Determining Ischemic Stroke From CT-Angiography Imaging Using Symmetry-Sensitive Convolutional Networks

Barman, Arko; Inam, Mehmet Enes; Lee, Songmi; Savitz, Sean I; Sheth, Sunil A; Giancardo, Luca

doi:10.1109/isbi.2019.8759475

Cited by 43 publications

(43 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The Deep Symmetry-Sensitive Convolutional Neural Network (DeepSymNet) architecture (Figure 2) used in this study was inspired from a model designed to identify spatial symmetries in brain angiograms (Barman et al, 2019). We applied this architecture to identify changes through time as opposed to spatial symmetries.…”

Section: Methodsmentioning

confidence: 99%

“…In this work, we propose to use a DeepSymNet-based model, a novel end-to-end deep learning architecture, to identify longitudinal neurodegenerative progression between structural MRI images with minimal preprocessing at two-time points. We adapt the DeepSymNet architecture presented by Barman et al (2019) to identify structural brain differences by learning time-sensitive representation on a subject-level. The imaging preprocessing pipeline required by the architecture does not use a priori brain regions or non-rigid registration algorithms making the process more robust by having fewer steps throughout the pipeline and more efficient in terms of time required to generate hypotheses and computational time than common longitudinal processing pipelines such as Freesurfer.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Quantifying Neurodegenerative Progression With DeepSymNet, an End-to-End Data-Driven Approach

et al. 2019

Self Cite

View full text Add to dashboard Cite

Alzheimer's disease (AD) is the most common neurodegenerative disorder worldwide and is one of the leading sources of morbidity and mortality in the aging population. There is a long preclinical period followed by mild cognitive impairment (MCI). Clinical diagnosis and the rate of decline is variable. Progression monitoring remains a challenge in AD, and it is imperative to create better tools to quantify this progression. Brain magnetic resonance imaging (MRI) is commonly used for patient assessment. However, current approaches for analysis require strong a priori assumptions about regions of interest used and complex preprocessing pipelines including computationally expensive non-linear registrations and iterative surface deformations. These preprocessing steps are composed of many stacked processing layers. Any error or bias in an upstream layer will be propagated throughout the pipeline. Failures or biases in the non-linear subject registration and the subjective choice of atlases of specific regions are common in medical neuroimaging analysis and may hinder the translation of many approaches to the clinical practice. Here we propose a data-driven method based on an extension of a deep learning architecture, DeepSymNet, that identifies longitudinal changes without relying on prior brain regions of interest, an atlas, or non-linear registration steps. Our approach is trained end-to-end and learns how a patient's brain structure dynamically changes between two-time points directly from the raw voxels. We compare our approach with Freesurfer longitudinal pipelines and voxel-based methods using the Alzheimer's Disease Neuroimaging Initiative (ADNI) database. Our model can identify AD progression with comparable results to existing Freesurfer longitudinal pipelines without the need of predefined regions of interest, non-rigid registration algorithms, or iterative surface deformation at a fraction of the processing time. When compared to other voxel-based methods which share some of the same benefits, our model showed a statistically significant performance improvement. Additionally, we show that our model can differentiate between healthy subjects and patients with MCI. The model's decision was investigated using the epsilon layer-wise propagation algorithm. We found that the predictions were driven by the pallidum, putamen, and the superior temporal gyrus. Our novel longitudinal based, deep learning approach has the potential to diagnose patients earlier and enable new computational tools to monitor neurodegeneration in clinical practice.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Quantifying Neurodegenerative Progression With DeepSymNet, an End-to-End Data-Driven Approach

et al. 2019

Self Cite

View full text Add to dashboard Cite

show abstract

“…We anticipate these segmentations to be a particularly useful addition to the corpus of training data for image processing pipelines, including the ones leveraging deep learning algorithms. Indeed, this project began as an effort to provide VOIs to study image feature symmetry between left and right thorax anatomy as a clinical outcomes predictor, building on previous work from our group that localized stroke cores by comparing and contrasting brain hemisphere information extracted by “symmetry‐sensitive convolutional neural networks.” 58 …”

Section: Discussionmentioning

confidence: 99%

“…Indeed, this project began as an effort to provide VOIs to study image feature symmetry between left and right thorax anatomy as a clinical outcomes predictor, building on previous work from our group that localized stroke cores by comparing and contrasting brain hemisphere information extracted by "symmetry-sensitive convolutional neural networks." 58 Our pleural effusion segmentations are likely to be useful for investigating two questions surrounding a CT or PET/CT finding of pleural effusion: (a) the prognostic significance of pleural effusion in various cancer types, 59 and (b) the capacity of CT to discriminate between benign and malignant effusions. [60][61][62][63] Regarding the first question, Ryu et al 59 showed that in small cell lung cancer with stage I-III disease, the presence of even minimal pleural effusion confers an increased risk of death.…”

Section: Discussionmentioning

confidence: 99%

PleThora: Pleural effusion and thoracic cavity segmentations in diseased lungs for benchmarking chest CT processing pipelines

et al. 2020

Self Cite

View full text Add to dashboard Cite

This manuscript describes a dataset of thoracic cavity segmentations and discrete pleural effusion segmentations we have annotated on 402 computed tomography (CT) scans acquired from patients with non-small cell lung cancer. The segmentation of these anatomic regions precedes fundamental tasks in image analysis pipelines such as lung structure segmentation, lesion detection, and radiomics feature extraction. Bilateral thoracic cavity volumes and pleural effusion volumes were manually segmented on CT scans acquired from The Cancer Imaging Archive "NSCLC Radiomics" data collection. Four hundred and two thoracic segmentations were first generated automatically by a U-Net based algorithm trained on chest CTs without cancer, manually corrected by a medical student to include the complete thoracic cavity (normal, pathologic, and atelectatic lung parenchyma, lung hilum, pleural effusion, fibrosis, nodules, tumor, and other anatomic anomalies), and revised by a radiation oncologist or a radiologist. Seventy-eight pleural effusions were manually segmented by a medical student and revised by a radiologist or radiation oncologist. Interobserver agreement between the radiation oncologist and radiologist corrections was acceptable. All expert-vetted segmentations are publicly available in NIfTI format through The Cancer Imaging Archive at https://doi.org/10. 7937/tcia.2020.6c7y-gq39. Tabular data detailing clinical and technical metadata linked to segmentation cases are also available. Thoracic cavity segmentations will be valuable for developing image analysis pipelines on pathologic lungswhere current automated algorithms struggle most. In conjunction with gross tumor volume segmentations already available from "NSCLC Radiomics," pleural effusion segmentations may be valuable for investigating radiomics profile differences between effusion and primary tumor or training algorithms to discriminate between them.

show abstract

“…To this end, we previously developed a machine learning model called DeepSymNet (DSN) that successfully predicted infarct core, as compared with concurrently acquired CTP, using a much more widely available modality, CT angiography (CTA), in an automated fashion. 6,7 In this study, we hypothesized that NCHCT-ASPECTS and CTA with DSN perform adequately well in identifying the infarct core relative to CTP with Rapid (iSchemaView, Inc.) in patients with LVO AIS. We compared the performance of these three modalities at predicting the final infarct volume (FIV) in patients who underwent successful EVT.…”

mentioning

confidence: 99%

Automated prediction of final infarct volume in patients with large-vessel occlusion acute ischemic stroke

Abdelkhaleq¹,

Kim²,

Khose³

et al. 2021

Neurosurgical Focus

Self Cite

View full text Add to dashboard Cite

OBJECTIVE In patients with large-vessel occlusion (LVO) acute ischemic stroke (AIS), determinations of infarct size play a key role in the identification of candidates for endovascular stroke therapy (EVT). An accurate, automated method to quantify infarct at the time of presentation using widely available imaging modalities would improve screening for EVT. Here, the authors aimed to compare the performance of three measures of infarct core at presentation, including an automated method using machine learning. METHODS Patients with LVO AIS who underwent successful EVT at four comprehensive stroke centers were identified. Patients were included if they underwent concurrent noncontrast head CT (NCHCT), CT angiography (CTA), and CT perfusion (CTP) with Rapid imaging at the time of presentation, and MRI 24 to 48 hours after reperfusion. NCHCT scans were analyzed using the Alberta Stroke Program Early CT Score (ASPECTS) graded by neuroradiology or neurology expert readers. CTA source images were analyzed using a previously described machine learning model named DeepSymNet (DSN). Final infarct volume (FIV) was determined from diffusion-weighted MRI sequences using manual segmentation. The primary outcome was the performance of the three infarct core measurements (NCHCT-ASPECTS, CTA with DSN, and CTP-Rapid) to predict FIV, which was measured using area under the receiver operating characteristic (ROC) curve (AUC) analysis. RESULTS Among 76 patients with LVO AIS who underwent EVT and met inclusion criteria, the median age was 67 years (IQR 54–76 years), 45% were female, and 37% were White. The median National Institutes of Health Stroke Scale score was 16 (IQR 12–22), and the median NCHCT-ASPECTS on presentation was 8 (IQR 7–8). The median time between when the patient was last known to be well and arrival was 156 minutes (IQR 73–303 minutes), and between NCHCT/CTA/CTP to groin puncture was 73 minutes (IQR 54–81 minutes). The AUC was obtained at three different cutoff points: 10 ml, 30 ml, and 50 ml FIV. At the 50-ml FIV cutoff, the AUC of ASPECTS was 0.74; of CTP core volume, 0.72; and of DSN, 0.82. Differences in AUCs for the three predictors were not significant for the three FIV cutoffs. CONCLUSIONS In a cohort of patients with LVO AIS in whom reperfusion was achieved, determinations of infarct core at presentation by NCHCT-ASPECTS and a machine learning model analyzing CTA source images were equivalent to CTP in predicting FIV. These findings have suggested that the information to accurately predict infarct core in patients with LVO AIS was present in conventional imaging modalities (NCHCT and CTA) and accessible by machine learning methods.

show abstract

Determining Ischemic Stroke From CT-Angiography Imaging Using Symmetry-Sensitive Convolutional Networks

Cited by 43 publications

References 18 publications

Quantifying Neurodegenerative Progression With DeepSymNet, an End-to-End Data-Driven Approach

Quantifying Neurodegenerative Progression With DeepSymNet, an End-to-End Data-Driven Approach

PleThora: Pleural effusion and thoracic cavity segmentations in diseased lungs for benchmarking chest CT processing pipelines

Automated prediction of final infarct volume in patients with large-vessel occlusion acute ischemic stroke

Contact Info

Product

Resources

About