ObjectiveTo discuss the value of diffusion weighted imaging (DWI) in the diagnosis of placenta increta.
MethodsThe clinical data of 42 patients with placenta increta admitted to Sichuan Provincial Hospital for Women and Children between May 2012 and January 2014 were retrospectively analyzed. All the patients were examined by prenatal magnetic resonance scans and DWI scans for subsequent comparison between ADC of the local convex placental region and ADC of the normal placental region and between the results of the two imaging methods.
ResultsADC of the implantation area was significantly different from that of the normal placenta, so it could be used as a quantitative index. DWI had a higher sensitivity of diagnosis than conventional MRI.
ConclusionCompared with conventional magnetic resonance imaging, DWI is more valuable in the clinical diagnosis of placenta increta, which provides a reliable basis for clinical treatment.
Objective
To explore the MRI features of juvenile-onset ankylosing spondylitis (JoAS) for improving the level of diagnosis and therapy.
Methods
MRI findings of JoAS in 25 patients confirmed by clinical and laboratory results between October 2010 and September 2014 were retrospectively analyzed.
Results
There were a total of 67 locations of lesion in the 25 cases, including 19 in sacroiliac joint, 21 in hip joint, 6 in ischial tuberosity, 6 in crista iliaca, 9 in knee joint, 4 in ankle joint, and 2 in foot. MRI showed 63 locations with bone marrow edema, 36 with joint effusion, 26 with bone destruction, and 19 with enthesitis.
Conclusions
Bone marrow edema is the most common MRI manifestation of JoAS, and the main bone destruction is middle axis joint. MRI is sensitive but not specific to lesions, so differential diagnosis is needed and helpful.
ObjectiveTo investigate the efficacy of vagus nerve stimulation (VNS) in patients with refractory magnetic resonance imaging (MRI)-negative epilepsy and to evaluate potential clinical predictors. MethodsA retrospective collection of efficacy data was conducted on 24 patients with intractable MRI-negative epilepsy treated with VNS, who were followed up for more than six months, at Beijing Tiantan Hospital and Beijing Fengtai Hospital from January 2016 to September 2023. Patients were divided into two subgroups based on their response to VNS: responders (≥50% reduction in seizure frequency) and non-responders (<50% reduction in seizure frequency). The relationship between preoperative clinical data and VNS efficacy was further analyzed to identify potential predictors of VNS efficacy. ResultsA total of 24 patients were included, with an average age of (14.26±8.39) years old. Seizure frequency was reduced by more than 50% in 37.5% of patients, and 8.3% of patients achieved seizure-free after VNS treatment. Preoperative seizure frequency and interictal epileptiform discharge type were significantly associated with VNS efficacy (P<0.05). Multivariate regression analysis showed that a monthly seizure frequency of less than 100 and focal interictal epileptiform discharges were independent predictors of VNS efficacy (P<0.05). ConclusionVNS is an effective treatment for patients with refractory MRI-negative epilepsy. Lower monthly seizure frequency and focal interictal epileptiform discharges are potential predictors of VNS efficacy. These findings provide important references for clinicians in selecting and evaluating patients for VNS treatment.
Objective
To discuss the clinical value of whole spine magnetic resonance imaging (WSMRI) in practice of neurosurgical spinal surgery.
Method
A total of 70 cases examined using WSMRI between January 2015 and December 2016 were collected and analyzed retrospectively.
Results
All patients got clear images of WSMRI. Eighteen cases got important information, including spinal variation (1 case), multiple lesions (3 cases), combined lesions (6 cases) and large range multi-segmental lesions (8 cases), which were missed by single-segment MRI .
Conclusions
WSMRI can show all the spine, spinal cord and surrounding tissue in one image at one time. It has high clinical value because of its accurate positioning, comprehensiveness, time saving, and low rate of misdiagnosis and missed diagnosis.
ObjectiveTo observe the correlation between posterior myopic retinoschisis(MRS) and posterior scleral staphyma (PS) in pathological myopia (PM), and to preliminarily explore the influencing factors of MRS.MethodsA retrospective case series study. From November 2016 to November 2019, 38 patients with PM with MRS diagnosed in Henan Eye Hospital & Henan Eye Institute from were included in the study. There were 10 males and 28 females; 13 patients were binocular and 25 patients were monocular. The average age was (49±13) years old. BCVA, retinoscopy optometry, frequency domain OCT, three-dimensional magnetic resonance imaging (3D-MRI) examination and axial length (AL) measurement were performed. According to the frequency domain OCT inspection results, MRS was divided into inner splitting, outer splitting and mixed splitting; based on the 3D-MRI scan results, PS was divided into broad macula, narrow macula,discoid, nasal, subdisc and other types. The correlation between MRS and PS was tested by χ2 test or Fisher exact test.ResultsAmong 60 eyes, 58 eyes (96.77%) of MRS combined with PS. Among them, the wide macula, narrow macula, discoid, nasal, subdisc, and other types were 30 (51.72%), 19 (32.75%), 1 (1.72%), 2 (3.48%), 2 (3.48%) and 4 (6.85%) eyes; inner split, outer split, and mixed split were 10 (17.24%), 24 (41.38%), 24 (41.38%) eyes. Of the 19 eyes with narrow macular PS, MRS involved the fovea in 16 eyes; of the 39 eyes with PS of other forms, MRS involved the fovea in 22 eyes. There was a statistically significant difference between the narrow macular type and other types involving foveal eyes (P=0.044). The correlation between MRS involving the fovea and narrow macular PS was moderate (Cramer's V=0.275). The ages of patients with inner split, outer split, and mixed split were 44±12, 56±10, and 44±13 years, respectively. Patients with inner splitting were younger than those with outer splitting, and those with outer splitting were older than those with inner splitting and mixed splitting. The differences were statistically significant (P=0.010, 0.010, 0.060).ConclusionPM with MRS mostly occur in PS-affected eyes, and mainly macular PS (wide macula, narrow macula).
Objective
To investigate the application of the dynamic contrast enhanced MRI (DCE-MRI ) combined with magnetic resonance spectroscopy (MRS) in the diagnosis of prostate cancer.
Method
A total of 60 patients with prostate cancer and 60 patients with benign prostatic hyperplasia diagnoses in Sichuan Cancer Hospital from January 2011 to January 2014 were included as prostate cancer group and proliferative group respectively. Sixty healthy individuals during the same period were included as the control group. We used Siemens Avanto 1.5 T high field superconducting MRI for DCE-MRI scan and MRS scan. After the MRS scan was finished, we used the workstation spectroscopy tab spectral analysis. Eventually we got the crest lines of prostate metabolites choline (Cho), creatine (Cr) and citrate (Cit). Then we calculated Cho/Cit, (Cho+Cr)/Cit and their average.
Results
Comparing the signal value in 21 seconds, 1 minute, 2 minutes of DCE-MRI, the differences among the three groups were statistically significant (P<0.05). Comparing the results of spectral analysis, the differences among the three groups were statistically significant (P<0.05). The sensitivity was 89.67%, the specificity was 95.45% and the accuracy was 94.34% when using DCE-MRI combined with MRS.
Conclusion
DCE-MRI combined with MRS greatly improves the sensitivity, specificity and accuracy of the diagnosis of prostate cancer; it has a great application value in the diagnosis of prostate cancer.
Objective To observe the changes of visual pathway of central nervous system in patients with glaucoma.Methods Thirty-five subjects were enrolled in this prospective study. There were seven patients of chronic primary angle closure glaucoma with bilateral visual field defects, 12 patients of primary open angle glaucoma with bilateral visual field defects, and 16 normal subjects. GE Signa HD 1.5 T magnetic resonance scanners and head eight phased array were used. The mean fractional anisotropy (FA) and the mean average diffusion coefficient (DCavg) of white matter tracts in visual pathway of bilateral optic nerve, optic chiasma, bilateral optic tract，bilateral optic radiation were measured by diffusion tensor imaging, meanwhile the white matter tracts were reconstructed by fiber tracking system. The volumes of lateral geniculate body were measured by coronal proton density weighted magnetic resonance imaging.Results The differences of FA among bilateral optic nerve, optic chiasma, bilateral optic tract，bilateral optic radiation were statistically significant (F=25.985，20.626，12.262，22.399，21.708，24.994，22.774；P＜0.05). There was no difference of DCavg among bilateral optic nerve, optic chiasma, bilateral optic tract，bilateral optic radiation（F=2.097，2.178，0.530，0.983，0.608, 0.866, 1.504；P＞0.05）. The differences of volume of lateral geniculate body among three groups were statistically significant （F=18.631，17.274；P＜0.05）. Conclusion There is degeneration in visual pathway of central nervous system in patients with glaucoma.
Objective To investigate the value of MR diffusion-weighted imaging (DWI) in differentiating pancreatic carcinoma from chronic focal pancreatitis on 3.0 T MR system. Methods Thirteen patients with proved pancreatic carcinoma, 7 patients with confirmed chronic focal pancreatitis, and 14 healthy volunteers, were included in this study. MR examination including the routine abdomen scanning protocol and DWI was performed for both patients and volunteers. The SE-EPI sequence and ASSET technique were used for DWI. The b values of 400, 600, 800 and 1 000 s/mm2 were selected to acquire the DWI. The corresponding apparent diffusion coefficient (ADC) values were measured in each designated region of interest and statistically analyzed. Results ①DWI of the healthy volunteers showed intermediate signals of pancreas. ②DWI of pancreatic tumor masses showed homogenous high signal intensity relative to the surrounding pancreatic tissue with clear boundary. Under different b values, the tumor ADC values were (1.63±0.235)×10-3 mm2/s, (1.42±0.126)×10-3mm2 /s, (1.36±0.170)×10-3 mm2 /s and (1.26±0.178)×10-3 mm2 /s respectively, which were significantly lower than those of non-tumor region 〔(2.11±0.444)×10-3 mm2 /s, (1.83±0.230)×10-3 mm2 /s, (1.81±0.426)×10-3 mm2 /s, (1.60±0.230)×10-3 mm2 /s〕 and of the normal pancreas 〔(1.85±0.350)×10-3 mm2 /s, (1.69±0.290)×10-3 mm2 /s, (1.67±0.268)×10-3 mm2 /s, (1.42±0.221)×10-3 mm2 /s〕， P＜0.05. ③DWI of chronic focal pancreatitis showed inhomogeneous slightly hyper-intense signal with blurring borders. Under different b values, the ADC values of the inflammatory masses of chronic pancreatitis were (169±0.150)×10-3 mm2 /s, (1.56±0.119)×10-3 mm2 /s, (1.59±0.172)×10-3 mm2/s and (1.35±0.080)×10-3 mm2 /s respectively, which were higher than those of pancreatic carcinoma. When b value was set to 800 s/mm2 , the difference in ADC values between pancreatic carcinoma and chronic focal pancreatitis was statistically significant (P＜0.05). Conclusion MR DWI can clearly depict the tumor mass of pancreatic carcinoma. In addition, the measurement of ADC values can provide useful information for the differential diagnosis between pancreatic carcinoma and chronic focal pancreatitis.
When applying deep learning algorithms to magnetic resonance (MR) image segmentation, a large number of annotated images are required as data support. However, the specificity of MR images makes it difficult and costly to acquire large amounts of annotated image data. To reduce the dependence of MR image segmentation on a large amount of annotated data, this paper proposes a meta-learning U-shaped network (Meta-UNet) for few-shot MR image segmentation. Meta-UNet can use a small amount of annotated image data to complete the task of MR image segmentation and obtain good segmentation results. Meta-UNet improves U-Net by introducing dilated convolution, which can increase the receptive field of the model to improve the sensitivity to targets of different scales. We introduce the attention mechanism to improve the adaptability of the model to different scales. We introduce the meta-learning mechanism, and employ a composite loss function for well-supervised and effective bootstrapping of model training. We use the proposed Meta-UNet model to train on different segmentation tasks, and then use the trained model to evaluate on a new segmentation task, where the Meta-UNet model achieves high-precision segmentation of target images. Meta-UNet has a certain improvement in mean Dice similarity coefficient (DSC) compared with voxel morph network (VoxelMorph), data augmentation using learned transformations (DataAug) and label transfer network (LT-Net). Experiments show that the proposed method can effectively perform MR image segmentation using a small number of samples. It provides a reliable aid for clinical diagnosis and treatment.
【Abstract】ObjectiveTo investigate the value of volumetric interpolated breathhold examination (3DVIBE) MRI sequence in the diagnosis of functional islet cell tumors of the pancreas. MethodsDedicated MRI scan was performed for 3 patients suspected to have functional islet cell tumors of the pancreas on clinical and laboratory basis. The MRI scan protocol included routine axial T1W and T2W, coronal true fast imaging with steady state procession (TrueFISP) and MRCP, gadoliniumenhanced 3DVIBE dynamic triphasic acquisitions and enhanced 2D GRE T1W scan. The three phases images of 3DVIBE sequence were acquired at 15 s, 40 s and 65 s after injection of contrast agent, corresponding to the early arterial, late arterial and portal venous phase respectively. The imaging features were compared with surgical and pathological findings. ResultsThe triphasic images of 3DVIBE sequence depicted clearly the morphology of small functional islet cell tumors of the pancreas and reflected accurately the characteristics of tumor blood supply, while other MRI sequences might miss these small lesions. ConclusionThinslice and fast dynamic MRI sequence, as exemplified by 3DVIBE sequence, is very useful in the detection and characterization of pancreatic functional islet cell tumors.
