NIMG-25. IMPROVING THE NONINVASIVE CLASSIFICATION OF GLIOMA GENETIC SUBTYPE WITH DEEP LEARNING AND DIFFUSION-WEIGHTED IMAGING

2021 ◽  
Vol 23 (Supplement_6) ◽  
pp. vi133-vi134
Author(s):  
Julia Cluceru ◽  
Joanna Phillips ◽  
Annette Molinaro ◽  
Yannet Interian ◽  
Tracy Luks ◽  
...  
Keyword(s):  
Deep Learning ◽  
Network Architecture ◽  
Diffusion Weighted Imaging ◽  
Color Image ◽  
Genetic Alterations ◽  
Molecular Subgroup ◽  
Post Contrast ◽  
Diffusion Weighted ◽  
Mutational Status ◽  
Anatomical Images

Abstract In contrast to the WHO 2016 guidelines that use genetic alterations to further stratify patients within a designated grade, new recommendations suggest that IDH mutation status, followed by 1p19q-codeletion, should be used before grade when differentiating gliomas. Although most gliomas will be resected and their tissue evaluated with genetic profiling, non-invasive characterization of genetic subgroup can benefit patients where surgery is not otherwise advised or a fast turn-around is required for clinical trial eligibility. Prior studies have demonstrated the utility of using anatomical images and deep learning to distinguish either IDH-mutant from IDH-wildtype tumors or 1p19q-codeleted from non-codeleted lesions separately, but not combined or using the most recent recommendations for stratification. The goal of this study was to evaluate the effects of training strategy and incorporation of Apparent Diffusion Coefficient (ADC) maps from diffusion-weighted imaging on predicting new genetic subgroups with deep learning. Using 414 patients with newly-diagnosed glioma (split 285/50/49 training/validation/test) and optimized training hyperparameters, we found that a 3-class approach with T1-post-contrast, T2-FLAIR, and ADC maps as inputs achieved the best performance for molecular subgroup classification, with overall accuracies of 86.0%[CI:0.839,1.0], 80.0%[CI:0.720,1.0], and 85.7%[CI:0.771,1.0] on training, validation, and test sets, respectively, and final test class accuracies of 95.2%(IDH-wildtype), 88.9%(IDH-mutated,1p19qintact), and 60%(IDHmutated,1p19q-codeleted). Creating an RGB-color image from 3 MRI images and applying transfer learning with a residual network architecture pretrained on ImageNet resulted in an 8% averaged increase in overall accuracy. Although classifying both IDH and 1p19q mutations together was overall advantageous compared with a tiered structure that first classified IDH mutational status, the 2-tiered approach better generalized to an independent multi-site dataset when only anatomical images were used. Including biologically relevant ADC images improved model generalization to our test set regardless of modeling approach, highlighting the utility of incorporating diffusion-weighted imaging in future multi-site analyses of molecular subgroup.

scholarly journals NIMG-36. AUTOMATIC STRATIFICATION OF ENHANCING AND NON-ENHANCING GLIOMAS INTO GENETIC SUBTYPES USING DEEP NEURAL NETWORKS AND DIFFUSION-WEIGHTED IMAGING

2020 ◽  
Vol 22 (Supplement_2) ◽  
pp. ii155-ii155
Author(s):  
Julia Cluceru ◽  
Paula Alcaide-Leon ◽  
Yannet Interian ◽  
Valentina Pedoia ◽  
Joanna Phillips ◽  
...  
Keyword(s):  
Neural Networks ◽  
A Priori ◽  
Genetic Alterations ◽  
Model Framework ◽  
Molecular Subgroup ◽  
Post Contrast ◽  
Channel Input ◽  
Anatomical Images ◽  
Diffuse Gliomas ◽  
Resected Tissue

Abstract INTRODUCTION Current WHO guidelines emphasize classification of diffuse gliomas by genetic alterations into three subgroups: 1) IDH-wildtype; 2) IDH-mutant, 1p/19q-codeleted; and 3) IDH-mutant, 1p/19q-non-codeleted. Non-invasive genetic characterization can benefit patients with inoperable lesions or who are administered molecularly-targeted therapy before surgery. Prior studies that use anatomical images and convolutional neural networks (CNNs) to distinguish either IDH-mutant from IDH-wildtype tumors, or 1p/19q-codeleted from non-codeleted tumors have resulted in misclassification of nonenhancing IDH-wildtype and enhancing IDH-mutant tumors. This study investigated the benefit of a priori separation of enhancing from nonenhancing lesions and the inclusion of ADC maps from diffusion MRI to genetic subgroup classification. METHODS 3D T2-weighted, T2-FLAIR, and post-contrast T1-weighted images were acquired preoperatively from 254 patients with newly-diagnosed gliomas. IDH1R132H mutations[VJ1] [CJ2], 1p19q-codeletions, ATRX alterations, and p53 mutations were assessed from the resected tissue to determine subtype stratification: IDH-wildtype (n=95), IDH-mutant, 1p/19q-codeleted (n=62), and IDH-mutant, non-codeleted (n=97). 3-channel input images were constructed for each patient using T2-FLAIR, T1-post-contrast, and either T2-weighted or ADC images. Three VGG-16 CNNs pre-trained on ImageNet were re-trained for: 1) lesions without enhancement, 2) enhancing lesions, and 3) all lesions together[VJ3]. RESULTS A network trained on only enhancing lesions predicted the IDH-wildtype subtype with the highest class accuracy (ADC 94%, T2-weighted 100%) compared to using all lesions combined (ADC 90%, T2-weighted 90%). Models trained using non-enhancing lesions and ADC yielded the highest accuracy classifying 1p/19q-codeleted/non-codeleted subgroups (87%/90% for the non-enhancing network vs 83%/81% for combined network). CONCLUSIONS Our results support a strategy that first considers whether a lesion is enhancing when predicting molecular subgroup and includes ADC if the lesion is non-enhancing. Analysis is underway to test this model framework on independent TCIA data.

Effects of contrast-enhancement on diffusion weighted imaging and apparent diffusion coefficient measurements in 3-T magnetic resonance imaging of breast lesions

2017 ◽  
Vol 59 (8) ◽  
pp. 902-908
Author(s):  
Valentina Cipolla ◽  
Daniele Guerrieri ◽  
Giacomo Bonito ◽  
Simone Celsa ◽  
Carlo de Felice
Keyword(s):  
Diffusion Coefficient ◽  
Apparent Diffusion Coefficient ◽  
Diffusion Weighted Imaging ◽  
Breast Lesions ◽  
Post Contrast ◽  
Diffusion Weighted ◽  
Signed Rank ◽  
Apparent Diffusion ◽  
Malignant Breast ◽  
Adc Values

Background The effect of gadolinium-based contrast agents on diffusion-weighted imaging (DWI) measurements of breast lesions is still not clear. Purpose To investigate gadolinium effects on DWI and apparent diffusion coefficient (ADC) in breast lesions and normal parenchyma with 3 Tesla contrast-enhanced MRI. Material and Methods Pre- and post-contrast DWI (b = 0 and b = 1000 s/mm2) were acquired in 47 patients. Measured ADC values, pre- and post-contrast T2 signal intensity (T2 SI) and contrast-to-noise ratio (CNR) were compared with Wilcoxon signed-rank and rank-sum test ( P < 0.05). Results Post-contrast ADC was reduced only in malignant lesions (−34%), T2 SI was reduced both in malignant (−50%) and benign (−36%) lesions. Post-contrast CNR was reduced in all groups except for benign lesions. Conclusion Gadolinium-based contrast agent causes a significant reduction in ADC values of malignant breast lesions.

Cerebral schistosomiasis: diffusion-weighted imaging helps to differentiate from brain glioma and metastasis

2017 ◽  
Vol 58 (11) ◽  
pp. 1371-1377 ◽  
Author(s):  
Jinbai Huang ◽  
Jing Luo ◽  
Jie Peng ◽  
Tao Yang ◽  
Huanghua Zheng ◽  
...  
Keyword(s):  
Brain Metastasis ◽  
Diffusion Weighted Imaging ◽  
Brain Mri ◽  
High Grade ◽  
Post Contrast ◽  
Adc Value ◽  
Diffusion Weighted ◽  
Perifocal Edema ◽  
Adc Values ◽  
High Grade Gliomas

Background Diffusion-weighted imaging (DWI) was introduced into clinical use some years ago. However, its use in the diagnosis of cerebral schistosomiasis has not been reported. Purpose To investigate the ability of the apparent diffusion coefficient (ADC) value of DWI in the diagnosis of cerebral schistosomiasis, and to differentiate it from brain high-grade gliomas and metastasis. Material and Methods Conventional brain MRI with pre-contrast, post-contrast, and DWI was performed on 50 cases of cerebral schistosomiasis, high-grade glioma, and brain metastasis. The ADC values of the three lesions, the proximal and the distal perifocal edema were measured. In order to remove the individual difference effect of ADC values, relative ADC (rADC) values were calculated through dividing the ADC value of the lesion area by that of the contralateral normal white matter. rADC values were used to evaluate the differences among cerebral schistosomiasis, brain high-grade gliomas, and metastasis. Results rADC of cerebral schistosomiasis was significantly lower than rADC of brain metastasis ( P < 0.05), without any significant differences when compared with high-grade gliomas. rADC of proximal perifocal edema in cerebral schistosomiasis was significantly higher than in high-grade gliomas ( P < 0.010), but not different compared with brain metastasis. Conclusion DWI examination with ADC values of lesions and proximal perifocal edema might be helpful in the exact diagnosis of cerebral schistosomiasis.

scholarly journals Automatic Assessment of ASPECTS Using Diffusion-Weighted Imaging in Acute Ischemic Stroke Using Recurrent Residual Convolutional Neural Network

Diagnostics ◽  
2020 ◽  
Vol 10 (10) ◽  
pp. 803
Author(s):  
Luu-Ngoc Do ◽  
Byung Hyun Baek ◽  
Seul Kee Kim ◽  
Hyung-Jeong Yang ◽  
Ilwoo Park ◽  
...  
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Convolutional Neural Network ◽  
Diffusion Weighted Imaging ◽  
Learning Algorithm ◽  
Binary Classification ◽  
Automatic Assessment ◽  
Anterior Circulation ◽  
Deep Learning Algorithm ◽  
Diffusion Weighted

The early detection and rapid quantification of acute ischemic lesions play pivotal roles in stroke management. We developed a deep learning algorithm for the automatic binary classification of the Alberta Stroke Program Early Computed Tomographic Score (ASPECTS) using diffusion-weighted imaging (DWI) in acute stroke patients. Three hundred and ninety DWI datasets with acute anterior circulation stroke were included. A classifier algorithm utilizing a recurrent residual convolutional neural network (RRCNN) was developed for classification between low (1–6) and high (7–10) DWI-ASPECTS groups. The model performance was compared with a pre-trained VGG16, Inception V3, and a 3D convolutional neural network (3DCNN). The proposed RRCNN model demonstrated higher performance than the pre-trained models and 3DCNN with an accuracy of 87.3%, AUC of 0.941, and F1-score of 0.888 for classification between the low and high DWI-ASPECTS groups. These results suggest that the deep learning algorithm developed in this study can provide a rapid assessment of DWI-ASPECTS and may serve as an ancillary tool that can assist physicians in making urgent clinical decisions.

scholarly journals Diffusion-Weighted Imaging and Post-contrast Enhancement in Differentiating Optic Neuritis and Non-arteritic Anterior Optic Neuropathy

2017 ◽  
Vol 42 (2) ◽  
pp. 90-98 ◽  
Author(s):  
Ore-ofe O. Adesina ◽  
J. Scott McNally ◽  
Karen L. Salzman ◽  
Bradley J. Katz ◽  
Judith E. A. Warner ◽  
...  
Keyword(s):  
Optic Neuritis ◽  
Contrast Enhancement ◽  
Optic Neuropathy ◽  
Diffusion Weighted Imaging ◽  
Post Contrast ◽  
Diffusion Weighted

MRI findings, including diffusion-weighted imaging, in seven cats with nasal lymphoma and two cats with nasal adenocarcinoma

2020 ◽  
pp. 1098612X2093281
Author(s):  
Toshiyuki Tanaka ◽  
Kazuna Ashida ◽  
Yasumasa Iimori ◽  
Hiroki Yamazaki ◽  
Keiichiro Mie ◽  
...  
Keyword(s):  
Diffusion Weighted Imaging ◽  
Case Series ◽  
Mri Findings ◽  
Post Contrast ◽  
Adc Value ◽  
Diffusion Weighted ◽  
Adc Values ◽  
Nasal Adenocarcinoma ◽  
Nasal Lymphoma ◽  
Heterogeneous Enhancement

Objectives Case series summary Primary nasal tumours in cats are rare, with lymphoma being the most common feline nasal tumour, followed by adenocarcinoma. Although CT can reliably detect feline nasal tumours, there are no specific CT features that identify each tumour type. To our knowledge, there have been no reports describing MRI findings, including diffusion-weighted imaging (DWI), for nasal lymphomas and adenocarcinomas in cats. Therefore, this retrospective study aimed to evaluate the MRI findings of nasal lymphoma and adenocarcinoma, including qualitative and quantitative analysis of DWI. Methods MRI examination was performed on seven cats with histologically confirmed lymphoma and on two with adenocarcinoma. The MRI protocol included T2-weighted imaging (T2WI), T1-weighted imaging (T1WI) and DWI. Apparent diffusion coefficient (ADC) values were measured using DWI. Contrast agent was not used in one cat with lymphoma. Results In those with lymphoma, three (43%) were iso- and hyperintense on T2WI, seven (100%) were iso-intense on T1WI, five (83%) exhibited mild heterogeneous enhancement, including a prominent region of non-enhancement on post-contrast T1WI, and seven (100%) cats exhibited hyperintensity on DWI. The median ADC values were 0.45 × 10−3 mm2/s (range 0.37–0.53 × 10−3 mm2/s). For adenocarcinoma, two (100%) were iso- and hyperintense on T2WI, two (100%) were iso-intense on T1WI, two (100%) exhibited marked heterogeneous enhancement on post-contrast T1WI and two (100%) were iso-intense on DWI. The median ADC values were 1.08 × 10−3 mm2/s (range 0.88–1.27 × 10−3 mm2/s). The median ADC values of lymphoma tended to be lower than adenocarcinoma ( P = 0.056). Conclusions and relevance Determining ADC value and tumours with a large area of non-enhancement may be helpful in differentiating nasal lymphoma from nasal adenocarcinoma.

scholarly journals Magnetic resonance texture analysis utility in differentiating intraparenchymal neurosarcoidosis from primary central nervous system lymphoma: a preliminary analysis

2019 ◽  
Vol 32 (3) ◽  
pp. 203-209 ◽  
Author(s):  
Girish Bathla ◽  
Neetu Soni ◽  
Raymondo Endozo ◽  
Balaji Ganeshan
Keyword(s):  
Central Nervous System ◽  
Nervous System ◽  
Diffusion Coefficient ◽  
Magnetic Resonance ◽  
Spatial Scale ◽  
Texture Analysis ◽  
Diffusion Weighted Imaging ◽  
Post Contrast ◽  
Diffusion Weighted ◽  
Apparent Diffusion

Purpose Neurosarcoidosis and primary central nervous system lymphomas, although distinct disease entities, can both have overlapping neuroimaging findings. The purpose of our preliminary study was to assess if magnetic resonance texture analysis can differentiate parenchymal mass-like neurosarcoidosis granulomas from primary central nervous system lymphomas. Methods A total of nine patients was evaluated, four with parenchymal neurosarcoidosis granulomas and five with primary central nervous system lymphomas. Magnetic resonance texture analysis was performed with commercial software using a filtration histogram technique. Texture features of different sizes and variations in signal intensity were extracted at six different spatial scale filters, followed by feature quantification using statistical and histogram parameters and 36 features were analysed for each sequence (T1-weighted, T2-weighted, fluid-attenuated inversion recovery, diffusion-weighted, apparent diffusion coefficient, T1-post contrast). The non-parametric Mann–Whitney test was used to evaluate the differences between different texture parameters. Results The differences in distribution of entropy on T2-weighted imaging, apparent diffusion coefficient and T1-weighted post-contrast images were statistically significant on all spatial scale filters. Magnetic resonance texture analysis using medium and coarse spatial scale filters was especially useful in discriminating neurosarcoidosis from primary central nervous system lymphomas for mean, mean positive pixels, kurtosis, and skewness on diffusion-weighted imaging ( P < 0.004–0.030). At spatial scale filter 5, entropy on T2-weighted imaging ( P = 0.001) was the most useful discriminator with a cut-off value of 6.12 ( P = 0.001, area under the curve (AUC)-1, sensitivity (Sn)-100%, specificity (Sp)-100%), followed by kurtosis and skewness on diffusion-weighted imaging with a cut-off value of −0.565 ( P = 0.011, AUC-0.97, Sn-100%, Sp-83%) and–0.365 ( P = 0.008, AUC-0.98, Sn-100%, Sp-100%) respectively. Conclusion Filtration histogram-based magnetic resonance texture analysis appears to be a promising modality to distinguish parenchymal neurosarcoidosis granulomas from primary central nervous system lymphomas.

Sign in / Sign up

Export Citation Format

Share Document