with CDCs and two patients with CBA underwent MRCP before the operation, and one patient with CDCs ruptured at birth and underwent ultrasound diagnosis. The sensitivity and diagnostic compliance of prenatal MRI signs for the location were higher when compared to postnatal MRCP (100% vs. 76.9% and 83.3% vs. 66.7%): the cyst was located at the porta hepatis, which was higher than the lowest edge of the liver, and parallel to the hepatoduodenal ligament. Conclusion Prenatal MRI is higher than that of US for diagnosing CDCs, specifically in identifying the location of the cyst and confirming the origin of the cyst. The length, width, and size of the CDC become slightly bigger in our study.

Deep and hybrid learning of MRI diagnosis for early detection of the progression stages in Alzheimer’s disease

Alzheimer’s, or so-called dementia, is one of the types of diseases that affects brain cells and causes memory loss, difficulty in thinking, and forgetfulness. Thus far, there is no effective treatment for AD, but treatment could be helpful in impeding the progression of the disease. Therefore, early AD diagnosis is effective in limiting the disease from progressing to advanced and dangerous stages. Physicians and radiologists face difficulties in diagnosing healthy nerve cells from soft tissue, and it requires substantial expertise and a long time to decipher the MRI images. Thus, artificial intelligence techniques can play a key role in diagnosing MRI images for early detection of AD. In this study, four proposed systems with different methodologies and materials for tracking the stages of AD development are presented. The first proposed system is to classify a data set using artificial neural networks (ANNs) and feed-forward neural networks (FFNN) based on the features extracted in a hybrid manner by using a combination of Local Binary Pattern (LBP), Discrete Wavelet Transform (DWT), and Gray Level Co-occurrence Matrix (GLCM) algorithms. The second proposed system is to classify the data set using two deep learning models – ResNet-18 and AlexNet – that are pre-trained based on deep feature map extraction. The third proposed system is to diagnose the data set using a hybrid technology between ResNet-18 and AlexNet models to extract feature maps and machine learning (SVM) to classify feature maps. The fourth proposed system diagnoses the data set using ANN and FFNN algorithms based on the hybrid features of ResNet-18 and AlexNet deep learning models and traditional algorithms (LBP, DWT, and GLCM). All the proposed techniques achieved superior results in the diagnosis of MRI images for early detection of AD. The FFNN algorithms based on the hybrid features extracted by ResNet-18 with features extracted using traditional algorithms achieved an accuracy of 99.8%, precision of 99.9%, sensitivity of 99.75%, specificity of 100%, and AUC of 99.94%.