Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) may provide more information in diagnosis of malignant tumor compared to conventional magnetic resonance imaging (MRI). Nowadays, in order to utilize the information expediently and efficiently, many researchers are aiming at the development of computer-aided diagnosis (CAD) of malignant tumor based on DCE-MRI. In this review, we survey the research in this field and summarize the literature in four parts, i.e. ① image preprocessing——noise reduction and image registration; ② region of interests (ROI) segmentation; ③ feature extraction——exploring the image characteristics by analyzing the ROI quantitatively; ④ tumor lesion recognition and classification——distinguishing and classifying tumor lesions by learning the features of ROI. We summarize the application of CAD techniques of DCE-MRI for cancer diagnosis and, finally, give some discussion on how to improve the efficiency of CAD in the future research.
How to extract high discriminative features that help classification from complex resting-state fMRI (rs-fMRI) data is the key to improving the accuracy of brain disease recognition such as schizophrenia. In this work, we use a weighted sparse model for brain network construction, and utilize the Kendall correlation coefficient (KCC) to extract the discriminative connectivity features for schizophrenia classification, which is conducted with the linear support vector machine. Experimental results based on the rs-fMRI of 57 schizophrenia patients and 64 healthy controls show that our proposed method is more effective (i.e., achieving a significantly higher classification accuracy, 81.82%) than other competing methods. Specifically, compared with the traditional network construction methods (Pearson’s correlation and sparse representation) and the commonly used feature selection methods (two-sample t-test and Least absolute shrinkage and selection operator (Lasso)), the algorithm proposed in this paper can more effectively extract the discriminative connectivity features between the schizophrenia patients and the healthy controls, and further improve the classification accuracy. At the same time, the discriminative connectivity features extracted in the work could be used as the potential clinical biomarkers to assist the identification of schizophrenia.
Objective To observe the characteristics of magnetic resonance diffusion tensor imaging（MR-DTI）for optic nerves and optic radiation in blind patients.Methods The optic nerves and optic radiation of 20 blind patients（blind group）and 20 controls（control group） were scanned by MR-DTI. Fractional anisotropy (FA) and directional encoded color (DEC) maps were acquired through postprocessing with the aid of volumeone 1.72 software. The signal intensity of optic nerves and optic radiation were then observed. The FA, mean diffusivity (MD), lambda;∥ and lambda;perp; value of bilateral optic nerves and optic radiation in two groups were measured in the DEC maps.Results While the high signal intensity was found in bilateral optic nerves in FA and DEC maps in control group，the signal decreased markedly in the blind group. The FA and lambda;∥ value of optic nerves in the blind group were declined obviously compared to that in the control group. The difference was statistically significant (t=16.294, 14.660;P=0.000). The MD and lambda;perp; value of optic nerves in the blind group were increased obviously compared to that in the control group, the difference was also statistically significant (t=8.096, 8.538; P=0.000). The high signal intensity was found in bilateral optic radiation in FA and DEC maps in both the blind and control groups. There were no statistic differences in FA and MD value in bilateral optic radiation between the blind and control groups (Left：t=1.456，1.811；P=0.152，0.076. Right：t=0.779，0.073；P=0.440，0.942). Conclusion A low signal intensity of bilateral optic nerves and a high signal intensity of bilateral optic radiation were found in blind patients.
ObjectiveTo analyze and conclude CT and MRI imaging features of ectopic pancreas in gastrointestinal tract so as to improve the understanding of the features.MethodsThe clinical, imaging, and pathological data of 12 patients with ectopic pancreas in the gastrointestinal tract confirmed by the pathology in the Sichuan Provincial People’s Hospital from November 2016 to June 2019 were retrospectively analyzed. The characteristics of image presentation were summarized.Results① The anatomical distribution: all patients had a single lesion. Of the 12 cases, 6 cases located in the gastric body lesser curvature, 3 cases located in the gastric angle, 1 case located in the posterior wall of gastric antrum, 1 case occurred in the upper jejunum, and 1 case occurred in the terminal ileum; 8 cases located in the submucosa, 2 cases located in the submucosa and muscular layer simultaneously, 1 case located in the submucosa, muscular and serous layer simultaneously, and 1 case located in the muscular layer. ② Size of the lesions: the maxium dimensions of the lesions were all 3 cm or less, and the long axes of the lesions were parallel to the gastrointestinal tract wall in 10 cases. ③ The internal characteristics: the results of 9 of 11 cases showed the isodensity compared to main pancreas on the plain CT scan. The results of 8 patients with enhanced CT showed the moderate to obvious enhancement, with 2 cases showed the slightly enhanced flaky or tube-like foci. In the arterial phase and portal venous phase, 6 cases showed the isodensity compared to main pancreas respectively. The result of MRI in 1 patient showed the isointensity compared to main pancreas on the plain scan and obviously heterogeneous enhancement.ConclusionCT and MRI could provide some information about location, size, and internal density or intensity of ectopic pancreas, and could be helpful for diagnosis.
ObjectiveTo investigate the value of gadolinium ethoxybenzyl diethylenetriamine pentaacetic acid enhanced magnetic resonance imaging (Gd-EOB-DTPA-MRI) in gross morphological classification of hepatocellular carcinoma (HCC). MethodsThe clinicopathologic data of patients with HCC who received surgical treatment in the Affiliated Huai’an Hospital of Xuzhou Medical University from January 2017 to December 2022 were retrospectively gathered. The Gd-EOB-DTPA-MRI was performed before operation. Two radiologists independently assessed the gross morphological classification of HCC according to the imaging performance. The tumors were cut into sections in a coronal plane and were taken pictures for recording pathological features after operation. The tumors were assigned into 4 types according to the references and clinical experiences: single nodular type (SN), single nodular with extranodular growth type (SN-EG), confluent multi-nodular type (CMN), and infiltration type (IF). Matching degree of morphological classification was analyzed between by the Gd-EOB-DTPA-MRI and resected specimen. The pathological features of 4 types of HCC were also analyzed. ResultsA total of 87 patients with HCC were included. The gross morphological classification by the Gd-EOB-DTPA-MRI was 28 (32.2%) patients with SN, 28 (32.2%) patients with SN-EG, 21 (24.1%) patients with CMN, 10 (11.5%) patients with IF, which by the resected specimen was 33 (37.9%) patients with SN, 24 (27.6%) patients with SN-EG, 21 (24.1%) patients with CMN, and 9 (10.4%) patients with IF in the 87 patients with HCC. The Kappa’s coefficient of agreement between the results of Gd-EOB-DTPA-MRI and postoperative resection specimens was 0.776 (P=0.199). There were statistical differences in the tumor diameter and microvascular invasion (MVI) among the 4 types of gross morphology classification (F=2.937, P=0.038; χ2=16.852, P=0.001), the MVI rate was highest and tumor diameter was biggest in the patients with IF among the 4 types of gross morphology classification (P<0.05). ConclusionsFrom the results of this study, the gross morphological classification of HCC is closely related to the tumor diameter and MVI. Results of Gd-EOB-DTPA-MRI and postoperative resection specimens in assessing the gross morphological classification are good agreement. Therefore, an accurate preoperative planning and better therapy strategy for the patients with HCC can be provided according to gross morphological classification by preoperative Gd-EOB-DTPA-MRI.
Objective To explore the accuracy of contrast-enhanced magnetic resonance imaging (MRI) in predicting pathological complete remission (pCR) in breast cancer patients after neoadjuvant therapy (NAC). Methods The clinicopathological data of 245 patients with invasive breast cancer who had completed the surgical resection after NAC in the Affiliated Hospital of Southwest Medical University from March 2020 to April 2022 were collected retrospectively. According to the results of hormone receptor (HR) and human epidermal growth factor receptor 2 (HER2) detected by immunohistochemistry, all patients were divided into four subgroups: HR+/HER2–, HR+/HER2+, HR–/HER2+ and HR–/HER2–. The value of MRI in evaluating the efficacy of NAC was analyzed by comparing the postoperative pathological results as the gold standard with the residual tumor size assessed by preoperative MRI. Meanwhile, the sensitivity, specificity and positive predictive value (PPV) of pCR predicted by the evaluation results of enhanced MRI were analyzed, and further analyzed its predictive value for pCR of different subtypes of breast cancer. Results There were 88 cases (35.9%) achieved radiological complete response (rCR) and 106 cases (43.3%) achieved pCR in 245 patients. Enhanced MRI in assessing the size of residual tumors overestimated and underestimated 12.7% (31/245) and 9.8% (24/245) of patients, respectively. When setting rCR as the MRI assessment index the specificity, sensitivity and PPV were 84.2% (117/139), 62.3% (66/106) and 75.0% (66/88), respectively. When setting near-rCR as the MRI assessment index the specificity, sensitivity and PPV were 70.5% (98/139), 81.1% (86/106), and 67.7% (86/127), respectively. The positive predictive value of both MRI-rCR and MRI-near-rCR in evaluating pCR of each subtype subgroup of breast cancer was the highest in the HR–/HER2+ subgroup (91.7% and 83.3%, respectively). In each subgroup, compared with rCR, the specificity of near-rCR to predict pCR decreased to different degrees, while the sensitivity increased to different degrees. Conclusions Breast contrast-enhanced MRI can more accurately evaluate the efficacy of localized breast lesions after NAC, and can also more accurately predict the breast pCR after NAC. The HR–/HER2+ subgroup may be a potentially predictable population with pCR exemption from breast surgery. However, the accuracy of the evaluation of pCR by breast enhancement MRI in HR+/HER2– subgroup is low.
Objective To investigate the CT and MR imaging manifestation of rare pancreatic tumors in order to deepen the understanding of their imaging characteristics and improve the diagnostic accuracy. Methods Clinical and image date of 34 cases with rare pancreatic tumors proved by surgery and histopathologically were analyzed retrospectively. Including neuroendocrine tumors of the pancreas (NETP,n=13), solid-pseudopapillary tumors of pancreas (SPTP,n=10), intraductal papillary mucinous neoplasms (IPMN,n=2), serous cystadenoma (SCA,n=7), and mucinous cystadenoma (MCA,n=2). Examined by CT in 19 cases, by MRI in 11 cases, examined by CT and MRI at the same time in 4 cases. The characterized imaging features of each kind of tumors were analyzed emphatically. Results Of the 13 cases of NETP, solid lesions in 6 cases, predominantly soild in 4 cases, predominantly cystic in 3 cases. Homogenous enhancement in 6 cases, heterogeneous enhancement in 7cases, the soild constituent of all cases were showed moderate to marked enhancement. Of the 10 cases of SPTP, predominantly soild in 2 cases, soild and cystic in 5 cases, predominantly cystic in 3 cases. The solid part of 10 cases presented as gradually enhancement, 2 cases appeared hemorrhage, 1 case appeared stippled calcification. Of the 2 cases of IPMN, both of them were combined type, showed multilocular cystic tumors due to the dilated of the pancreatic duct. Of the 7 cases of SCA, microcystic partten in 3 cases and single cyst partten in 4 cases, showed unilocular or multilocular cystic with clear boundary. The 2 cases of MCA, showed unilocular cystic with clear boundary. Conclusion Different histological types of pancreatic rare tumor appeared different kinds of imaging characteristic, we may improve the diagnostic accuracy by analyzing their features.
Alzheimer’s disease (AD) is an irreversible neurodegenerative disorder that damages patients’ memory and cognitive abilities. Therefore, the diagnosis of AD holds significant importance. The interactions between regions of interest (ROIs) in the brain often involve multiple areas collaborating in a nonlinear manner. Leveraging these nonlinear higher-order interaction features to their fullest potential contributes to enhancing the accuracy of AD diagnosis. To address this, a framework combining nonlinear higher-order feature extraction and three-dimensional (3D) hypergraph neural networks is proposed for computer-assisted diagnosis of AD. First, a support vector machine regression model based on the radial basis function kernel was trained on ROI data to obtain a base estimator. Then, a recursive feature elimination algorithm based on the base estimator was applied to extract nonlinear higher-order features from functional magnetic resonance imaging (fMRI) data. These features were subsequently constructed into a hypergraph, leveraging the complex interactions captured in the data. Finally, a four-dimensional (4D) spatiotemporal hypergraph convolutional neural network model was constructed based on the fMRI data for classification. Experimental results on the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database demonstrated that the proposed framework outperformed the Hyper Graph Convolutional Network (HyperGCN) framework by 8% and traditional two-dimensional (2D) linear feature extraction methods by 12% in the AD/normal control (NC) classification task. In conclusion, this framework demonstrates an improvement in AD classification compared to mainstream deep learning methods, providing valuable evidence for computer-assisted diagnosis of AD.
摘要:目的:探讨平山病的MRI影像特点及其临床应用价值。方法:5例临床确诊平山病病例组和10例正常对照组进行颈椎自然位及屈颈位MRI检查,矢状位T1WI、T2WI及轴位T2WI,观察颈髓、蛛网膜下腔及硬膜外腔变化情况。结果:病例组的5例平山病均系16~20岁男性。自然位:5例下位颈髓均萎缩变扁,硬膜外间隙未显示增宽。屈颈位:5例C5~7颈髓前移变扁中,将变扁颈髓又分为上中下三段,以中段最窄,上下段渐移行至正常;C5~7蛛网膜下腔亦变窄,硬脊膜伴随前移;而C4~7硬脊膜后间隙则增宽,呈新月形影,增宽程度分为轻、中、重三度,最重者位于C6椎体平面,T2加权像及T1WI增强呈高信号,其中1例内见血管流空信号影。对照组为10例志愿者,自然位: 4例C3~7颈髓形态、大小基本一致,6例颈髓自颈3逐渐移行与胸1脊髓其大小一致;屈颈位:颈髓和蛛网膜下腔大小与自然位比较无明显变化,硬膜后间隙自C3平面向下延至T1平面,T2WI上呈均匀线样高信号影。结论:下位颈髓萎缩变扁,屈颈位颈髓及硬脊膜前移、硬脊膜后间隙增宽呈新月形影等,是临床诊断平山病较特征性的MRI表现。Abstract: Objective: To evaluate clinical value and MRI features of Hirayama disease. Methods: Five cases of hirayama disease, which had been clinically confirmed using siemens sonata 1.5T MRI scan, physiological condition and flexional condition, Sagittal view T1WI, T2WI and Axial View T2WI, and GdDTPA enhanced examination, for MRI changes of spinal cord, subarachnoid cavity, duramater of spinal membrae and extra dural space, etc were studied. Results: In case group of 5 cases of hirayama disease, age was mainly in 16–20 years old, All of 5 cases were men. Which were pressed and become thinner of spinal cord, strictic changes of subarachnoid cavity, new moony shape and enlargement changes and pushed forward of extra duramater space, and higher intensity signal of GdDTPA enhancement, and vascular flow effect (one case ) in C5–C7. but also, for contrast group 10 cases of normal volunteer, physiological condition:4 cases in cervical spinal cord with shape and structure were uniformity, and duramater, subarachnoid cavity, extra duramater space etc in C3–C7 were abnormal. Six cases in cervical spinal cord with shape and structure gradully changed from C3 to T1; flexional condition: 10 cases of MRI changes of spinal cord, subarachnoid cavity were as same as it in physiological condition,all of T2 WI, higher intensity signal were homogeneous of extraduramater space in C3–T1. Conclusion: The feature findings of cervical spinal cord became thinner, and cervical cord, durameter pussed forward, new moony shape and enlargment of extradurameter space, vascular flow effusion, etc in MRI were useful value for clinical diagnosis.
The incidence of prostate cancer ranks the second in malignant tumors among elderly males. Multi-parametric MRI (Mp-MRI) is an important mean for detection, staging, and grading of prostate cancer. In order to standardize the collection, interpretation, and reporting of prostate MRI data, the European Urogenital Radiology Society launched the Prostate Imaging Reporting and Data System (PI-RADS) in 2012. Due to some limitations in the application process, the Joint Committee of the American Society of Radiology and the European Society of Radiology issued an updated version of PI-PADS V2 in 2014. In recent years, some studies have been carried out on the effectiveness, accuracy, and consistency of the diagnosis of prostate cancer. This article will review the application and research status of PI-RADS V2 system in the diagnosis of Mp-MRI for prostate cancer.