Improved early stroke detection: Wavelet-based perception enhancement of computerized tomography exams

Hybrid Ensemble Deep Learning Model for Advancing Ischemic Brain Stroke Detection and Classification in Clinical Application

Ischemic brain strokes are severe medical conditions that occur due to blockages in the brain's blood flow, often caused by blood clots or artery blockages. Early detection is crucial for effective treatment. This study aims to improve the detection and classification of ischemic brain strokes in clinical settings by introducing a new approach that integrates the stroke precision enhancement, ensemble deep learning, and intelligent lesion detection and segmentation models. The proposed hybrid model was trained and tested using a dataset of 10,000 computed tomography scans. A 25-fold cross-validation technique was employed, while the model's performance was evaluated using accuracy, precision, recall, and F1 score. The findings indicate significant improvements in accuracy for different stages of stroke images when enhanced using the SPEM model with contrast-limited adaptive histogram equalization set to 4. Specifically, accuracy showed significant improvement (from 0.876 to 0.933) for hyper-acute stroke images; from 0.881 to 0.948 for acute stroke images, from 0.927 to 0.974 for sub-acute stroke images, and from 0.928 to 0.982 for chronic stroke images. Thus, the study shows significant promise for the detection and classification of ischemic brain strokes. Further research is needed to validate its performance on larger datasets and enhance its integration into clinical settings.

Taxonomy of Acute Stroke: Imaging, Processing, and Treatment

Stroke management employs a variety of diagnostic imaging modalities, image processing and analysis methods, and treatment procedures. This work categorizes methods for stroke imaging, image processing and analysis, and treatment, and provides their taxonomies illustrated by a state-of-the-art review. Imaging plays a critical role in stroke management, and the most frequently employed modalities are computed tomography (CT) and magnetic resonance (MR). CT includes unenhanced non-contrast CT as the first-line diagnosis, CT angiography, and CT perfusion. MR is the most complete method to examine stroke patients. MR angiography is useful to evaluate the severity of artery stenosis, vascular occlusion, and collateral flow. Diffusion-weighted imaging is the gold standard for evaluating ischemia. MR perfusion-weighted imaging assesses the penumbra. The stroke image processing methods are divided into non-atlas/template-based and atlas/template-based. The non-atlas/template-based methods are subdivided into intensity and contrast transformations, local segmentation-related, anatomy-guided, global density-guided, and artificial intelligence/deep learning-based. The atlas/template-based methods are subdivided into intensity templates and atlases with three atlas types: anatomy atlases, vascular atlases, and lesion-derived atlases. The treatment procedures for arterial and venous strokes include intravenous and intraarterial thrombolysis and mechanical thrombectomy. This work captures the state-of-the-art in stroke management summarized in the form of comprehensive and straightforward taxonomy diagrams. All three introduced taxonomies in diagnostic imaging, image processing and analysis, and treatment are widely illustrated and compared against other state-of-the-art classifications.

Novel artificial intelligence-based hypodensity detection tool improves clinician identification of hypodensity on non-contrast computed tomography in stroke patients

Introduction

In acute stroke, identifying early changes (parenchymal hypodensity) on non-contrast CT (NCCT) can be challenging. We aimed to identify whether the accuracy of clinicians in detecting acute hypodensity in ischaemic stroke patients on a non-contrast CT is improved with the use of an Artificial Intelligence (AI) based, automated hypodensity detection algorithm (HDT) using MRI-DWI as the gold standard.

Methods

The study employed a case-crossover within-clinician design, where 32 clinicians were tasked with identifying hypodensity lesions on NCCT scans for five a priori selected patient cases, before and after viewing the AI-based HDT. The DICE similarity coefficient (DICE score) was the primary measure of accuracy. Statistical analysis compared DICE scores with and without AI-based HDT using mixed-effects linear regression, with individual NCCT scans and clinicians as nested random effects.

Results

The AI-based HDT had a mean DICE score of 0.62 for detecting hypodensity across all NCCT scans. Clinicians' overall mean DICE score was 0.33 (SD 0.31) before AI-based HDT implementation and 0.40 (SD 0.27) after implementation. AI-based HDT use was associated with an increase of 0.07 (95% CI: 0.02-0.11, p = 0.003) in DICE score accounting for individual scan and clinician effects. For scans with small lesions, clinicians achieved a mean increase in DICE score of 0.08 (95% CI: 0.02, 0.13, p = 0.004) following AI-based HDT use. In a subgroup of 15 trainees, DICE score improved with AI-based HDT implementation [mean difference in DICE 0.09 (95% CI: 0.03, 0.14, p = 0.004)].

Discussion

AI-based automated hypodensity detection has potential to enhance clinician accuracy of detecting hypodensity in acute stroke diagnosis, especially for smaller lesions, and notably for less experienced clinicians.

Detecting brain lesions in suspected acute ischemic stroke with CT-based synthetic MRI using generative adversarial networks

Background

Difficulties in detecting brain lesions in acute ischemic stroke (AIS) have convinced researchers to use computed tomography (CT) to scan for and magnetic resonance imaging (MRI) to search for these lesions. This work aimed to develop a generative adversarial network (GAN) model for CT-to-MR image synthesis and evaluate reader performance with synthetic MRI (syn-MRI) in detecting brain lesions in suspected patients.

Methods

Patients with primarily suspected AIS were randomly assigned to the training (n=140) or testing (n=53) set. Emergency CT and follow-up MR images in the training set were used to develop a GAN model to generate syn-MR images from the CT data in the testing set. The standard reference was the manual segmentations of follow-up MR images. Image similarity was evaluated between syn-MRI and the ground truth using a 4-grade visual rating scale, the peak signal-to-noise ratio (PSNR), and the structural similarity index measure (SSIM). Reader performance with syn-MRI and CT was evaluated and compared on a per-patient (patient detection) and per-lesion (lesion detection) basis. Paired t-tests or Wilcoxon signed-rank tests were used to compare reader performance in lesion detection between the syn-MRI and CT data.

Results

Grade 2–4 brain lesions were observed on syn-MRI in 92.5% (49/53) of the patients, while the remaining syn-MRI data showed no lesions compared to the ground truth. The GAN model exhibited a weak PSNR of 24.30 dB but a favorable SSIM of 0.857. Compared with CT, syn-MRI led to an increase in the overall sensitivity from 38% (57/150) to 82% (123/150) in patient detection and from 4% (68/1,620) to 16% (262/1,620) in lesion detection (R=0.32, corrected P<0.001), but the specificity in patient detection decreased from 67% (6/9) to 33% (3/9). An additional 75% (70/93) of patients and 15% (77/517) of lesions missed on CT were detected on syn-MRI.

Conclusions

The GAN model holds potential for generating synthetic MR images from noncontrast CT data and thus could help sensitively detect individuals among patients with suspected AIS. However, the image similarity performance of the model needs to be improved, and further expert discrimination is strongly recommended.

Modified Computed Tomography Classification for Chronic Subdural Hematoma Features Good Interrater Agreement: A Single-Center Retrospective Cohort Study

OBJECTIVE

The present study aimed to establish whether our modified Nakaguchi computed tomography (CT) classification improves the interrater agreement of chronic subdural hematoma (CSDH) classification and prediction of CSDH recurrence relative to 2 other CT classifications.

METHODS

This retrospective study considered 277 consecutive patients with CSDH and 307 hematomas treated with burr-hole surgery between January 2009 and December 2018. Two neurosurgeons blinded to patients' clinical data classified the CT scans of patients with CSDH into 4 or 5 types according to the Nomura classification (high, iso, low, mixed, and layering), Nakaguchi classification (homogenous, laminar, separated, and trabecular), and our modified Nakaguchi classification (homogenous, gradation, laminar, separated, and trabecular). The κ statistic was used to evaluate the interrater agreement of the 3 CT classifications. Univariable and multivariable logistic regression analyses were used to calculate odds ratios for CSDH recurrence.

RESULTS

κ values of the modified, Nakaguchi, and Nomura classification were 0.78, 0.63, and 0.70, respectively. During the 3 months follow-up, the recurrence rate for CSDH was 11.4% (35/307 hematomas). Of the types defined by the modified classification, the gradation type was associated with the highest recurrence (mean recurrence rate, 15.9% ± 0.3%). Multivariable logistic regression analyses showed that a gradation-type hematoma, as defined with the modified classification, was an independent risk factor associated with recurrence (odds ratio, 2.36; 95% confidence interval, 1.11-4.98; P = 0.025).

CONCLUSIONS

The modified classification was useful for preoperative CT classification of CSDH and the prediction of recurrence, with high agreement between raters.

Identification of invisible ischemic stroke in noncontrast CT based on novel two-stage convolutional neural network model

PURPOSE

Early identification of ischemic stroke lesion regions plays a vital role in its treatments like thrombolytic therapy and patients' recovery. Noncontrast computed tomography (ncCT) is the most widespread imaging modality in emergency departments. Unfortunately, it is extremely hard to distinguish the lesion from healthy tissue during the hyper-acute phase of stroke. In this paper, a two-stage convolutional neural network-based method was proposed to identify the invisible ischemic stroke from ncCT.

METHODS

In order to combine the global and local information of images effectively, a cascaded structure with two coordinated networks was used to detect the suspicious stroke regions on the whole and optimize the detailed localization. In the first stage, an end-to-end U-net with adaptive threshold was proposed to integrate global position, symmetry and gray texture information to detect the suspicious regions. After reducing the interference from most normal regions, a ResNet-based patch classification network was used to eliminate some false positive samples on suspicious regions by mining deeper image features, contributing to a more precise localization of stroke. Finally, a MAP model was used to optimize the result by combining the classification results of each patch with their spatial constraint information.

RESULTS

Three independent experiments, that is, training and testing on dataset from one hospital, on the combination of two, and on the two respectively, were performed on a total of 277 cases from two hospitals to validate the proposed model, The proposed method achieved identification accuracy of 91.89%, 87.21%, and 85.71% in the three experiments, and the final localization accuracy in terms of precise localization of stroke were 82.35%, 83.02%, and 81.40%, respectively, which indicated the robustness and clinical values of the method.

CONCLUSIONS

There are some deep image feature differences between stroke region and normal region on ncCT images. The proposed two-stage convolutional neural network model can well seize these features and use them to effectively identify and locate stroke.

Ischemic infarct detection, localization, and segmentation in noncontrast CT human brain scans: review of automated methods

Noncontrast Computed Tomography (NCCT) of the brain has been the first-line diagnosis for emergency evaluation of acute stroke, so a rapid and automated detection, localization, and/or segmentation of ischemic lesions is of great importance. We provide the state-of-the-art review of methods for automated detection, localization, and/or segmentation of ischemic lesions on NCCT in human brain scans along with their comparison, evaluation, and classification. Twenty-two methods are (1) reviewed and evaluated; (2) grouped into image processing and analysis-based methods (11 methods), brain atlas-based methods (two methods), intensity template-based methods (1 method), Stroke Imaging Marker-based methods (two methods), and Artificial Intelligence-based methods (six methods); and (3) properties of these groups of methods are characterized. A new method classification scheme is proposed as a 2 × 2 matrix with local versus global processing and analysis, and density versus spatial sampling. Future studies are necessary to develop more efficient methods directed toward deep learning methods as well as combining the global methods with a high sampling both in space and density for the merged radiologic and neurologic data.

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

Background Identifying the presence and extent of infarcted brain tissue at baseline plays a crucial role in the treatment of patients with acute ischemic stroke (AIS). Patients with extensive infarction are unlikely to benefit from thrombolysis or thrombectomy procedures. Purpose To develop an automated approach to detect and quantitate infarction by using non-contrast-enhanced CT scans in patients with AIS. Materials and Methods Non-contrast-enhanced CT images in patients with AIS (<6 hours from symptom onset to CT) who also underwent diffusion-weighted (DW) MRI within 1 hour after AIS were obtained from May 2004 to July 2009 and were included in this retrospective study. Ischemic lesions manually contoured on DW MRI scans were used as the reference standard. An automatic segmentation approach involving machine learning (ML) was developed to detect infarction. Randomly selected nonenhanced CT images from 157 patients with the lesion labels manually contoured on DW MRI scans were used to train and validate the ML model; the remaining 100 patients independent of the derivation cohort were used for testing. The ML algorithm was quantitatively compared with the reference standard (DW MRI) by using Bland-Altman plots and Pearson correlation. Results In 100 patients in the testing data set (median age, 69 years; interquartile range [IQR]: 59-76 years; 59 men), baseline non-contrast-enhanced CT was performed within a median time of 48 minutes from symptom onset (IQR, 27-93 minutes); baseline MRI was performed a median of 38 minutes (IQR, 24-48 minutes) later. The algorithm-detected lesion volume correlated with the reference standard of expert-contoured lesion volume in acute DW MRI scans (r = 0.76, P < .001). The mean difference between the algorithm-segmented volume (median, 15 mL; IQR, 9-38 mL) and the DW MRI volume (median, 19 mL; IQR, 5-43 mL) was 11 mL (P = .89). Conclusion A machine learning approach for segmentation of infarction on non-contrast-enhanced CT images in patients with acute ischemic stroke showed good agreement with stroke volume on diffusion-weighted MRI scans. © RSNA, 2020 Online supplemental material is available for this article. See also the editorial by Nael in this issue.

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

BACKGROUND AND PURPOSE

Alberta Stroke Program Early CT Score (ASPECTS) was devised as a systematic method to assess the extent of early ischemic change on noncontrast CT (NCCT) in patients with acute ischemic stroke (AIS). Our aim was to automate ASPECTS to objectively score NCCT of AIS patients.

MATERIALS AND METHODS

We collected NCCT images with a 5-mm thickness of 257 patients with acute ischemic stroke (<8 hours from onset to scans) followed by a diffusion-weighted imaging acquisition within 1 hour. Expert ASPECTS readings on DWI were used as ground truth. Texture features were extracted from each ASPECTS region of the 157 training patient images to train a random forest classifier. The unseen 100 testing patient images were used to evaluate the performance of the trained classifier. Statistical analyses on the total ASPECTS and region-level ASPECTS were conducted.

RESULTS

For the total ASPECTS of the unseen 100 patients, the intraclass correlation coefficient between the automated ASPECTS method and DWI ASPECTS scores of expert readings was 0.76 (95% confidence interval, 0.67-0.83) and the mean ASPECTS difference in the Bland-Altman plot was 0.3 (limits of agreement, -3.3, 2.6). Individual ASPECTS region-level analysis showed that our method yielded κ = 0.60, sensitivity of 66.2%, specificity of 91.8%, and area under curve of 0.79 for 100 × 10 ASPECTS regions. Additionally, when ASPECTS was dichotomized (>4 and ≤4), κ = 0.78, sensitivity of 97.8%, specificity of 80%, and area under the curve of 0.89 were generated between the proposed method and expert readings on DWI.

CONCLUSIONS

The proposed automated ASPECTS scoring approach shows reasonable ability to determine ASPECTS on NCCT images in patients presenting with acute ischemic stroke.

Ischemic stroke enhancement using a variational model and the expectation maximization method

OBJECTIVES

In order to enable less experienced physicians to reliably detect early signs of stroke, A novel approach was proposed to enhance the visual perception of ischemic stroke in non-enhanced CT.

METHODS

A set of 39 retrospective CT scans were used, divided into 23 cases of acute ischemic stroke and 16 normal patients. Stroke cases were obtained within 4.5 h of symptom onset and with a mean NIHSS of 12.9±7.4. After selection of adjunct slices from the CT exam, image averaging was performed to reduce the noise and redundant information. This was followed by a variational decomposition model to keep the relevant component of the image. The expectation maximization method was applied to generate enhanced images.

RESULTS

We determined a test to evaluate the performance of observers in a clinical environment with and without the aid of enhanced images. The overall sensitivity of the observer's analysis was 64.5 % and increased to 89.6 % and specificity was 83.3 % and increased to 91.7 %.

CONCLUSION

These results show the importance of a computational tool to assist neuroradiology decisions, especially in critical situations such as the diagnosis of ischemic stroke.

KEY POINTS

• Diagnosing patients with stroke requires high efficiency to avoid irreversible cerebral damage. • A computational algorithm was proposed to enhance the visual perception of stroke. • Observers' performance was increased with the aid of enhanced images.

New approach to detect and classify stroke in skull CT images via analysis of brain tissue densities

BACKGROUND AND OBJECTIVE

Cerebral vascular accident (CVA), also known as stroke, is an important health problem worldwide and it affects 16 million people worldwide every year. About 30% of those that have a stroke die and 40% remain with serious physical limitations. However, recovery in the damaged region is possible if treatment is performed immediately. In the case of a stroke, Computed Tomography (CT) is the most appropriate technique to confirm the occurrence and to investigate its extent and severity. Stroke is an emergency problem for which early identification and measures are difficult; however, computer-aided diagnoses (CAD) can play an important role in obtaining information imperceptible to the human eye. Thus, this work proposes a new method for extracting features based on radiological density patterns of the brain, called Analysis of Brain Tissue Density (ABTD).

METHODS

The proposed method is a specific approach applied to CT images to identify and classify the occurrence of stroke diseases. The evaluation of the results of the ABTD extractor proposed in this paper were compared with extractors already established in the literature, such as features from Gray-Level Co-Occurrence Matrix (GLCM), Local binary patterns (LBP), Central Moments (CM), Statistical Moments (SM), Hu's Moment (HM) and Zernike's Moments (ZM). Using a database of 420 CT images of the skull, each extractor was applied with the classifiers such as MLP, SVM, kNN, OPF and Bayesian to classify if a CT image represented a healthy brain or one with an ischemic or hemorrhagic stroke.

RESULTS

ABTD had the shortest extraction time and the highest average accuracy (99.30%) when combined with OPF using the Euclidean distance. Also, the average accuracy values for all classifiers were higher than 95%.

CONCLUSIONS

The relevance of the results demonstrated that the ABTD method is a useful algorithm to extract features that can potentially be integrated with CAD systems to assist in stroke diagnosis.

A quantitative symmetry-based analysis of hyperacute ischemic stroke lesions in noncontrast computed tomography

PURPOSE

Early identification of ischemic stroke plays a significant role in treatment and potential recovery of damaged brain tissue. In noncontrast CT (ncCT), the differences between ischemic changes and healthy tissue are usually very subtle during the hyperacute phase (< 8 h from the stroke onset). Therefore, visual comparison of both hemispheres is an important step in clinical assessment. A quantitative symmetry-based analysis of texture features of ischemic lesions in noncontrast CT images may provide an important information for differentiation of ischemic and healthy brain tissue in this phase.

METHODS

One hundred thirty-nine (139) ncCT scans of hyperacute ischemic stroke with follow-up magnetic resonance diffusion-weighted (MR-DW) images were collected. The regions of stroke were identified in the MR-DW images, which were spatially aligned to corresponding ncCT images. A state-of-the-art symmetric diffeomorphic image registration was utilized for the alignment of CT and MR-DW, for identification of individual brain hemispheres, and for localization of the region representing healthy tissue contralateral to the stroke cores. Texture analysis included extraction and classification of co-occurrence and run-length texture-based image features in the regions of ischemic stroke and their contralateral regions.

RESULTS

The classification schemes achieved area under the receiver operating characteristic [Az] ≈ 0.82 for the whole dataset. There was no statistically significant difference in the performance of classifiers for the data sets with time between 2 and 8 hours from symptom onset. The performance of the classifiers did not depend on the size of the stroke regions.

CONCLUSIONS

The results provide a set of optimal texture features which are suitable for distinguishing between hyperacute ischemic lesions and their corresponding contralateral brain tissue in noncontrast CT. This work is an initial step toward development of an automated decision support system for detection of hyperacute ischemic stroke lesions on noncontrast CT of the brain.

Quantification of image contrast of infarcts on computed tomography scans

Introduction Accurate identification of infarcts in non-contrast computed tomography (NC-CT) scans of the brain is fundamental in the diagnosis and management of patients with stroke. Quantification of image contrast properties at the boundaries of ischemic infarct regions in NC-CT can contribute to a more precise manual or automatic delineation of these regions. Here we explore these properties quantitatively. Methods We retrospectively investigated 519 NC-CT studies of 425 patients with clinically confirmed ischemic strokes. The average and standard deviation (SD) of patients' age was 67.5 ± 12.4 years and the average(median)±SD time from symptoms onset to NC-CT examination was 27.4(12)±35.7 h. For every scan with an ischemic lesion identified by experts, the image contrast of the lesion vs. normal surrounding parenchyma was calculated as a difference of mean Hounsfield Unit (HU) of 1-5 consecutive voxels (the contrast window width) belonging to the lesion and to the parenchyma. This contrast was calculated at each single voxel of ischemic lesion boundaries (previously delineated by the experts) in horizontal and vertical directions in each image. The distributions of obtained horizontal, vertical and both contrasts combined were calculated among all 519 NC-CTs. Results The highest applicative contrast window width was identified as 5 voxels. The ischemic infarcts were found to be characterized by 6.60 HU, 8.28 HU and 7.55 HU mean values for distributions of horizontal, vertical and combined contrasts. Approximately 40-50% of the infarct boundary voxels were found to refer to the image contrast below 5 HU. Conclusion Low image contrast of ischemic lesions prevents accurate delineation of the infarcts in NC-CT.

A Quantitative Method Using Head Noncontrast CT Scans to Detect Hyperacute Nonvisible Ischemic Changes in Patients With Stroke

PURPOSE

Because clinical evaluation of noncontrast computed tomography (CT) has a poor sensitivity in the evaluation of acute ischemic stroke, computer-aided diagnosis may be able to facilitate the performance. Recently, we introduced a computational method for the detection and localization of visible infarcts. Herein, we aimed to evaluate and extend a previous method, the Stroke Imaging Marker (SIM), to localize nonvisible hyperacute ischemia.

MATERIALS AND METHODS

On the basis of the SIM and its components-the ratio of percentile differences in subranges of Hounsfield Unit (HU) distribution (P-ratio), ratio of voxels count in ranges of brain CT intensity, median HU attenuation value-the infarct localization was performed in 140 early and follow-up scans of 70 patients. In none of the early scans was the infarct visible to a radiologist or an experienced stroke neuroradiologist. The infarcted hemisphere detection rate (HDR) and sensitivity of infarct localization were measured by overlapping the region of detected tissue in the initial scan, with the gold standard set for the fully visible stroke in the follow-up scan.

RESULTS

The best performance of the algorithm was found for the P-ratio including seven percentile subranges within the range of 35th-75th percentile. The modified SIM provided a 76% ischemic HDR and 54% sensitivity in spatial localization of hyperacute ischemia (68% among properly detected infarct sides).

CONCLUSION

The improved SIM is a dedicated and potentially useful tool for hyperacute nonvisible brain infarct detection from CT scans and may contribute to reduction of image-to-needle time in patients eligible for revascularization therapy.

Performance of e-ASPECTS software in comparison to that of stroke physicians on assessing CT scans of acute ischemic stroke patients

BACKGROUND

The Alberta Stroke Program Early CT score (ASPECTS) is an established 10-point quantitative topographic computed tomography scan score to assess early ischemic changes. We compared the performance of the e-ASPECTS software with those of stroke physicians at different professional levels.

METHODS

The baseline computed tomography scans of acute stroke patients, in whom computed tomography and diffusion-weighted imaging scans were obtained less than two hours apart, were retrospectively scored by e-ASPECTS as well as by three stroke experts and three neurology trainees blinded to any clinical information. The ground truth was defined as the ASPECTS on diffusion-weighted imaging scored by another two non-blinded independent experts on consensus basis. Sensitivity and specificity in an ASPECTS region-based and an ASPECTS score-based analysis as well as receiver-operating characteristic curves, Bland-Altman plots with mean score error, and Matthews correlation coefficients were calculated. Comparisons were made between the human scorers and e-ASPECTS with diffusion-weighted imaging being the ground truth. Two methods for clustered data were used to estimate sensitivity and specificity in the region-based analysis.

RESULTS

In total, 34 patients were included and 680 (34 × 20) ASPECTS regions were scored. Mean time from onset to computed tomography was 172 ± 135 min and mean time difference between computed tomographyand magnetic resonance imaging was 41 ± 31 min. The region-based sensitivity (46.46% [CI: 30.8;62.1]) of e-ASPECTS was better than three trainees and one expert (p ≤ 0.01) and not statistically different from another two experts. Specificity (94.15% [CI: 91.7;96.6]) was lower than one expert and one trainee (p < 0.01) and not statistically different to the other four physicians. e-ASPECTS had the best Matthews correlation coefficient of 0.44 (experts: 0.38 ± 0.08 and trainees: 0.19 ± 0.05) and the lowest mean score error of 0.56 (experts: 1.44 ± 1.79 and trainees: 1.97 ± 2.12).

CONCLUSION

e-ASPECTS showed a similar performance to that of stroke experts in the assessment of brain computed tomographys of acute ischemic stroke patients with the Alberta Stroke Program Early CT score method.

Feasibility and Diagnostic Accuracy of Ischemic Stroke Territory Recognition Based on Two-Dimensional Projections of Three-Dimensional Diffusion MRI Data

This study was conducted to assess the feasibility and diagnostic accuracy of brain artery territory recognition based on geoprojected two-dimensional maps of diffusion MRI data in stroke patients. In this retrospective study, multiplanar diffusion MRI data of ischemic stroke patients was used to create a two-dimensional map of the entire brain. To guarantee correct representation of the stroke, a computer-aided brain artery territory diagnosis was developed and tested for its diagnostic accuracy. The test recognized the stroke-affected brain artery territory based on the position of the stroke in the map. The performance of the test was evaluated by comparing it to the reference standard of each patient's diagnosed stroke territory on record. This study was designed and conducted according to Standards for Reporting of Diagnostic Accuracy (STARD). The statistical analysis included diagnostic accuracy parameters, cross-validation, and Youden Index optimization. After cross-validation on a cohort of 91 patients, the sensitivity of this territory diagnosis was 81% with a specificity of 87%. With this, the projection of strokes onto a two-dimensional map is accurate for representing the affected stroke territory and can be used to provide a static and printable overview of the diffusion MRI data. The projected map is compatible with other two-dimensional data such as EEG and will serve as a useful visualization tool.

Recovery of CT stroke hypodensity--An adaptive variational approach

The present research was directed to effective image restoration with the extraction of ischemic edema signs. Computerized support of hyperacute stroke diagnosis based on routinely used computerized tomography (CT) scans was optimized to visualize the infarct extent more precisely. In particular, a beneficial support of time-limited appropriate decision of whether to treat the patient by thrombolysis is expected. Because of a limited accuracy in determining the area of core infarction, particularly in the early hours of symptoms' onset, a variational approach to sensed data recovery was applied. Proposed methodology adjusts fidelity norms and regularization priors integrated with simulated sensing procedures in a compressed sensing framework. Experimental study confirmed almost perfect recognition of ischemic stroke in a test set of over 500 CT scans.

Automated brain computed tomographic densitometry of early ischemic changes in acute stroke

The Alberta Stroke Program Early CT score (ASPECTS) scoring method is frequently used for quantifying early ischemic changes (EICs) in patients with acute ischemic stroke in clinical studies. Varying interobserver agreement has been reported, however, with limited agreement. Therefore, our goal was to develop and evaluate an automated brain densitometric method. It divides CT scans of the brain into ASPECTS regions using atlas-based segmentation. EICs are quantified by comparing the brain density between contralateral sides. This method was optimized and validated using CT data from 10 and 63 patients, respectively. The automated method was validated against manual ASPECTS, stroke severity at baseline and clinical outcome after 7 to 10 days (NIH Stroke Scale, NIHSS) and 3 months (modified Rankin Scale). Manual and automated ASPECTS showed similar and statistically significant correlations with baseline NIHSS ([Formula: see text] and [Formula: see text], respectively) and with follow-up mRS ([Formula: see text] and [Formula: see text]), except for the follow-up NIHSS. Agreement between automated and consensus ASPECTS reading was similar to the interobserver agreement of manual ASPECTS (differences [Formula: see text] point in 73% of cases). The automated ASPECTS method could, therefore, be used as a supplementary tool to assist manual scoring.

Ischemic stroke detection system with a computer-aided diagnostic ability using an unsupervised feature perception enhancement method

We propose an ischemic stroke detection system with a computer-aided diagnostic ability using a four-step unsupervised feature perception enhancement method. In the first step, known as preprocessing, we use a cubic curve contrast enhancement method to enhance image contrast. In the second step, we use a series of methods to extract the brain tissue image area identified during preprocessing. To detect abnormal regions in the brain images, we propose using an unsupervised region growing algorithm to segment the brain tissue area. The brain is centered on a horizontal line and the white matter of the brain's inner ring is split into eight regions. In the third step, we use a coinciding regional location method to find the hybrid area of locations where a stroke may have occurred in each cerebral hemisphere. Finally, we make corrections and mark the stroke area with red color. In the experiment, we tested the system on 90 computed tomography (CT) images from 26 patients, and, with the assistance of two radiologists, we proved that our proposed system has computer-aided diagnostic capabilities. Our results show an increased stroke diagnosis sensitivity of 83% in comparison to 31% when radiologists use conventional diagnostic images.

Hypodensity extractor: a phantom study

We report on the extraction procedures of low-contrast symptomatic hypodensity optimized for a computed tomography-based diagnosis. The specific application is brain imaging with enhanced perception of hypodense areas which are direct symptoms of acute ischemia. A standard low-contrast phantom, as commonly employed in dosimetry and imaging quality evaluation, was used to derive numeric criteria for assessing the extraction effectiveness. Our proposed procedure is based on multiscale analysis of the image data expanded over the frames of wavelets, curvelets or complex wavelets, followed by nonlinear approximation of the symptom signatures. Apparent subtle density changes in the phantom were evaluated using computational metrics and subjective ratings. We discuss the advantages and disadvantages of our proposed optimized hypodensity extraction procedures.

An active contour model for medical image segmentation with application to brain CT image

PURPOSE

Cerebrospinal fluid (CSF) segmentation in computed tomography (CT) is a key step in computer-aided detection (CAD) of acute ischemic stroke. Because of image noise, low contrast and intensity inhomogeneity, CSF segmentation has been a challenging task. A region-based active contour model, which is insensitive to contour initialization and robust to intensity inhomogeneity, was developed for segmenting CSF in brain CT images.

METHODS

The energy function of the region-based active contour model is composed of a range domain kernel function, a space domain kernel function, and an edge indicator function. By minimizing the energy function, the region of edge elements of the target could be automatically identified in images with less dependence on initial contours. The energy function was optimized by means of the deepest descent method with a level set framework. An overlap rate between segmentation results and the reference standard was used to assess the segmentation accuracy. The authors evaluated the performance of the proposed method on both synthetic data and real brain CT images. They also compared the performance level of our method to those of region-scalable fitting (RSF) and global convex segment (GCS) models.

RESULTS

For the experiment of CSF segmentation in 67 brain CT images, their method achieved an average overlap rate of 66% compared to the average overlap rates of 16% and 46% from the RSF model and the GCS model, respectively.

CONCLUSIONS

Their region-based active contour model has the ability to achieve accurate segmentation results in images with high noise level and intensity inhomogeneity. Therefore, their method has great potential in the segmentation of medical images and would be useful for developing CAD schemes for acute ischemic stroke in brain CT images.