scholarly journals Evaluation of a CTA-based convolutional neural network for infarct volume prediction in anterior cerebral circulation ischaemic stroke

2021 ◽  
Vol 5 (1) ◽  
Author(s):  
Lasse Hokkinen ◽  
Teemu Mäkelä ◽  
Sauli Savolainen ◽  
Marko Kangasniemi
Keyword(s):  
Neural Network ◽  
Thrombolytic Therapy ◽  
Supportive Care ◽  
Convolutional Neural Network ◽  
Ct Perfusion ◽  
Infarct Volume ◽  
University Hospital ◽  
Current Guideline ◽  
Infarct Core

Abstract Background Computed tomography angiography (CTA) imaging is needed in current guideline-based stroke diagnosis, and infarct core size is one factor in guiding treatment decisions. We studied the efficacy of a convolutional neural network (CNN) in final infarct volume prediction from CTA and compared the results to a CT perfusion (CTP)-based commercially available software (RAPID, iSchemaView). Methods We retrospectively selected 83 consecutive stroke cases treated with thrombolytic therapy or receiving supportive care that presented to Helsinki University Hospital between January 2018 and July 2019. We compared CNN-derived ischaemic lesion volumes to final infarct volumes that were manually segmented from follow-up CT and to CTP-RAPID ischaemic core volumes. Results An overall correlation of r = 0.83 was found between CNN outputs and final infarct volumes. The strongest correlation was found in a subgroup of patients that presented more than 9 h of symptom onset (r = 0.90). A good correlation was found between the CNN outputs and CTP-RAPID ischaemic core volumes (r = 0.89) and the CNN was able to classify patients for thrombolytic therapy or supportive care with a 1.00 sensitivity and 0.94 specificity. Conclusions A CTA-based CNN software can provide good infarct core volume estimates as observed in follow-up imaging studies. CNN-derived infarct volumes had a good correlation to CTP-RAPID ischaemic core volumes.

Abstract WP49: Defining the Acute Infarct Core with CT Perfusion in Ischemic Stroke: Physiological and Technical Issues

Stroke ◽  
2013 ◽  
Vol 44 (suppl_1) ◽  
Author(s):  
Christopher D d’Esterre ◽  
Enrico Fainardi ◽  
Ting Yim Lee
Keyword(s):  
Optimal Parameter ◽  
Ct Perfusion ◽  
Infarct Volume ◽  
Mean Transit Time ◽  
Reactive Hyperemia ◽  
Poor Correlation ◽  
Infarct Core ◽  
Ischemic Lesion ◽  
Ischemic Time ◽  
Acute Infarct

Background: CT Perfusion (CTP) defined hemodynamic parameters used to delineate admission infarct core can be affected by truncated data acquisition, recanalization status and reactive hyperemia. We determined the optimal CTP parameter for infarct demarcation while taking these variables into account. Methods: 30 patients had CTP/NCCT scanning within 6 hours of ictus, a 24 hour CTA and an NCCT at 3 months post stroke to define final infarct. Patients were analyzed according to: 1) the percent wash out (truncation) of the ischemic time density curve (ITDC) and 2) recanalization status defined using the 24 hour CTA. CTP functional maps were generated using delay insensitive CTP software (GE Healthcare). For all patients, the total ischemic lesion (infarct+penumbra+benign oligemia) was defined using the contrast delay plus mean transit time map. Cerebral blood flow (CBF), cerebral blood volume (CBV) and the product of CBF and CBV (CBFxCBV) were used to define the infarct core defect, according to established thresholds, and compared with the infarct volume defined on the 3 month NCCT. The coefficients of correlation (R2) of linear regression models were used for the comparisons. Results: R2 values for admission CBF, CBV, and CBFxCBV defect versus final infarct volume for patients with and without truncation of the ITDC were 0.89, 0.49, 0.65 and 0.90, 0.42, 0.68, respectively; while R2 values for patients with and without recanalization at 24 hours were 0.73, 0.33, 0.44 and 0.84, 0.54, 0.45, respectively. In addition, for the recanalization group with and without truncation of the ITDC, R2 for CBF, CBV, CBFxCBV versus final infarct volume were 0.73, 0.12, 0.31 and 0.79, 0.58, 0.56, respectively. Hyperemia, defined as an increase in CBV relative to the contralateral hemisphere, was observed in 30% of patients. Both hyperemia and ITDC truncation led to poor correlation between the CBV defect and NCCT defined infarct volume. Conclusion: CBF is the optimal parameter for determining the size of the acute infarct core as it is not affected by truncation of the ITDC and autoregulatory vasodilation causing reactive hyperemia.

scholarly journals CTA Source Images as a Predictor of Final Infarct Volume are Time-Dependent

2010 ◽  
Vol 37 (6) ◽  
pp. 866-867 ◽  
Author(s):  
Dylan Blacquiere ◽  
Miguel Bussière ◽  
Cheemun Lum ◽  
Dar Dowlatshahi
Keyword(s):  
Acute Stroke ◽  
Ct Angiography ◽  
Ct Perfusion ◽  
Infarct Volume ◽  
Artery Occlusion ◽  
Time Dependent ◽  
Infarct Core ◽  
Time Resolved

Avascularity on CT angiography source images (CTASI) may better predict final infarct volume in acute stroke as compared to early ischemic changes on non-contract CT. These CTASI findings may represent infarct core and help determine the extent of salvageable tissue. However, the extent of avascularity on CTASI may overestimate infarct volume if transit of contrast is prolonged due to proximal artery occlusion. We present a case where CT-perfusion (CTP) and time-resolved CT-angiography (CTA) identified salvageable tissue thought to be infarcted on CTASI.

Abstract 38: Brain CT Perfusion is Superior to Non-contrast CT Aspects in the Evaluation of Hyperacute Infarct Volume

Stroke ◽  
2017 ◽  
Vol 48 (suppl_1) ◽  
Author(s):  
Jelle Demeestere ◽  
Carlos Garcia-Esperon ◽  
Pablo Garcia-Bermejo ◽  
Fouke Ombelet ◽  
Patrick McElduff ◽  
...  
Keyword(s):  
Computed Tomography ◽  
Ischemic Stroke ◽  
Ct Perfusion ◽  
Infarct Volume ◽  
Lesion Volume ◽  
Infarct Core ◽  
Anterior Circulation ◽  
C Statistic ◽  
Contrast Ct ◽  
Core Volume

Objective: To compare the predictive capacity to detect established infarct in acute anterior circulation stroke between the Alberta Stroke Program Early Computed Tomography Score (ASPECTS) on non-contrast computed tomography (CT) and CT perfusion. Methods: Fifty-nine acute anterior circulation ischemic stroke patients received brain non-contrast CT, CT perfusion and hyperacute magnetic resonance imaging (MRI) within 100 minutes from CT imaging. ASPECTS scores were calculated by 4 independent vascular neurologists, blinded from CT perfusion and MRI data. CT perfusion infarct core volumes were calculated by MIStar software. The accuracy of commonly used ASPECTS cut-off scores and a CT perfusion core volume of ≥ 70 mL to detect a hyperacute MRI diffusion lesion of ≥ 70 ml was evaluated. Results: Median ASPECTS score was 9 (IQR 7-10). Median CT perfusion core volume was 22 ml (IQR 10.4-71.9). Median MRI diffusion lesion volume was 24,5 ml (IQR 10-63.9). ASPECTS score of < 6 had a sensitivity of 0.37, specificity of 0.95 and c-statistic of 0.66 to predict an acute MRI lesion ≥ 70 ml. In comparison, a CT perfusion core lesion of ≥ 70 ml had a sensitivity of 0.76, specificity of 0.98 and c-statistic of 0.92. The CT perfusion core lesion covered a median of 100% of the acute MRI lesion volume (IQR 86-100%). Conclusions: CT perfusion is superior to ASPECTS to predict hyperacute MRI lesion volume in ischemic stroke.

Use of a novel convolutional neural network-based mammographic evaluation to assess response to adjuvant endocrine therapy in women with early-stage breast cancer.

2021 ◽  
Vol 39 (15_suppl) ◽  
pp. 530-530
Author(s):  
Julia Elizabeth McGuinness ◽  
Vicky Ro ◽  
Simukayi Mutasa ◽  
Richard Ha ◽  
Katherine D. Crew
Keyword(s):  
Breast Cancer ◽  
Neural Network ◽  
Prognostic Factors ◽  
Convolutional Neural Network ◽  
Endocrine Therapy ◽  
Risk Score ◽  
Early Stage ◽  
Stage I ◽  
Risk Scores

530 Background: The standard of care for early-stage hormone receptor (HR)-positive breast cancer (BC) is 5-10 years of adjuvant endocrine therapy (ET), which leads to a 50-60% relative risk reduction in BC recurrence. However, 10-40% of patients may relapse up to 20 years (y) after diagnosis, and there is a need for biomarkers of response to ET. We developed a novel, fully-automated convolutional neural network (CNN)-based mammographic evaluation that accurately predicts BC risk, which is being evaluated as a pharmacodynamic response biomarker to adjuvant ET. Methods: We conducted a retrospective cohort study among women with HR-positive stage I-III unilateral BC diagnosed at Columbia University Irving Medical Center from 2007-2017, who received adjuvant ET and had at least 2 mammograms of the contralateral breast (baseline and annual follow-up). Demographics, clinical characteristics, BC treatments, and relapse status were extracted from the electronic health record and New York-Presbyterian Hospital Tumor Registry. We performed CNN analysis of mammograms at baseline (start of ET) and annual follow-up. Our primary endpoint was change in CNN risk score, expressed as a continuous variable (range, 0-1). We used two-sample t-tests to assess for differences in mean CNN scores between patients who relapsed or remained in remission. We evaluated if CNN score at baseline and change from baseline were associated with relapse using logistic regression, with adjustment for known prognostic factors. Results: Among 870 evaluable women, mean age at diagnosis was 59.5y (standard deviation [SD], 12.4); 60.3% had stage I tumors, 72.6% underwent lumpectomy, and 45.8% received chemotherapy. With a median follow-up of 4.9y, there were 68 (7.9%) breast cancer relapses (36 distant, 26 local, 6 new primary). Median number of evaluable mammograms per patient was 5 (range, 2-13). Mean baseline CNN risk scores were significantly higher among women who relapsed compared to those in remission (0.258 vs 0.237, p = 0.022), which remained significant after adjustment for known prognostic factors. There was a significant difference in mean absolute change in CNN risk score from baseline to 1y follow-up between those who relapsed vs. remained in remission (0.001 vs. -0.022, p = 0.027), but this was no longer significant in multivariable analysis. Conclusions: We demonstrated that higher baseline CNN risk score was an independent predictor of BC relapse. A greater decrease in mean CNN risk scores at 1-year follow-up after initiating adjuvant ET was seen among BC patients who remained in remission compared to those who relapsed. Therefore, baseline CNN risk scores may identify patients at high-risk for breast cancer recurrence to target for more intensive adjuvant treatment. Early changes in CNN risk scores may be used to predict response to long-term ET in the adjuvant setting.

A Fully-Automatic Segmentation of the Carpal Tunnel from Magnetic Resonance Images Based on the Convolutional Neural Network-Based Approach

2021 ◽  
Author(s):  
Tai-Hua Yang ◽  
Cheng-Wei Yang ◽  
Yung-Nien Sun ◽  
Ming-Huwi Horng
Keyword(s):  
Neural Network ◽  
Magnetic Resonance ◽  
Convolutional Neural Network ◽  
Median Nerve ◽  
Carpal Tunnel ◽  
Flexor Tendon ◽  
Magnetic Resonance Images ◽  
University Hospital ◽  
Feature Maps ◽  
Flexor Tendons

Abstract Purpose Carpal tunnel syndrome is one of the common peripheral neuropathies. For magnetic resonance imaging, segmentation of the carpal tunnel and its contents, including flexor tendons and the median nerve for magnetic resonance images is an important issue. In this study, a convolutional neural network (CNN) model, which was modified by the original DeepLabv3 + model to segment three primary structures of the carpal tunnel: the carpal tunnel, flexor tendon, and median nerve. Methods To extract important feature maps for segmentation of the carpal tunnel, flexor tendon, and median nerve, the proposed CNN model termed modified DeepLabv3 + uses DenseNet-121 as a backbone and adds dilated convolution to the original spatial pyramid pooling module. A MaskTrack method was used to refine the segmented results generated by modified DeepLabv3 + , which have a small and blurred appearance. For evaluation of the segmentation results, the average Dice similarity coefficients (ADSC) were used as the performance index. Results Sixteen MR images corresponding to different subjects were obtained from the National Cheng Kung University Hospital. Our proposed modified DeepLabv3 + generated the following ADSCs: 0.928 for carpal tunnel, 0.872 for flexor tendons and 0.785 for the median nerve. The ADSC value of 0.8053 generated the MaskTrack that 0.22 ADSC measure were improved for measuring the median nerve. Conclusions The experimental results showed that the proposed modified DeepLabv3 + model can promote segmentations of the carpal tunnel and its contents. The results are superior to the results generated by original DeepLabv3 + . Additionally, MaskTrack can also effectively refine median nerve segmentations.

scholarly journals Neural Network-derived perfusion maps: a Model-free approach to computed tomography perfusion in patients with acute ischemic stroke

2021 ◽  
Author(s):  
Umberto A. Gava ◽  
Federico D’Agata ◽  
Enzo Tartaglione ◽  
Marco Grangetto ◽  
Francesca Bertolino ◽  
...  
Keyword(s):  
Neural Network ◽  
Ischemic Stroke ◽  
Acute Ischemic Stroke ◽  
State Of The Art ◽  
Ct Perfusion ◽  
Pearson Correlation ◽  
Ground Truth ◽  
Infarct Core ◽  
Model Free ◽  
Perfusion Maps

AbstractPurposeIn this study we investigate whether a Convolutional Neural Network (CNN) can generate clinically relevant parametric maps from CT perfusion data in a clinical setting of patients with acute ischemic stroke.MethodsTraining of the CNN was done on a subset of 100 perfusion data, while 15 samples were used as validation. All the data used for the training/validation of the network and to generate ground truth (GT) maps, using a state-of-the-art deconvolution-algorithm, were previously pre-processed using a standard pipeline. Validation was carried out through manual segmentation of infarct core and penumbra on both CNN-derived maps and GT maps. Concordance among segmented lesions was assessed using the Dice and the Pearson correlation coefficients across lesion volumes.ResultsMean Dice scores from two different raters and the GT maps were > 0.70 (good-matching). Inter-rater concordance was also high and strong correlation was found between lesion volumes of CNN maps and GT maps (0.99, 0.98).ConclusionOur CNN-based approach generated clinically relevant perfusion maps that are comparable to state-of-the-art perfusion analysis methods based on deconvolution of the data. Moreover, the proposed technique requires less information to estimate the ischemic core and thus might allow the development of novel perfusion protocols with lower radiation dose.

scholarly journals Computed tomography angiography-based deep learning method for treatment selection and infarct volume prediction in anterior cerebral circulation large vessel occlusion

2021 ◽  
Vol 10 (11) ◽  
pp. 205846012110603
Author(s):  
Lasse Hokkinen ◽  
Teemu Mäkelä ◽  
Sauli Savolainen ◽  
Marko Kangasniemi
Keyword(s):  
Neural Network ◽  
Computed Tomography ◽  
Convolutional Neural Network ◽  
Computed Tomography Angiography ◽  
Time Window ◽  
Infarct Volume ◽  
Correlation Coefficients ◽  
Late Time ◽  
Large Vessel Occlusion ◽  
Vessel Occlusion

Background Computed tomography perfusion (CTP) is the mainstay to determine possible eligibility for endovascular thrombectomy (EVT), but there is still a need for alternative methods in patient triage. Purpose To study the ability of a computed tomography angiography (CTA)-based convolutional neural network (CNN) method in predicting final infarct volume in patients with large vessel occlusion successfully treated with endovascular therapy. Materials and Methods The accuracy of the CTA source image-based CNN in final infarct volume prediction was evaluated against follow-up CT or MR imaging in 89 patients with anterior circulation ischemic stroke successfully treated with EVT as defined by Thrombolysis in Cerebral Infarction category 2b or 3 using Pearson correlation coefficients and intraclass correlation coefficients. Convolutional neural network performance was also compared to a commercially available CTP-based software (RAPID, iSchemaView). Results A correlation with final infarct volumes was found for both CNN and CTP-RAPID in patients presenting 6–24 h from symptom onset or last known well, with r = 0.67 ( p < 0.001) and r = 0.82 ( p < 0.001), respectively. Correlations with final infarct volumes in the early time window (0–6 h) were r = 0.43 ( p = 0.002) for the CNN and r = 0.58 ( p < 0.001) for CTP-RAPID. Compared to CTP-RAPID predictions, CNN estimated eligibility for thrombectomy according to ischemic core size in the late time window with a sensitivity of 0.38 and specificity of 0.89. Conclusion A CTA-based CNN method had moderate correlation with final infarct volumes in the late time window in patients successfully treated with EVT.

Abstract P357: Deep Learning Building on Prior Ischemic Core Segmentation Improves Prediction of Infarction After Stroke

Stroke ◽  
2021 ◽  
Vol 52 (Suppl_1) ◽  
Author(s):  
Julian Klug ◽  
Guillaume Leclerc ◽  
Elisabeth Dirren ◽  
Preti Maria Gulia ◽  
Dimitri Van De Ville ◽  
...  
Keyword(s):  
Machine Learning ◽  
Deep Learning ◽  
Infarct Volume ◽  
Machine Learning Algorithms ◽  
University Hospital ◽  
Baseline Model ◽  
Stroke Treatment ◽  
Efficient Integration ◽  
Ischemic Core

Introduction: Imaging studies are used to guide patient selection for acute stroke treatment. Perfusion CT (pCT) is widely used to identify the acute ischemic core and penumbra, but the prediction of the final infarct remains challenging. With the advent of machine learning, algorithms learning the prediction of the final lesion from imaging data collected in the acute phase have been proposed. We aimed to investigate whether machine learning methods that integrate prior ischemic core segmentation improve the prediction of the final infarct after stroke. Methodology: We retrospectively included all stroke patients admitted to the Geneva University Hospital for intravenous and/or endovascular treatment from 01.2016 to 12.2017. All patients had acute pCT and follow-up MRI. An Attention-Gated 3D Unet was used as the baseline model on which the effect of access to a threshold-based ischemic core segmentation was tested. To ensure the efficient integration of information contained in voxels from the ischemic core, we extended the baseline model with a bayesian skip connection allowing only the prior to bypass most of the network. This modifies the model’s task to predict divergence from the prior representation. All models were evaluated for the prediction of the final infarct on follow-up MRI, given acute pCT maps as input. The output of each model was compared to finals lesions manually delineated by expert neurologists. Dice score was used to assess performance. Results: A total of 144 patients were included. Median hypoperfused tissue volume (Tmax > 6s) was 60 ml [17-134], median ischemic core (relative CBF < 38%) volume was 23 ml [17-33] and median final infarct volume was 13 ml [3-38]. Dice score for the threshold based ischemic core segmentation was 0.1. The baseline model with and without prior segmentation as input achieved a Dice score of 0.19. Adding the proposed bayesian skip connection lead to a more efficient integration of the prior segmentation ensuring faster convergence and better performance with a final Dice score of 0.21. Conclusion: The evaluated deep learning model can effectively leverage the information contained in a prior segmentation of the ischemic core to enhance the learning process and improve the prediction of the final infarct after stroke.

scholarly journals Stroke progression and clinical outcome in ischemic stroke patients with a history of migraine

2019 ◽  
Vol 14 (9) ◽  
pp. 946-955 ◽  
Author(s):  
Inge A Mulder ◽  
Ghislaine Holswilder ◽  
Marianne AA van Walderveen ◽  
Irene C van der Schaaf ◽  
Edwin Bennink ◽  
...  
Keyword(s):  
Ischemic Stroke ◽  
Blood Brain Barrier ◽  
Infarct Volume ◽  
Hemorrhagic Transformation ◽  
Perfusion Deficit ◽  
Infarct Core ◽  
Severe Stroke ◽  
Brain Barrier Permeability ◽  
History Of

Background Patients with migraine might be more susceptible of spreading depolarizations, which are known to affect vascular and neuronal function and penumbra recovery after stroke. We investigated whether these patients have more severe stroke progression and less favorable outcomes after recanalization therapy. Methods We included patients from a prospective multicenter ischemic stroke cohort. Lifetime migraine history was based on the International Classification of Headache Disorders II criteria. Patients without confirmed migraine diagnosis were excluded. Patients underwent CT angiography and CT perfusion <9 h of onset and follow-up CT after three days. On admission, presence of a perfusion deficit, infarct core and penumbra volume, and blood brain barrier permeability (BBBP) were assessed. At follow-up we assessed malignant edema, hemorrhagic transformation, and final infarct volume. Outcome at three months was evaluated with the modified Rankin Scale (mRS). We calculated adjusted relative risks (aRR) or difference of means (aB) with regression analyses. Results We included 600 patients of whom 43 had migraine. There were no differences between patients with or without migraine in presence of a perfusion deficit on admission (aRR: 0.98, 95%CI: 0.77–1.25), infarct core volume (aB: -10.8, 95%CI: -27.04–5.51), penumbra volume (aB: -11.6, 95%CI: -26.52–3.38), mean blood brain barrier permeability (aB: 0.08, 95%CI: -3.11–2.96), malignant edema (0% vs. 5%), hemorrhagic transformation (aRR: 0.26, 95%CI: 0.04–1.73), final infarct volume (aB: -14.8, 95%CI: 29.9–0.2) or outcome after recanalization therapy (mRS > 2, aRR: 0.50, 95%CI: 0.21–1.22). Conclusion Elderly patients with a history of migraine do not seem to have more severe stroke progression and have similar treatment outcomes compared with patients without migraine.

scholarly journals Machine Learning Classification of Inflammatory Bowel Disease in Children Based on a Large Real-World Pediatric Cohort CEDATA-GPGE® Registry

2021 ◽  
Vol 8 ◽  
Author(s):  
Nicolas Schneider ◽  
Keywan Sohrabi ◽  
Henning Schneider ◽  
Klaus-Peter Zimmer ◽  
Patrick Fischer ◽  
...  
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Support Vector Machine ◽  
Convolutional Neural Network ◽  
Support Vector ◽  
Machine Learning Classification ◽  
Extreme Gradient Boosting ◽  
Inflammatory Bowel

Introduction: The rising incidence of pediatric inflammatory bowel diseases (PIBD) facilitates the need for new methods of improving diagnosis latency, quality of care and documentation. Machine learning models have shown to be applicable to classifying PIBD when using histological data or extensive serology. This study aims to evaluate the performance of algorithms based on promptly available data more suited to clinical applications.Methods: Data of inflammatory locations of the bowels from initial and follow-up visitations is extracted from the CEDATA-GPGE registry and two follow-up sets are split off containing only input from 2017 and 2018. Pre-processing excludes patients in remission and encodes the categorical data numerically. For classification of PIBD diagnosis, a support vector machine (SVM), a random forest algorithm (RF), extreme gradient boosting (XGBoost), a dense neural network (DNN) and a convolutional neural network (CNN) are employed. As best performer, a convolutional neural network is further improved using grid optimization.Results: The achieved accuracy of the optimized neural network reaches up to 90.57% on data inserted into the registry in 2018. Less performant methods reach 88.78% for the DNN down to 83.94% for the XGBoost. The accuracy of prediction for the 2018 follow-up dataset is higher than those for older datasets. Neural networks yield a higher standard deviation with 3.45 for the CNN compared to 0.83–0.86 of the support vector machine and ensemble methods.Discussion: The displayed accuracy of the convolutional neural network proofs the viability of machine learning classification in PIBD diagnostics using only timely available data.

Sign in / Sign up

Export Citation Format

Share Document