Context-Aware Convolutional Neural Networks for Stroke Sign Detection in Non-contrast CT Scans

Lisowska, Aneta; O’Neil, Alison Q.; Dilys, Vismantas; Daykin, Matthew; Beveridge, Erin; Muir, Keith W.; McLaughlin, Stephen; Poole, Ian

doi:10.1007/978-3-319-60964-5_43

Cited by 30 publications

(19 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The reported sensitivity of AI algorithm ASPECTS range from 45% to 98%, mean 68%, and specificity ranges from 57% to 95%, mean 81% 9 10 14 16 17 22–27. Use of a convolutional neural network (CNN) for a combined asymmetric middle cerebral artery territory hypodensity and dense vessel detection may have higher performance; however, only area under the curve (AUC) metrics are reported (receiver operating characteristic AUC 92–96%) 28. Collectively among these studies, there is significant heterogeneity in study data, including cohort size, gold standard comparison, and time to initial head CT acquisition (table 2, online supplementary table I).…”

Section: Resultsmentioning

confidence: 99%

Artificial intelligence to diagnose ischemic stroke and identify large vessel occlusions: a systematic review

Murray

Unberath

Hager

et al. 2019

J NeuroIntervent Surg

224

163

View full text Add to dashboard Cite

Background and purposeAcute stroke caused by large vessel occlusions (LVOs) requires emergent detection and treatment by endovascular thrombectomy. However, radiologic LVO detection and treatment is subject to variable delays and human expertise, resulting in morbidity. Imaging software using artificial intelligence (AI) and machine learning (ML), a branch of AI, may improve rapid frontline detection of LVO strokes. This report is a systematic review of AI in acute LVO stroke identification and triage, and characterizes LVO detection software.MethodsA systematic review of acute stroke diagnostic-focused AI studies from January 2014 to February 2019 in PubMed, Medline, and Embase using terms: ‘artificial intelligence’ or ‘machine learning or deep learning’ and ‘ischemic stroke’ or ‘large vessel occlusion’ was performed.ResultsVariations of AI, including ML methods of random forest learning (RFL) and convolutional neural networks (CNNs), are used to detect LVO strokes. Twenty studies were identified that use ML. Alberta Stroke Program Early CT Score (ASPECTS) commonly used RFL, while LVO detection typically used CNNs. Image feature detection had greater sensitivity with CNN than with RFL, 85% versus 68%. However, AI algorithm performance metrics use different standards, precluding ideal objective comparison. Four current software platforms incorporate ML: Brainomix (greatest validation of AI for ASPECTS, uses CNNs to automatically detect LVOs), General Electric, iSchemaView (largest number of perfusion study validations for thrombectomy), and Viz.ai (uses CNNs to automatically detect LVOs, then automatically activates emergency stroke treatment systems).ConclusionsAI may improve LVO stroke detection and rapid triage necessary for expedited treatment. Standardization of performance assessment is needed in future studies.

show abstract

Section: Resultsmentioning

confidence: 99%

Artificial intelligence to diagnose ischemic stroke and identify large vessel occlusions: a systematic review

Murray

Unberath

Hager

et al. 2019

J NeuroIntervent Surg

224

163

View full text Add to dashboard Cite

show abstract

Section: Key Resultsmentioning

confidence: 99%

“…A relatively recent work (11) explored the feasibility of applying a deep learning technique, convolutional neural network architecture, to this challenging problem. Unfortunately, that work could be used only to qualitatively detect the presence or absence of ischemic lesions at the hemisphere level (11). In our study, we investigated ML techniques with manually defined features combined with deep learning-based features to automatically detect and quantitate infarction on baseline non-contrast-enhanced CT images, with diffusion-weighted (DW) MRI used as the reference standard.…”

Section: Key Resultsmentioning

confidence: 99%

“…Most published methods have used ML-based approaches and have relied heavily on comparison of the ipsilateral and contralateral brain regions, as this is how humans interpret images (16,17,22). Lisowska et al (11) explored the feasibility of applying a deep learning technique, namely a convolutional neural network-based method, to this problem. Their work, however, could only estimate the presence or absence of ischemic lesions at the hemisphere level.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Machine Learning for Detecting Early Infarction in Acute Stroke with Non–Contrast-enhanced CT

Qiu¹,

et al. 2020

View full text Add to dashboard Cite

NEURORADIOLOGYI dentification of the presence and extent of infarcted brain tissue at baseline plays a crucial role in the management of acute ischemic stroke (AIS). Patients with extensive infarction at baseline are unlikely to benefit from thrombolysis or thrombectomy procedures (1-4). Non-contrastenhanced CT is the most widely used imaging modality to assess the extent of infarction in patients with acute stroke (depicted as hypoattenuation). The Alberta Stroke Program Early CT Score (ASPECTS) offers a semiquantitative approach to subjectively assess infarction extent on non-contrast-enhanced CT images (1,2,5). ASPECTS assessment with this method, however, requires expertise, something not readily available in smaller hospitals or during off hours (6). Moreover, because it is semiquantitative, ASPECTS offers only an approximation of the total extent of early ischemic changes. For example, a patient with an ASPECTS of 6 could have a large infarct if most of the cortical regions are involved; another patient with the same ASPECTS of 6 could have a relatively small infarct if only subcortical regions are involved.Quantitative estimation of infarction with non-contrastenhanced CT is challenging because the density and texture variations in the involved brain regions are subtle and can be confounded by normal physiologic changes or old lesions (7,8). Low signal-to-noise ratio, thick slices, and low contrast in images of brain tissue make most traditional image-based segmentation approaches difficult (eg, thresholding-derived, region-based, edge-based, or model-based methods) (7). Recent advances in image processing that use deep learningor machine learning (ML)-based algorithms have potential

show abstract

“…Of these, Support Vector Machine (SVM ) and neural networks are the most popular techniques used in medical applicat ions [6], [7]. Several works such as those discussed in [8], [9], [10] only use med ical images such as CT scans in order to train their models. Yet others, such as those discussed in [11], [12], use a combination of med ical images along with other clinical data in order to train the model.…”

Section: Related Workmentioning

confidence: 99%

Adaptive Neuro-Fuzzy Inference System for Prediction of Surgery Time for Ischemic Stroke Patients

Ali

Qidwai

Ilyas

et al. 2019

IJIE

View full text Add to dashboard Cite

There is a worldwide trend in the increase of the number of stroke cases reported every year. According to the World Health Organizat ion, stroke is one of the leading causes of death in 2016 with appro ximately 5.8 million deaths reported in 2016 alone. While a stroke could happen to anyone, mu ltiple risk factors such as age, hypertension, obesity and diabetes contribute significantly to its occurrence.

show abstract

Context-Aware Convolutional Neural Networks for Stroke Sign Detection in Non-contrast CT Scans

Cited by 30 publications

References 16 publications

Artificial intelligence to diagnose ischemic stroke and identify large vessel occlusions: a systematic review

Artificial intelligence to diagnose ischemic stroke and identify large vessel occlusions: a systematic review

Machine Learning for Detecting Early Infarction in Acute Stroke with Non–Contrast-enhanced CT

Adaptive Neuro-Fuzzy Inference System for Prediction of Surgery Time for Ischemic Stroke Patients

Contact Info

Product

Resources

About