scholarly journals COMP-03. QUANTITATIVE IMAGE FEATURE ANALYSIS IN DIFFUSE GLIOMA – A VALUABLE MR IMAGING BIOMARKER FOR PREOPERATIVE IDH MUTATION CLASSIFICATION

2019 ◽  
Vol 21 (Supplement_6) ◽  
pp. vi61-vi61
Author(s):  
Eric Carver ◽  
James Snyder ◽  
Brent Griffith ◽  
Ning Wen
Keyword(s):  
Mr Imaging ◽  
Image Features ◽  
Image Feature ◽  
Feature Analysis ◽  
Imaging Features ◽  
Idh Mutation ◽  
Mr Images ◽  
Mutation Status ◽  
Non Invasive ◽  
Quantitative Image

Abstract INTRODUCTION Pre-operative differentiation of IDH mutant gliomas from similar appearing pathologies on imaging prior to definitive surgical diagnosis may aid treatment navigation, maximize the surgical approach, and provide diagnostic support for inoperable tumors. Quantitative image feature analysis offers a potential non-invasive method to identify diagnostic, prognostic, and predictive imaging biomarkers. We investigated the use of radiomic MR imaging features to classify tumors based on IDH mutation status. METHOD Pre-operative T1-weighted (T1W), T2-weighted (T2W), T1-contrast enhanced (T1CE), and fluid attenuated inversion recovery (FLAIR) MR brain images, along with patient IDH mutation status (mutant/wildtype) were obtained for 128 glioma patients from The Cancer Genome Atlas (TCGA). Enhancing tumor was delineated by GLISTRboost. GlistrBoost is a hybrid-discriminative model that segments tumors based on an expectation-maximization framework with a classification scheme and uses a probabilistic Bayesian strategy for segmentation refinement. MR studies for 78 glioma patients from six institutions were used for training and 50 glioma patients from a different institution were used for validation. Pre-processing included registration, resampling, and normalization. Cancer Imaging Phenomics Toolkit (CaPTK) extracted 938 radiomic image features per sequence for the enhancing tumor contour. Relevance of each individual feature was determined by the least absolute shrinkage and selection operator (LASSO). The ability of relevant radiomic image features to identify mutation status of IDH was assessed by logistic regression. RESULTS LASSO identified one highly informative radiomic imaging feature, the minimum of the mean absolute histogram deviation on T1 MR images, which was able to predict IDH mutation status with an accuracy of 0.74, precision of 1.0, and recall of 0.32. CONCLUSION Non-invasive prediction of IDH mutation status from pre-surgical MR images offers potential diagnostic, therapeutic, and prognostic benefits for glioma patients. Quantitative image feature analysis is a feasible method for identifying potential radiomic imaging features.

Hemorrhagic and nonhemorrhagic Rathke cleft cysts mimicking pituitary apoplexy

2008 ◽  
Vol 108 (1) ◽  
pp. 3-8 ◽  
Author(s):  
Mandy J. Binning ◽  
James K. Liu ◽  
John Gannon ◽  
Anne G. Osborn ◽  
William T. Couldwell
Keyword(s):  
Mr Imaging ◽  
Pituitary Tumor ◽  
Pituitary Apoplexy ◽  
Pituitary Tumors ◽  
Imaging Features ◽  
Mr Images ◽  
Acute Hemorrhage ◽  
Consistent Finding ◽  
Imaging Characteristics ◽  
Intracystic Hemorrhage

Object Rathke cleft cysts (RCCs) are infrequently symptomatic, and apoplexy is one of the most unusual presentations. Only a few cases of apoplexy associated with RCCs have been reported, and their clinical, imaging, surgical, and pathological features are poorly understood. In the cases that have been reported, intracystic hemorrhage has been a consistent finding. The authors report 6 cases of RCCs in which the presenting clinical and imaging features indicated pituitary apoplexy, both with and without intracystic hemorrhage. Methods The authors retrospectively reviewed charts and magnetic resonance (MR) imaging studies obtained in patients who underwent transsphenoidal surgery for RCC. Six patients were identified who presented with symptoms and MR imaging characteristics consistent with pituitary apoplexy but were found intraoperatively to have an RCC. All 6 patients presented with a sudden headache, 2 with visual loss, and 1 with diplopia. Review of the preoperative MR images demonstrated mixed signal intensities in the sellar masses suggestive of a hemorrhagic pituitary tumor. In all patients there was a presumed clinical diagnosis of pituitary tumor apoplexy and an imaging-documented diagnosis of hemorrhagic pituitary tumor. Results All 6 patients underwent transsphenoidal resection to treat the suspected pituitary apoplexy. Intraoperative and histopathological findings were consistent with the diagnosis of an RCC in all cases. Only 2 cases showed evidence of hemorrhage intraoperatively. In all cases, an intracystic nodule was found within the RCC at surgery, and this intracystic nodule was present on the initial MR imaging when retrospectively reviewed. The imaging characteristics of the intracystic nodules were similar to those of acute hemorrhage seen in cases of pituitary apoplexy. Conclusions The clinical and imaging features of RCCs appear similar to those of hemorrhagic pituitary tumors, making them often indistinguishable from pituitary apoplexy.

scholarly journals Differential Diagnosis of Sinonasal Extramedullary Plasmacytoma, Non-Hodgkin’s Lymphoma, and Squamous Cell Carcinoma Using CT Scan and MRI

2021 ◽  
Vol 18 (2) ◽  
Author(s):  
Dong-Joo Lee ◽  
Sang Duk Hong ◽  
Myeong Sang Yu ◽  
Sung Jae Heo ◽  
Joo-Yeon Kim ◽  
...  
Keyword(s):  
Squamous Cell Carcinoma ◽  
Mr Imaging ◽  
Cell Carcinoma ◽  
Tumor Volume ◽  
Cell Lymphoma ◽  
Bone Destruction ◽  
Extramedullary Plasmacytoma ◽  
Imaging Features ◽  
Mr Images ◽  
Waldeyer’S Ring

Background: The imaging features of sinonasal extramedullary plasmacytoma (EMP) are non-specific and similar to those of other lesions, such as sinonasal non-Hodgkin’s lymphoma (NHL) and squamous cell carcinoma (SCC). Objectives: To analyze the computed tomography (CT) and magnetic resonance (MR) images of patients with EMP, NHL, and SCC to identify the radiological characteristics differentiating sinonasal EMP from NHL and SCC. Patients and Methods: In this cross-sectional study, the CT and MR imaging features of 37 patients with sinonasal EMP, 46 patients with NHL, and 44 patients with SCC were analyzed. Sinonasal NHL was categorized into two distinct types, namely, natural killer/T-cell lymphoma (n = 32) and diffuse large B-cell lymphoma (n = 14). The tumor volume was determined by measuring the region of interest (ROI) in the PACS program. Besides, homogeneity, apparent diffusion coefficient (ADC) in the ADC maps, degree of enhancement, adjacent bone destruction, and invasion to Waldeyer’s ring and cervical or retropharyngeal lymph nodes were evaluated. Results: Although the tumor volume was larger in the EMP group as compared to the NHL and SCC groups, the difference was not statistically significant. The NHL group showed the highest tumor homogeneity on both CT and MR images. EMP was more heterogenous than NHL, with moderate signal intensity on T1-weighted MR images. On the other hand, EMP and NHL showed significantly lower ADCs as compared to SCC. The majority of patients with sinonasal EMP, NHL, and SCC showed an avid enhancement. Also, destructive tumor growth involving the adjacent bone was more frequent in SCC than in EMP or NHL. However, there were no significant differences among sinonasal EMP, NHL, and SCC in terms of invasion to Waldeyer’s ring and cervical or retropharyngeal lymph node metastasis. Conclusion: Marked heterogeneity on T1-weighted images, low ADCs, and lack of adjacent bone destruction were the CT and MR imaging features that favored the diagnosis of EMP over NHL or SCC.

scholarly journals Image Feature Analysis and Dynamic Measurement of Plantar Pressure Based on Fusion Feature Extraction

2021 ◽  
Vol 38 (6) ◽  
pp. 1829-1835
Author(s):  
Ji Zou ◽  
Chao Zhang ◽  
Zhongjing Ma ◽  
Lei Yu ◽  
Kaiwen Sun ◽  
...  
Keyword(s):  
Feature Extraction ◽  
Plantar Pressure ◽  
Dynamic Measurement ◽  
Image Features ◽  
Image Feature ◽  
Feature Analysis ◽  
Fusion Algorithm ◽  
Massive Datasets ◽  
Processing Power ◽  
Fusion Feature

Footprint recognition and parameter measurement are widely used in fields like medicine, sports, and criminal investigation. Some results have been achieved in the analysis of plantar pressure image features based on image processing. But the common algorithms of image feature extraction often depend on computer processing power and massive datasets. Focusing on the auxiliary diagnosis and treatment of foot rehabilitation of foot laceration patients, this paper explores the image feature analysis and dynamic measurement of plantar pressure based on fusion feature extraction. Firstly, the authors detailed the idea of extracting image features with a fusion algorithm, which integrates wavelet transform and histogram of oriented gradients (HOG) descriptor. Next, the plantar parameters were calculated based on plantar pressure images, and the measurement steps of plantar parameters were given. Finally, the feature extraction effect of the proposed algorithm was verified, and the measured results on plantar parameters were obtained through experiments.

Resection of insular gliomas: the importance of lenticulostriate artery position

2008 ◽  
Vol 109 (5) ◽  
pp. 825-834 ◽  
Author(s):  
Yaron A. Moshel ◽  
Joshua D.S. Marcus ◽  
Erik C. Parker ◽  
Patrick J. Kelly
Keyword(s):  
Mr Imaging ◽  
Surgical Techniques ◽  
Brain Parenchyma ◽  
Growth Patterns ◽  
Normal Brain ◽  
Imaging Features ◽  
Mr Images ◽  
Presenting Symptoms ◽  
Group I ◽  
Neurological Morbidity

Object The object of this study was to identify characteristic preoperative angiographic and MR imaging features of safely resectable insular gliomas and describe the surgical techniques and postoperative clinical outcomes. Methods Thirty-eight patients with insular gliomas underwent transsylvian resection between 1995 and 2007. Patient demographics, presenting symptoms, pathological findings, and neurological outcomes were retrospectively reviewed. Preoperative MR imaging–defined tumor volumes were superimposed onto the preoperative stereotactic cerebral angiograms to determine whether the insular tumor was confined lateral to (Group I) or extended medially around (Group II) the lenticulostriate arteries (LSAs). Results Twenty-five patients (66%) had tumors situated lateral to the LSAs and 13 (34%) had tumors encasing the LSAs. Insular gliomas situated lateral to the LSAs led to significant medial displacement of these vessels (161 ± 39%). In 20 (80%) of these 25 cases the boundaries between tumor and brain parenchyma were well demarcated on preoperative T2-weighted MR images. In contrast, there was less displacement of the LSAs (130 ± 14%) in patients with insular gliomas extending around the LSAs on angiography. In 11 (85%) of these 13 cases, the tumor boundaries were diffuse on T2-weighted MR images. Postoperative hemiparesis or worsening of a preexisting hemiparesis, secondary to LSA compromise, occurred in 5 patients, all of whom had tumor volumes that extended medial to the LSAs. Gross-total or near-total resection was achieved more frequently in cases in which the insular glioma remained lateral to the LSAs (84 vs 54%). Conclusions Insular gliomas with an MR imaging–defined tumor volume located lateral to the LSAs on stereotactic angiography displace the LSAs medially by expanding the insula, have well-demarcated tumor boundaries on MR images, and can be completely resected with minimal neurological morbidity. In contrast, insular tumors that appear to surround the LSAs do not displace these vessels medially, are poorly demarcated from normal brain parenchyma on MR images, and are associated with higher rates of neurological morbidity if aggressive resection is pursued. Preoperative identification of these anatomical growth patterns can be of value in planning resection.

scholarly journals Peroneal and tibial intraneural ganglia: correlation between intraepineurial compartments observed on magnetic resonance images and the potential importance of these compartments

2007 ◽  
Vol 22 (6) ◽  
pp. 1-10 ◽  
Author(s):  
Robert J. Spinner ◽  
Kimberly K. Amrami ◽  
Diana Angius ◽  
Huan Wang ◽  
Stephen W. Carmichael
Keyword(s):  
Magnetic Resonance ◽  
Sciatic Nerve ◽  
Mr Imaging ◽  
Tibial Nerve ◽  
Cyst Formation ◽  
Magnetic Resonance Images ◽  
Imaging Features ◽  
Mr Images ◽  
Superior Tibiofibular Joint ◽  
Tibiofibular Joint

Object Previously the authors demonstrated that peroneal and tibial intraneural ganglia arising from the superior tibiofibular joint may occasionally extend proximally within the epineurium to reach the sciatic nerve. The dynamic nature of these cysts, dependent on intraarticular pressures, may give rise to differing clinical and imaging presentations that have remained unexplained until now. To identify the pathogenesis of these unusual cysts and to correlate their atypical magnetic resonance (MR) imaging appearance, the authors retrospectively reviewed their own experience as well as the published literature on these types of intraneural ganglia. Methods A careful review of MR images obtained in 22 patients with intraneural ganglia located about the knee region (18 peroneal and four tibial intraneural ganglia) allowed the authors to substantiate three different patterns: outer (epifascicular) epineurial (20 cases); inner (interfascicular) epineurial (one case); and combined outer and inner epineurial (one case). In these cases serial MR images allowed the investigators to track the movement of the cyst within the same layer of the epineurium. All lesions had connections to the superior tibiofibular joint. Nine patients were identified as having lesions with sciatic nerve extension. Seven patients harboring an outer epineurial cyst (six in whom the cyst involved the peroneal nerve and one in whom it involved the tibial nerve) had signs of sciatic nerve cross-over, with the cyst seen in the sciatic nerve and/or other terminal branches. In only two of these cases had the cyst previously been recognized to have sciatic nerve involvement. In contrast, in one case an inner epineurial cyst involving the tibial nerve ascended within the tibial division of the sciatic nerve and did not cross over. A single patient had a combination of both outer and inner epineurial cysts; these were easily distinguished by their distinctive imaging patterns. Conclusions This anatomical compartmentalization of intraneural cysts can be used to explain varied clinical and imaging patterns of cleavage planes for cyst formation and propagation. Compartmentalization elucidates the mechanism for cases of outer epineurial cysts in which there are primary ascent, sciatic cross-over, and descent of the lesion down terminal branches; correlates these cysts' atypical MR imaging features; and contrasts a different pattern of inner epineurial cysts in which ascent and descent occur without cross-over. The authors present data demonstrating that the dynamic phases of these intraneural ganglia frequently involve the sciatic nerve. Their imaging features are subtle and serve to explain the underrecognition and underreporting of the longitudinal extension of these cysts. Importantly, cysts extending to the sciatic nerve are still derived from the superior tibiofibular joint. Combined with the authors' previous experimental data, the current observations help the reader understand intraneural ganglia with a different, deeper degree of anatomical detail.

scholarly journals A Novel Fully Automated Mri-Based Deep Learning Method for Classification of Idh Mutation Status in Brain Gliomas

2019 ◽  
Author(s):  
Chandan Ganesh Bangalore Yogananda ◽  
Bhavya R Shah ◽  
Maryam Vejdani-Jahromi ◽  
Sahil S Nalawade ◽  
Gowtham K Murugesan ◽  
...  
Keyword(s):  
Deep Learning ◽  
Cross Validation ◽  
Brain Mri ◽  
The Cancer Genome Atlas ◽  
Tumor Segmentation ◽  
Idh Mutation ◽  
Mr Images ◽  
Post Contrast ◽  
Mutation Status ◽  
Brain Gliomas

Abstract Background Isocitrate dehydrogenase (IDH) mutation status has emerged as an important prognostic marker in gliomas. Currently, reliable IDH mutation determination requires invasive surgical procedures. The purpose of this study was to develop a highly-accurate, MRI-based, voxel-wise deep-learning IDH-classification network using T2-weighted (T2w) MR images and compare its performance to a multi-contrast network. Methods Multi-parametric brain MRI data and corresponding genomic information were obtained for 214 subjects (94 IDH-mutated, 120 IDH wild-type) from The Cancer Imaging Archive (TCIA) and The Cancer Genome Atlas (TCGA). Two separate networks were developed including a T2w image only network (T2-net) and a multi-contrast (T2w, FLAIR, and T1 post-contrast) network (TS-net) to perform IDH classification and simultaneous single label tumor segmentation. The networks were trained using 3D-Dense-UNets. Three-fold cross-validation was performed to generalize the networks’ performance. ROC analysis was also performed. Dice-scores were computed to determine tumor segmentation accuracy. Results T2-net demonstrated a mean cross-validation accuracy of 97.14% ±0.04 in predicting IDH mutation status, with a sensitivity of 0.97 ±0.03, specificity of 0.98 ±0.01, and an AUC of 0.98 ±0.01.  TS-net achieved a mean cross-validation accuracy of 97.12% ±0.09, with a sensitivity of 0.98 ±0.02, specificity of 0.97 ±0.001, and an AUC of 0.99 ±0.01. The mean whole tumor segmentation Dice-scores were 0.85 ±0.009 for T2-net and 0.89 ±0.006 for TS-net. Conclusion We demonstrate high IDH classification accuracy using only T2-weighted MR images. This represents an important milestone towards clinical translation.

Sign in / Sign up

Export Citation Format

Share Document