Automated ASPECTS on Noncontrast CT Scans in Patients with Acute Ischemic Stroke Using Machine Learning

Kuang, Hulin; Najm, Mohamed; Chakraborty, Debabrata; Maraj, Nicholas; Sohn, Sung Il; Goyal, Mayank; Hill, Michael D.; Demchuk, Andrew M.; Menon, Bijoy K; Qiu, Wu

doi:10.3174/ajnr.a5889

Cited by 98 publications

(77 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…4 With the introduction of software packages trained on deep learning algorithms, attempts have been made to use automated ASPECTS as a way to address variability associated with human interpretation with some success. [10][11][12][23][24][25] In this study, we showed that automated ASPECTS has a similar diagnostic performance to consensus reading of experienced neuroradiologists with excellent agreement (k = 0.84). In a recent study by Maegerlein et al 11 using a different software package, similar results were shown with substantial agreement (k = 0.9) between automated and consensus reads.…”

Section: Discussionsupporting

confidence: 53%

Automated ASPECTS in Acute Ischemic Stroke: A Comparative Analysis with CT Perfusion

Sundaram¹,

Goldstein²,

Wheelwright

et al. 2019

AJNR Am J Neuroradiol

View full text Add to dashboard Cite

BACKGROUND AND PURPOSE: Automated ASPECTS has the potential of reducing interobserver variability in the determination of early ischemic changes. We aimed to assess the performance of an automated ASPECTS software against the assessment of a neuroradiologist in a comparative analysis with concurrent CTP-based CBV ASPECTS. MATERIALS AND METHODS: Patients with anterior circulation stroke who had baseline NCCT and CTP and underwent successful mechanical thrombectomy were included. NCCT-ASPECTS was assessed by 2 neuroradiologists, and discrepancies were resolved by consensus. CTP-CBV ASPECTS was assessed by a different neuroradiologist. Automated ASPECTS was provided by Brainomix software. ASPECTS was dichotomized (ASPECTS $6 or ,6) and was also based on the time from onset (.6 or #6 hours). RESULTS: A total of 58 patients were included. The interobserver agreement for NCCT ASPECTS was moderate (k = 0.48) and marginally improved (k = 0.64) for dichotomized data. Automated ASPECTS showed excellent agreement with consensus reads (k = 0.84) and CTP-CBV ASPECTS (k = 0.84). Intraclass correlation coefficients for ASPECTS across all 3 groups were 0.84 (95% CI, 0.76-0.90, raw scores) and 0.94 (95% CI, 0.91-0.96, dichotomized scores). Automated scores were comparable with consensus reads and CTP-CBV ASPECTS in patients when grouped on the basis of time from symptom onset (.6 or #6 hours). There was significant (P , .001) negative correlation with final infarction volume and the 3 ASPECTS groups (r = À0.52, consensus reads; À0.58, CTP-CBV; and À0.66, automated). CONCLUSIONS: ASPECTS derived from an automated software performs equally as well as consensus reads of expert neuroradiologists and concurrent CTP-CBV ASPECTS and can be used to standardize ASPECTS reporting and minimize interpretation variability.

show abstract

Section: Discussionsupporting

confidence: 53%

Automated ASPECTS in Acute Ischemic Stroke: A Comparative Analysis with CT Perfusion

Sundaram¹,

Goldstein²,

Wheelwright

et al. 2019

AJNR Am J Neuroradiol

View full text Add to dashboard Cite

show abstract

“…The majority of current studies describe AI-mediated detection of core stroke lesion volume on CT and MRI scans, and some use these features to predict acute stroke growth 16–19. For non-contrast CT analysis specifically, automated ASPECTS from ML used in conjunction with clinical presentation accurately associates NIHSS score and can be used to select patients for ET 20 21.…”

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

217

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

“…The dice obtained by this model is 75.24%, significantly lower ( P < 0.01) than 79.42% reported by the proposed method in this study, which suggests the effectiveness of this feature. In order to assess the sensitivity and stability of the lesion occurrence probability feature, we calculated the proportion of the infarcted ASPECTS regions in both 30 training images and all 100 images. It was found that the proportions of ASPECTS regions with infarction in 30 training images are similar to the ones in all 100 images.…”

Section: Resultsmentioning

confidence: 99%

“…4 Ischemic stroke lesions were segmented based on texture features and classification. 5,6 Nowinski et al detected, localized, and quantified the stroke infarct by analyzing hemisphere attenuation value distributions using percentile difference ratios. 7 Tyan et al proposed an ischemic stroke detection system including contrast enhancement, the brain tissue image area extraction, and an unsupervised region growing algorithm, and tested this method on 90 CT scan slices from 26 ischemic stroke patients.…”

Section: Introductionmentioning

confidence: 99%

Semi‐automated infarct segmentation from follow‐up noncontrast CT scans in patients with acute ischemic stroke

2019

View full text Add to dashboard Cite

Purpose Cerebral infarct volume observed in follow‐up noncontrast computed tomography (NCCT) scans of acute ischemic stroke (AIS) patients is as an important radiologic outcome measure of the effectiveness of endovascular therapy (EVT). In this paper, our aim is to propose a semiautomated segmentation approach that can accurately measure ischemic infarct volume from NCCT images of AIS patients. Methods A novel cascaded random forest (RF) learning is first employed to classify each voxel into normal or ischemic voxel, leading to an infarct probability map. Four types of features: intensity, statistical information in local region, symmetric difference compared to the contralateral side, and spatial probability of infarct occurrence generated by the STAPLE method, are extracted. These features are input into RF to train a first‐stage classifier. The coarse segmentation results generated by the first‐stage classifier are then used to train a fine second‐stage classifier with fivefold cross validation. The RF estimated infarct probability map obtained in the second‐stage testing as well as user input high‐level knowledge are then incorporated into a convex optimization function to obtain final segmentation. One hundred AIS patients were collected in this study, of which 70 patient images were used for evaluation while the remaining 30 patient images were used for RF training. Results Quantitative results show that the proposed approach is capable of yielding a dice coefficient (DC) of 79.42%, significantly outperforming some state‐of‐the‐art automated segmentation methods, such as the RF‐based methods and convolutional neural network (CNN)‐based segmentation method, U‐net. The infarct volume obtained by the proposed method is strongly correlated with the manually segmented volume. In addition, interobserver variability analysis initialized by two observers suggests low user dependency. Conclusions Our proposed semiautomated segmentation method can accurately segment infarct from NCCT of AIS patients.

show abstract

Automated ASPECTS on Noncontrast CT Scans in Patients with Acute Ischemic Stroke Using Machine Learning

Cited by 98 publications

References 26 publications

Automated ASPECTS in Acute Ischemic Stroke: A Comparative Analysis with CT Perfusion

Automated ASPECTS in Acute Ischemic Stroke: A Comparative Analysis with CT Perfusion

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

Semi‐automated infarct segmentation from follow‐up noncontrast CT scans in patients with acute ischemic stroke

Contact Info

Product

Resources

About