Ensemble of Convolutional Neural Networks Improves Automated Segmentation of Acute Ischemic Lesions Using Multiparametric Diffusion-Weighted MRI

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.

Development of statistical auto-segmentation method for diffusion restriction gray matter lesions in patients with newly diagnosed sporadic Creutzfeldt-Jakob disease

Quantification of diffusion restriction lesions in sporadic Creutzfeldt-Jakob disease (sCJD) may provide information of the disease burden. We aim to develop an automatic segmentation model for sCJD and to evaluate the volume of disease extent as a prognostic marker for overall survival. Fifty-six patients (mean age ± SD, 61.2 ± 9.9 years) were included from February 2000 to July 2020. A threshold-based segmentation was used to obtain abnormal signal intensity masks. Segmented volumes were compared with the visual grade. The Dice similarity coefficient was calculated to measure the similarity between the automatic vs. manual segmentation. Cox proportional hazards regression analysis was performed to evaluate the volume of disease extent as a prognostic marker. The automatic segmentation showed good correlation with the visual grading. The cortical lesion volumes significantly increased as the visual grade aggravated (extensive: 112.9 ± 73.2; moderate: 45.4 ± 30.4; minimal involvement: 29.6 ± 18.1 mm3) (P < 0.001). The deep gray matter lesion volumes were significantly higher for positive than for negative involvement of the deep gray matter (5.6 ± 4.6 mm3 vs. 1.0 ± 1.3 mm3, P < 0.001). The mean Dice similarity coefficients were 0.90 and 0.94 for cortical and deep gray matter lesions, respectively. However, the volume of disease extent was not associated with worse overall survival (cortical extent: P = 0.07; deep gray matter extent: P = 0.12).

Scaling behaviours of deep learning and linear algorithms for the prediction of stroke severity

Deep learning has allowed for remarkable progress in many medical scenarios. Deep learning prediction models often require 105-107 examples. It is currently unknown whether deep learning can also enhance predictions of symptoms post-stroke in real-world samples of stroke patients that are often several magnitudes smaller. Such stroke outcome predictions however could be particularly instrumental in guiding acute clinical and rehabilitation care decisions. We here compared the capacities of classically used linear and novel deep learning algorithms in their prediction of stroke severity. Our analyses relied on a total of 1430 patients assembled from the MRI-Genetics Interface Exploration collaboration and a Massachusetts General Hospital-based study. The outcome of interest was National Institutes of Health Stroke Scale-based stroke severity in the acute phase after ischaemic stroke onset, which we predict by means of MRI-derived lesion location. We automatically derived lesion segmentations from diffusion-weighted clinical MRI scans, performed spatial normalization and included a principal component analysis step, retaining 95% of the variance of the original data. We then repeatedly separated a train, validation and test set to investigate the effects of sample size; we subsampled the train set to 100, 300 and 900 and trained the algorithms to predict the stroke severity score for each sample size with regularized linear regression and an eight-layered neural network. We selected hyperparameters on the validation set. We evaluated model performance based on the explained variance (R2) in the test set. While linear regression performed significantly better for a sample size of 100 patients, deep learning started to significantly outperform linear regression when trained on 900 patients. Average prediction performance improved by ∼20% when increasing the sample size 9× [maximum for 100 patients: 0.279 ± 0.005 (R2, 95% confidence interval), 900 patients: 0.337 ± 0.006]. In summary, for sample sizes of 900 patients, deep learning showed a higher prediction performance than typically employed linear methods. These findings suggest the existence of non-linear relationships between lesion location and stroke severity that can be utilized for an improved prediction performance for larger sample sizes.

Selective ensemble methods for deep learning segmentation of major vessels in invasive coronary angiography

BACKGROUND

Invasive coronary angiography (ICA) is a primary imaging modality that visualizes the lumen area of coronary arteries for diagnosis and interventional guidance. In the current practice of quantitative coronary analysis (QCA), semi-automatic segmentation tools require labor-intensive and time-consuming manual correction, limiting their application in the catheterization room.

PURPOSE

This study aims to propose rank-based selective ensemble methods that improve the segmentation performance and reduce morphological errors that limit fully automated quantification of coronary artery using deep-learning segmentation of ICA.

METHODS

Two selective ensemble methods proposed in this work integrated the weighted ensemble approach with per-image quality estimation. The segmentation outcomes from five base models with different loss functions were ranked either by mask morphology or estimated dice similarity coefficient (DSC). The final output was determined by imposing different weights according to the ranks. The ranking criteria based on mask morphology were formulated from empirical insight to avoid frequent types of segmentation errors (MSEN), while the estimation of DSCs was performed by comparing the pseudo-ground truth generated from a meta-learner (ESEN). Five-fold cross-validation was performed with the internal dataset of 7426 coronary angiograms from 2924 patients, and prediction model was externally validated with 556 images of 226 patients.

RESULTS

The selective ensemble methods improved the segmentation performance with DSCs up to 93.07% and provided a better delineation of coronary lesion with local DSCs of up to 93.93%, outperforming all individual models. Proposed methods also minimized the chances of mask disconnection in the most narrowed regions to 2.10%. The robustness of the proposed methods was also evident in the external validation. Inference time for major vessel segmentation was approximately one-sixth of a second.

CONCLUSION

Proposed methods successfully reduced morphological errors in the predicted masks and were able to enhance the robustness of the automatic segmentation. The results suggest better applicability of real-time QCA-based diagnostic methods in routine clinical settings.

The role of input imaging combination and ADC threshold on segmentation of acute ischemic stroke lesion using U-Net

BACKGROUND

To evaluate the effect of the weighting of input imaging combo and ADC threshold on the performance of the U-Net and to find an optimized input imaging combo and ADC threshold in segmenting acute ischemic stroke (AIS) lesion.

METHODS

This study retrospectively enrolled a total of 212 patients having AIS. Four combos, including ADC-ADC-ADC (AAA), DWI-ADC-ADC (DAA), DWI-DWI-ADC (DDA), and DWI-DWI-DWI (DDD), were used as input images, respectively. Three ADC thresholds including 0.6, 0.8 and 1.8 × 10-3 mm2/s were applied. Dice similarity coefficient (DSC) was used to evaluate the segmentation performance of U-Nets. Nonparametric Kruskal-Wallis test with Tukey-Kramer post-hoc tests were used for comparison. A p < .05 was considered statistically significant.

RESULTS

The DSC significantly varied among different combos of images and different ADC thresholds. Hybrid U-Nets outperformed uniform U-Nets at ADC thresholds of 0.6 × 10-3 mm2/s and 0.8 × 10-3 mm2/s (p < .001). The U-Net with imaging combo of DDD had segmentation performance similar to hybrid U-Nets at an ADC threshold of 1.8 × 10-3 mm2/s (p = .062 to 1). The U-Net using the imaging combo of DAA at the ADC threshold of 0.6 × 10-3 mm2/s achieved the highest DSC in the segmentation of AIS lesion.

CONCLUSIONS

The segmentation performance of U-Net for AIS varies among the input imaging combos and ADC thresholds. The U-Net is optimized by choosing the imaging combo of DAA at an ADC threshold of 0.6 × 10-3 mm2/s in segmentating AIS lesion with highest DSC.

KEY POINTS

• Segmentation performance of U-Net for AIS differs among input imaging combos. • Segmentation performance of U-Net for AIS differs among ADC thresholds. • U-Net is optimized using DAA with ADC = 0.6 × 10-3 mm2/s.

Artificial Intelligence for Automated DWI/FLAIR Mismatch Assessment on Magnetic Resonance Imaging in Stroke: A Systematic Review

We conducted this Systematic Review to create an overview of the currently existing Artificial Intelligence (AI) methods for Magnetic Resonance Diffusion-Weighted Imaging (DWI)/Fluid-Attenuated Inversion Recovery (FLAIR)-mismatch assessment and to determine how well DWI/FLAIR mismatch algorithms perform compared to domain experts. We searched PubMed Medline, Ovid Embase, Scopus, Web of Science, Cochrane, and IEEE Xplore literature databases for relevant studies published between 1 January 2017 and 20 November 2022, following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. We assessed the included studies using the Quality Assessment of Diagnostic Accuracy Studies 2 tool. Five studies fit the scope of this review. The area under the curve ranged from 0.74 to 0.90. The sensitivity and specificity ranged from 0.70 to 0.85 and 0.74 to 0.84, respectively. Negative predictive value, positive predictive value, and accuracy ranged from 0.55 to 0.82, 0.74 to 0.91, and 0.73 to 0.83, respectively. In a binary classification of ±4.5 h from stroke onset, the surveyed AI methods performed equivalent to or even better than domain experts. However, using the relation between time since stroke onset (TSS) and increasing visibility of FLAIR hyperintensity lesions is not recommended for the determination of TSS within the first 4.5 h. An AI algorithm on DWI/FLAIR mismatch assessment focused on treatment eligibility, outcome prediction, and consideration of patient-specific data could potentially increase the proportion of stroke patients with unknown onset who could be treated with thrombolysis.

The relevance of rich club regions for functional outcome post-stroke is enhanced in women

This study aimed to investigate the influence of stroke lesions in predefined highly interconnected (rich-club) brain regions on functional outcome post-stroke, determine their spatial specificity and explore the effects of biological sex on their relevance. We analyzed MRI data recorded at index stroke and ~3-months modified Rankin Scale (mRS) data from patients with acute ischemic stroke enrolled in the multisite MRI-GENIE study. Spatially normalized structural stroke lesions were parcellated into 108 atlas-defined bilateral (sub)cortical brain regions. Unfavorable outcome (mRS > 2) was modeled in a Bayesian logistic regression framework. Effects of individual brain regions were captured as two compound effects for (i) six bilateral rich club and (ii) all further non-rich club regions. In spatial specificity analyses, we randomized the split into "rich club" and "non-rich club" regions and compared the effect of the actual rich club regions to the distribution of effects from 1000 combinations of six random regions. In sex-specific analyses, we introduced an additional hierarchical level in our model structure to compare male and female-specific rich club effects. A total of 822 patients (age: 64.7[15.0], 39% women) were analyzed. Rich club regions had substantial relevance in explaining unfavorable functional outcome (mean of posterior distribution: 0.08, area under the curve: 0.8). In particular, the rich club-combination had a higher relevance than 98.4% of random constellations. Rich club regions were substantially more important in explaining long-term outcome in women than in men. All in all, lesions in rich club regions were associated with increased odds of unfavorable outcome. These effects were spatially specific and more pronounced in women.

Automatic identification of early ischemic lesions on non-contrast CT with deep learning approach

Early ischemic lesion on non-contrast computed tomogram (NCCT) in acute stroke can be subtle and need confirmation with magnetic resonance (MR) image for treatment decision-making. We retrospectively included the NCCT slices of 129 normal subjects and 546 ischemic stroke patients (onset < 12 h) with corresponding MR slices as reference standard from a prospective registry of Chang Gung Research Databank. In model selection, NCCT slices were preprocessed and fed into five different pre-trained convolutional neural network (CNN) models including Visual Geometry Group 16 (VGG16), Residual Networks 50, Inception-ResNet-v2, Inception-v3, and Inception-v4. In model derivation, the customized-VGG16 model could achieve an accuracy of 0.83, sensitivity 0.85, F-score 0.80, specificity 0.82, and AP 0.82 after using a tenfold cross-validation method, outperforming the pre-trained VGG16 model. In model evaluation, the customized-VGG16 model could correctly identify 53 in 58 subjects (91.37%) including 29 ischemic stroke patients and 24 normal subjects and reached the sensitivity of 86.95% in identifying ischemic NCCT slices (200/230), irrespective of supratentorial or infratentorial lesions. The customized-VGG16 CNN model can successfully identify the presence of early ischemic lesions on NCCT slices using the concept of automatic feature learning. Further study will be proceeded to detect the location of ischemic lesion.

Application of Machine Learning Techniques for Characterization of Ischemic Stroke with MRI Images: A Review

Magnetic resonance imaging (MRI) is a standard tool for the diagnosis of stroke, but its manual interpretation by experts is arduous and time-consuming. Thus, there is a need for computer-aided-diagnosis (CAD) models for the automatic segmentation and classification of stroke on brain MRI. The heterogeneity of stroke pathogenesis, morphology, image acquisition modalities, sequences, and intralesional tissue signal intensity, as well as lesion-to-normal tissue contrast, pose significant challenges to the development of such systems. Machine learning (ML) is increasingly being used in predictive neuroimaging diagnosis and prognostication. This paper reviews image processing and machine learning techniques that have been applied to detect ischemic stroke on brain MRI, including details on image acquisition, pre-processing, techniques to segment, extraction of features, and classification into stroke types. The main objective of this work is to find the state-of-art machine learning techniques used to predict the ischemic stroke and their application in clinical set-up. The article selection is performed according to PRISMA guideline. The state-of-the-art on automated MRI stroke diagnosis, with a focus on machine learning, is discussed, along with its advantages and limitations. We found that the various machine learning models discussed in this article are able to detect the infarcts with an acceptable accuracy of 70–90%. However, no one has highlighted the time complexity to predict the stroke in the model developed, which is an important factor. The work concludes with proposals for future recommendations for building efficient and robust deep learning (DL) models for quantitative brain MRI analysis. In recent work, with the application of DL approaches, using large datasets to train the models has improved the detection accuracy and reduced computational complexity. We suggest that the design of a decision support system based on artificial intelligence (AI) and clinical data presenting symptoms is essential to support clinicians to accelerate diagnosis and timeous therapy in the emergency management of stroke.

Face mask recognition system using CNN model

COVID-19 epidemic has swiftly disrupted our day-to-day lives affecting the international trade and movements. Wearing a face mask to protect one's face has become the new normal. In the near future, many public service providers will expect the clients to wear masks appropriately to partake of their services. Therefore, face mask detection has become a critical duty to aid worldwide civilization. This paper provides a simple way to achieve this objective utilising some fundamental Machine Learning tools as TensorFlow, Keras, OpenCV and Scikit-Learn. The suggested technique successfully recognises the face in the image or video and then determines whether or not it has a mask on it. As a surveillance job performer, it can also recognise a face together with a mask in motion as well as in a video. The technique attains excellent accuracy. We investigate optimal parameter values for the Convolutional Neural Network model (CNN) in order to identify the existence of masks accurately without generating over-fitting.

Semantic segmentation guided detector for segmentation, classification, and lesion mapping of acute ischemic stroke in MRI images

BACKGROUND AND PURPOSE

MRI images timely and accurately reflect ischemic injuries to the brain tissues and, therefore, can support clinical decision-making of acute ischemic stroke (AIS). To maximize the information provided by the MRI images, we leverage deep learning models to segment, classify, and map lesion distributions of AIS.

METHODS

We evaluated brain MRI images of AIS patients from 2017 to 2020 at a tertiary teaching hospital and developed the Semantic Segmentation Guided Detector Network (SGD-Net), composed of the first U-shaped model for segmentation in diffusion-weighted imaging (DWI) and the second model for binary classification of lesion size (lacune vs. non-lacune) and circulatory territory of lesion location (anterior vs. posterior circulation). Next, we modified the two-stage deep learning model into SGD-Net Plus by automatically segmenting AIS lesions in DWI images and registering the lesion in T1-weighted images and the brain atlases.

RESULTS

The final enrollment (216 patients with 4606 slices) was divided into 80% for model development and 20% for testing. S1 model segmented AIS lesions in DWI images accurately with a pixel accuracy > 99% (Dice 0.806-0.828 and IoU 0.675-707). In comprehensive evaluation of classification performance, the two-stage SGD-Net outperformed the traditional one-stage models in classifying AIS lesion size (accuracy 0.867-0.956 vs. 0.511-0.867, AUROC 0.962-0.992 vs. 0.528-0.937, AUPRC 0.964-0.994 vs. 0.549-0.938) and location (accuracy 0.860-0.930 vs. 0.326-0.721, AUROC 0.936-0.988 vs. 0.493-0.833, AUPRC 0.883-0.978 vs. 0.365-0.695). The precise lesion segmentation at the first stage of the deep learning model was the basis for further application. After that, the modified two-stage model SGD-Net Plus accurately reported the volume, region percentage, and lesion percentage of each region on the selected brain atlas. Its reports provided clear descriptions and quantifications of the AIS-related brain injuries on white matter tracts, Brodmann areas, and cytoarchitectonic areas.

CONCLUSION

Domain knowledge-oriented design of artificial intelligence applications can deepen our understanding of patients' conditions and strengthen the use of MRI for patient care. SGD-Net precisely segments AIS lesions on DWI and accurately classifies the lesions. In addition, SGD-Net Plus maps the AIS lesions and quantifies their occupancy in each brain region. They are practical tools to meet the clinical needs and enrich educational resources of neuroimage.

Automatic Segmentation in Acute Ischemic Stroke: Prognostic Significance of Topological Stroke Volumes on Stroke Outcome

BACKGROUND

Stroke infarct volume predicts patient disability and has utility for clinical trial outcomes. Accurate infarct volume measurement requires manual segmentation of stroke boundaries in diffusion-weighted magnetic resonance imaging scans which is time-consuming and subject to variability. Automatic infarct segmentation should be robust to rotation and reflection; however, prior work has not encoded this property into deep learning architecture. Here, we use rotation-reflection equivariance and train a deep learning model to segment stroke volumes in a large cohort of well-characterized patients with acute ischemic stroke in different vascular territories.

METHODS

In this retrospective study, patients were selected from a stroke registry at Houston Methodist Hospital. Eight hundred seventy-five patients with acute ischemic stroke in any brain area who had magnetic resonance imaging with diffusion-weighted imaging were included for analysis and split 80/20 for training/testing. Infarct volumes were manually segmented by consensus of 3 independent clinical experts and cross-referenced against radiology reports. A rotation-reflection equivariant model was developed based on U-Net and grouped convolutions. Segmentation performance was evaluated using Dice score, precision, and recall. Ninety-day modified Rankin Scale outcome prediction was also evaluated using clinical variables and segmented stroke volumes in different brain regions.

RESULTS

Segmentation model Dice scores are 0.88 (95% CI, 0.87-0.89; training) and 0.85 (0.82-0.88; testing). The modified Rankin Scale outcome prediction AUC using stroke volume in 30 refined brain regions based upon modified Rankin Scale-relevance areas adjusted for clinical variables was 0.80 (0.76-0.83) with an accuracy of 0.75 (0.72-0.78).

CONCLUSIONS

We trained a deep learning model with encoded rotation-reflection equivariance to segment acute ischemic stroke lesions in diffusion- weighted imaging using a large data set from the Houston Methodist stroke center. The model achieved competitive performance in 175 well-balanced hold-out testing cases that include strokes from different vascular territories. Furthermore, the location specific stroke volume segmentations from the deep learning model combined with clinical factors demonstrated high AUC and accuracy for 90-day modified Rankin Scale in an outcome prediction model.

Improving interobserver agreement and performance of deep learning models for segmenting acute ischemic stroke by combining DWI with optimized ADC thresholds

OBJECTIVES

To examine the role of ADC threshold on agreement across observers and deep learning models (DLMs) plus segmentation performance of DLMs for acute ischemic stroke (AIS).

METHODS

Twelve DLMs, which were trained on DWI-ADC-ADC combination from 76 patients with AIS using 6 different ADC thresholds with ground truth manually contoured by 2 observers, were tested by additional 67 patients in the same hospital and another 78 patients in another hospital. Agreement between observers and DLMs were evaluated by Bland-Altman plot and intraclass correlation coefficient (ICC). The similarity between ground truth (GT) defined by observers and between automatic segmentation performed by DLMs was evaluated by Dice similarity coefficient (DSC). Group comparison was performed using the Mann-Whitney U test. The relationship between the DSC and ADC threshold as well as AIS lesion size was evaluated by linear regression analysis. A p < .05 was considered statistically significant.

RESULTS

Excellent interobserver agreement and intraobserver repeatability in the manual segmentation (all ICC > 0.98, p < .001) were achieved. The 95% limit of agreement was reduced from 11.23 cm² for GT on DWI to 0.59 cm² for prediction at an ADC threshold of 0.6 × 10^-3 mm²/s combined with DWI. The segmentation performance of DLMs was improved with an overall DSC from 0.738 ± 0.214 on DWI to 0.971 ± 0.021 on an ADC threshold of 0.6 × 10^-3 mm²/s combined with DWI.

CONCLUSIONS

Combining an ADC threshold of 0.6 × 10^-3 mm²/s with DWI reduces interobserver and inter-DLM difference and achieves best segmentation performance of AIS lesions using DLMs.

KEY POINTS

• Higher Dice similarity coefficient (DSC) in predicting acute ischemic stroke lesions was achieved by ADC thresholds combined with DWI than by DWI alone (all p < .05). • DSC had a negative association with the ADC threshold in most sizes, both hospitals, and both observers (most p < .05) and a positive association with the stroke size in all ADC thresholds, both hospitals, and both observers (all p < .001). • An ADC threshold of 0.6 × 10^-3 mm²/s eliminated the difference of DSC at any stroke size between observers or between hospitals (p = .07 to > .99).

Development and clinical application of a deep learning model to identify acute infarct on magnetic resonance imaging

Stroke is a leading cause of death and disability. The ability to quickly identify the presence of acute infarct and quantify the volume on magnetic resonance imaging (MRI) has important treatment implications. We developed a machine learning model that used the apparent diffusion coefficient and diffusion weighted imaging series. It was trained on 6,657 MRI studies from Massachusetts General Hospital (MGH; Boston, USA). All studies were labelled positive or negative for infarct (classification annotation) with 377 having the region of interest outlined (segmentation annotation). The different annotation types facilitated training on more studies while not requiring the extensive time to manually segment every study. We initially validated the model on studies sequestered from the training set. We then tested the model on studies from three clinical scenarios: consecutive stroke team activations for 6-months at MGH, consecutive stroke team activations for 6-months at a hospital that did not provide training data (Brigham and Women’s Hospital [BWH]; Boston, USA), and an international site (Diagnósticos da América SA [DASA]; Brazil). The model results were compared to radiologist ground truth interpretations. The model performed better when trained on classification and segmentation annotations (area under the receiver operating curve [AUROC] 0.995 [95% CI 0.992–0.998] and median Dice coefficient for segmentation overlap of 0.797 [IQR 0.642–0.861]) compared to segmentation annotations alone (AUROC 0.982 [95% CI 0.972–0.990] and Dice coefficient 0.776 [IQR 0.584–0.857]). The model accurately identified infarcts for MGH stroke team activations (AUROC 0.964 [95% CI 0.943–0.982], 381 studies), BWH stroke team activations (AUROC 0.981 [95% CI 0.966–0.993], 247 studies), and at DASA (AUROC 0.998 [95% CI 0.993–1.000], 171 studies). The model accurately segmented infarcts with Pearson correlation comparing model output and ground truth volumes between 0.968 and 0.986 for the three scenarios. Acute infarct can be accurately detected and segmented on MRI in real-world clinical scenarios using a machine learning model.

A Review on Computer Aided Diagnosis of Acute Brain Stroke

Amongst the most common causes of death globally, stroke is one of top three affecting over 100 million people worldwide annually. There are two classes of stroke, namely ischemic stroke (due to impairment of blood supply, accounting for ~70% of all strokes) and hemorrhagic stroke (due to bleeding), both of which can result, if untreated, in permanently damaged brain tissue. The discovery that the affected brain tissue (i.e., 'ischemic penumbra') can be salvaged from permanent damage and the bourgeoning growth in computer aided diagnosis has led to major advances in stroke management. Abiding to the Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA) guidelines, we have surveyed a total of 177 research papers published between 2010 and 2021 to highlight the current status and challenges faced by computer aided diagnosis (CAD), machine learning (ML) and deep learning (DL) based techniques for CT and MRI as prime modalities for stroke detection and lesion region segmentation. This work concludes by showcasing the current requirement of this domain, the preferred modality, and prospective research areas.

Comparison of domain adaptation techniques for white matter hyperintensity segmentation in brain MR images

Robust automated segmentation of white matter hyperintensities (WMHs) in different datasets (domains) is highly challenging due to differences in acquisition (scanner, sequence), population (WMH amount and location) and limited availability of manual segmentations to train supervised algorithms. In this work we explore various domain adaptation techniques such as transfer learning and domain adversarial learning methods, including domain adversarial neural networks and domain unlearning, to improve the generalisability of our recently proposed triplanar ensemble network, which is our baseline model. We used datasets with variations in intensity profile, lesion characteristics and acquired using different scanners. For the source domain, we considered a dataset consisting of data acquired from 3 different scanners, while the target domain consisted of 2 datasets. We evaluated the domain adaptation techniques on the target domain datasets, and additionally evaluated the performance on the source domain test dataset for the adversarial techniques. For transfer learning, we also studied various training options such as minimal number of unfrozen layers and subjects required for fine-tuning in the target domain. On comparing the performance of different techniques on the target dataset, domain adversarial training of neural network gave the best performance, making the technique promising for robust WMH segmentation.

Applications of artificial intelligence for DWI and PWI data processing in acute ischemic stroke: Current practices and future directions

Multimodal Magnetic Resonance Imaging (MRI) techniques of Perfusion-Weighted Imaging (PWI) and Diffusion-Weighted Imaging (DWI) data are integral parts of the diagnostic workup in the acute stroke setting. The visual interpretation of PWI/DWI data is the most likely procedure to triage Acute Ischemic Stroke (AIS) patients who will access reperfusion therapy, especially in those exceeding 6 h of stroke onset. In fact, this process defines two classes of tissue: the ischemic core, which is presumed to be irreversibly damaged, visualized on DWI data and the penumbra which is the reversibly injured brain tissue around the ischemic tissue, visualized on PWI data. AIS patients with a large ischemic penumbra and limited infarction core have a high probability of benefiting from endovascular treatment. However, it is a tedious and time-consuming procedure. Consequently, it is subject to high inter- and intra-observer variability. Thus, the assessment of the potential risks and benefits of endovascular treatment is uncertain. Fast, accurate and automatic post-processing of PWI and DWI data is important for clinical diagnosis and is necessary to help the decision making for therapy. Therefore, an automated procedure that identifies stroke slices, stroke hemisphere, segments stroke regions in DWI, and measures hypoperfused tissue in PWI enhances considerably the reproducibility and the accuracy of stroke assessment. In this work, we draw an overview of several applications of Artificial Intelligence (AI) for the automation processing and their potential contributions in clinical practices. We compare the current approaches among each other's with respect to some key requirements.

Convolutional Neural Network-Processed MRI Images in the Diagnosis of Plastic Bronchitis in Children

Objective

The study focused on the features of the convolutional neural networks- (CNN-) processed magnetic resonance imaging (MRI) images for plastic bronchitis (PB) in children.

Methods

30 PB children were selected as subjects, including 19 boys and 11 girls. They all received the MRI examination for the chest. Then, a CNN-based algorithm was constructed and compared with Active Appearance Model (AAM) algorithm for segmentation effects of MRI images in 30 PB children, factoring into occurring simultaneously than (OST), Dice, and Jaccard coefficient.

Results

The maximum Dice coefficient of CNN algorithm reached 0.946, while that of active AAM was 0.843, and the Jaccard coefficient of CNN algorithm was also higher (0.894 vs. 0.758, P < 0.05). The MRI images showed pulmonary inflammation in all subjects. Of 30 patients, 14 (46.66%) had complicated pulmonary atelectasis, 9 (30%) had the complicated pleural effusion, 3 (10%) had pneumothorax, 2 (6.67%) had complicated mediastinal emphysema, and 2 (6.67%) had complicated pneumopericardium. Also, of 30 patients, 19 (63.33%) had lung consolidation and atelectasis in a single lung lobe and 11 (36.67%) in both two lung lobes.

Conclusion

The algorithm based on CNN can significantly improve the segmentation accuracy of MRI images for plastic bronchitis in children. The pleural effusion was a dangerous factor for the occurrence and development of PB.

Improving Ischemic Stroke Care With MRI and Deep Learning Artificial Intelligence

ABSTRACT

Advanced magnetic resonance imaging has been used as selection criteria for both acute ischemic stroke treatment and secondary prevention. The use of artificial intelligence, and in particular, deep learning, to synthesize large amounts of data and to understand better how clinical and imaging data can be leveraged to improve stroke care promises a new era of stroke care. In this article, we review common deep learning model structures for stroke imaging, evaluation metrics for model performance, and studies that investigated deep learning application in acute ischemic stroke care and secondary prevention.

Inter-vendor performance of deep learning in segmenting acute ischemic lesions on diffusion-weighted imaging: a multicenter study

There is little evidence on the applicability of deep learning (DL) in the segmentation of acute ischemic lesions on diffusion-weighted imaging (DWI) between magnetic resonance imaging (MRI) scanners of different manufacturers. We retrospectively included DWI data of patients with acute ischemic lesions from six centers. Dataset A (n = 2986) and B (n = 3951) included data from Siemens and GE MRI scanners, respectively. The datasets were split into the training (80%), validation (10%), and internal test (10%) sets, and six neuroradiologists created ground-truth masks. Models A and B were the proposed neural networks trained on datasets A and B. The models subsequently fine-tuned across the datasets using their validation data. Another radiologist performed the segmentation on the test sets for comparisons. The median Dice scores of models A and B were 0.858 and 0.857 for the internal tests, which were non-inferior to the radiologist’s performance, but demonstrated lower performance than the radiologist on the external tests. Fine-tuned models A and B achieved median Dice scores of 0.832 and 0.846, which were non-inferior to the radiologist's performance on the external tests. The present work shows that the inter-vendor operability of deep learning for the segmentation of ischemic lesions on DWI might be enhanced via transfer learning; thereby, their clinical applicability and generalizability could be improved.

A Detailed Analysis of Infarct Patterns and Volumes at 24-hour Noncontrast CT and Diffusion-weighted MRI in Acute Ischemic Stroke Due to Large Vessel Occlusion: Results from the ESCAPE-NA1 Trial

Background The effect of infarct pattern on functional outcome in acute ischemic stroke is incompletely understood. Purpose To investigate the association of qualitative and quantitative infarct variables at 24-hour follow-up noncontrast CT and diffusion-weighted MRI with 90-day clinical outcome. Materials and Methods The Safety and Efficacy of Nerinetide in Subjects Undergoing Endovascular Thrombectomy for Stroke, or ESCAPE-NA1, randomized controlled trial enrolled patients with large-vessel-occlusion stroke undergoing mechanical thrombectomy from March 1, 2017, to August 12, 2019. In this post hoc analysis of the trial, qualitative infarct variables (predominantly gray [vs gray and white] matter involvement, corticospinal tract involvement, infarct structure [scattered vs territorial]) and total infarct volume were assessed at 24-hour follow-up noncontrast CT or diffusion-weighted MRI. White and gray matter infarct volumes were assessed in patients by using follow-up diffusion-weighted MRI. Infarct variables were compared between patients with and those without good outcome, defined as a modified Rankin Scale score of 0-2 at 90 days. The association of infarct variables with good outcome was determined with use of multivariable logistic regression. Separate regression models were used to report effect size estimates with adjustment for total infarct volume. Results Qualitative infarct variables were assessed in 1026 patients (mean age ± standard deviation, 69 years ± 13; 522 men) and quantitative infarct variables were assessed in a subgroup of 358 of 1026 patients (mean age, 67 years ± 13; 190 women). Patients with gray and white matter involvement (odds ratio [OR] after multivariable adjustment, 0.19; 95% CI: 0.14, 0.25; P < .001), corticospinal tract involvement (OR after multivariable adjustment, 0.06; 95% CI: 0.04, 0.10; P < .001), and territorial infarcts (OR after multivariable adjustment, 0.22; 95% CI: 0.14, 0.32; P < .001) were less likely to achieve good outcome, independent of total infarct volume. Conclusion Infarct confinement to the gray matter, corticospinal tract sparing, and scattered infarct structure at 24-hour noncontrast CT and diffusion-weighted MRI were highly predictive of good 90-day clinical outcome, independent of total infarct volume. Clinical trial registration no. NCT02930018 © RSNA, 2021 Online supplemental material is available for this article. See also the editorial by Mossa-Basha in this issue.

Diffusion weighted imaging in acute ischemic stroke: A review of its interpretation pitfalls and advanced diffusion imaging application

Diffusion weighted imaging (DWI) is a widely used imaging technique to evaluate patients with stroke. It can detect brain ischemia within minutes of stroke onset. However, DWI has few potential pitfalls that should be recognized during interpretation. DWI lesion could be reversible in the early hours of stroke and the entire lesion may not represent ischemic core. False negative DWI could lead to diagnosis of DWI negative stroke or to a missed stroke diagnosis. Ischemic stroke mimics can occur on DWI with non-cerebrovascular neurological conditions. In this article, the history of DWI, its clinical applications, and potential pitfalls for use in acute ischemic stroke are reviewed. Advanced diffusion imaging techniques with reference to Diffusion Kurtosis Imaging and Diffusion Tensor Imaging that has been studied to evaluate ischemic core are discussed.

Deep neural network ensemble for on-the-fly quality control-driven segmentation of cardiac MRI T1 mapping

•

Quality control-driven framework for cardiac segmentation and quality control.

•

Exploiting variability within deep neural network ensemble to estimate uncertainty.

•

Novel on-the-fly selection mechanism for the final optimal segmentation.

•

Accurate, reliable, and fully automated analysis of T1 map with visualization.

•

Highlighting a potential flaw of the Pearson correlation to evaluate quality score.

Recent developments in artificial intelligence have generated increasing interest to deploy automated image analysis for diagnostic imaging and large-scale clinical applications. However, inaccuracy from automated methods could lead to incorrect conclusions, diagnoses or even harm to patients. Manual inspection for potential inaccuracies is labor-intensive and time-consuming, hampering progress towards fast and accurate clinical reporting in high volumes. To promote reliable fully-automated image analysis, we propose a quality control-driven (QCD) segmentation framework. It is an ensemble of neural networks that integrate image analysis and quality control. The novelty of this framework is the selection of the most optimal segmentation based on predicted segmentation accuracy, on-the-fly. Additionally, this framework visualizes segmentation agreement to provide traceability of the quality control process. In this work, we demonstrated the utility of the framework in cardiovascular magnetic resonance T1-mapping - a quantitative technique for myocardial tissue characterization. The framework achieved near-perfect agreement with expert image analysts in estimating myocardial T1 value (r=0.987,p<.0005; mean absolute error (MAE)=11.3ms), with accurate segmentation quality prediction (Dice coefficient prediction MAE=0.0339) and classification (accuracy=0.99), and a fast average processing time of 0.39 second/image. In summary, the QCD framework can generate high-throughput automated image analysis with speed and accuracy that is highly desirable for large-scale clinical applications.

Neuroimaging and deep learning for brain stroke detection - A review of recent advancements and future prospects

BACKGROUND AND OBJECTIVE

In recent years, deep learning algorithms have created a massive impact on addressing research challenges in different domains. The medical field also greatly benefits from the use of improving deep learning models which save time and produce accurate results. This research aims to emphasize the impact of deep learning models in brain stroke detection and lesion segmentation. This is achieved by discussing the state of the art approaches proposed by the recent works in this field.

METHODS

In this study, the advancements in stroke lesion detection and segmentation were focused. The survey analyses 113 research papers published in different academic research databases. The research articles have been filtered out based on specific criteria to obtain the most prominent insights related to stroke lesion detection and segmentation.

RESULTS

The features of the stroke lesion vary based on the type of imaging modality. To develop an effective method for stroke lesion detection, the features need to be carefully extracted from the input images. This review takes an attempt to categorize and discuss the different deep architectures employed for stroke lesion detection and segmentation, based on the underlying imaging modality. This further assists in understanding the relevance of the two-deep neural network components in medical image analysis namely Convolutional Neural Network (CNN) and Fully Convolutional Network (FCN). It hints at other possible deep architectures that can be proposed for better results towards stroke lesion detection. Also, the emerging trends and breakthroughs in stroke detection have been detailed in this evaluation.

CONCLUSION

This work concludes by examining the technical and non-technical challenges faced by researchers and indicate the future implications in stroke detection. It could support the bio-medical researchers to propose better solutions for stroke lesion detection.

Identification of White Matter Lesions in Patients With Acute Ischemic Lesions Using U-net

Background: White matter lesions (WML) have been proved to be significantly associated with many brain diseases. Precise evaluation of burden of WML at early stage could provide insights in the prognosis and assist in intervention. However, acute ischemic lesions (AIL) exhibit hyperintensities on FLAIR images either, and are detected by diffusion weighted imaging (DWI). It is challenging to identify and segment WML in the patients with WML and AIL. Convolutional neural network (CNN) based architecture has been validated as an efficient tool for automatic segmentation. This study aimed to evaluate the performance of U-net in evaluation of WML in the patients with WML and AIL.

Methods: A total of 208 cases from Chinese Atherosclerosis Risk Evaluation (CARE II) study were recruited in the present study. All subjects underwent imaging of FLAIR and DWI on 3.0 Tesla scanners. The contours of WML delineated by the observer and its scores rated by the observer were considered as gold standard. Among all 208 cases, 108 were randomly selected as train set, and the remaining 100 cases were used as test set. The performance of lesion segmentation toolbox (LST) and three U-net models were evaluated on three levels: pixel, lesion, and subject levels. The performance of all methods in WML identification and segmentation was also evaluated among the cases with different lesion volumes and between the cases with and without AIL.

Results: All U-net models outperformed LST on pixel, lesion, and subject levels, while no differences were found among three U-net models. All segmentation methods performed best in the cases with WML volume (WMLV) > 20 ml but worst in those with WMLV < 5 ml. In addition, all methods showed similar performance between the cases with and without AIL. The scores determined by U-net exhibited a strong correlation with the gold standard (all Spearman correlation coefficients >0.89, ICCs >0.88, p-values <0.001).

Conclusion: U-net performs well on identification and segmentation of WML in the patients with WML and AIL. The performance of U-net is validated by a dataset of multicenter study. Our results indicate that U-net has an advantage in assessing the burden of WML in the patients suffered from both WML and AIL.

Automatic Assessment of ASPECTS Using Diffusion-Weighted Imaging in Acute Ischemic Stroke Using Recurrent Residual Convolutional Neural Network

The early detection and rapid quantification of acute ischemic lesions play pivotal roles in stroke management. We developed a deep learning algorithm for the automatic binary classification of the Alberta Stroke Program Early Computed Tomographic Score (ASPECTS) using diffusion-weighted imaging (DWI) in acute stroke patients. Three hundred and ninety DWI datasets with acute anterior circulation stroke were included. A classifier algorithm utilizing a recurrent residual convolutional neural network (RRCNN) was developed for classification between low (1–6) and high (7–10) DWI-ASPECTS groups. The model performance was compared with a pre-trained VGG16, Inception V3, and a 3D convolutional neural network (3DCNN). The proposed RRCNN model demonstrated higher performance than the pre-trained models and 3DCNN with an accuracy of 87.3%, AUC of 0.941, and F1-score of 0.888 for classification between the low and high DWI-ASPECTS groups. These results suggest that the deep learning algorithm developed in this study can provide a rapid assessment of DWI-ASPECTS and may serve as an ancillary tool that can assist physicians in making urgent clinical decisions.

Automated segmentation of acute stroke lesions using a data-driven anomaly detection on diffusion weighted MRI

BACKGROUND

Successful delineation of lesions in acute ischemic strokes (AIS) is crucial for increasing the likelihood of good clinical outcome for the patient.

NEW METHODS

We developed a fully automated method to localize and segment AIS lesions in variable locations for 192 multimodal 3D-magnetic resonance images (MRI) including 106 stroke and 86 healthy cases. The method works based on the Crawford-Howell t-test and comparison of stroke images to healthy controls. We then developed a classifier to discriminate the images into stroke or non-stroke categories following the lesion segmentation.

RESULTS

The mean Dice similarity coefficient (DSC) for the test set was 0.50 ± 0.21 (min-max: 0.07-0.83) and mean net overlap was 0.66 ± 0.18 (min-max: 0.22-1). The experimental results for the classification of strokes from non-strokes showed mean accuracy, precision, sensitivity, and specificity of 73 %, 0.77 %, 84 %, and 69 %, respectively.

COMPARISON WITH EXISTING METHOD

The performance of our methods is comparable with previously published approaches based on machine learning and/or deep learning lesion segmentation techniques. However, most of the previously published methods have yielded low sensitivity, are computationally heavy, and difficult to interpret. The present approach is a significant improvement because it does not require high computation power and memory and can be implemented on a desktop workstation and integrated into the routine clinical diagnostic pipeline.

CONCLUSIONS

The current method is straightforward, fast, and shows good agreement with the lesions identified by human experts.

Big Data Approaches to Phenotyping Acute Ischemic Stroke Using Automated Lesion Segmentation of Multi-Center Magnetic Resonance Imaging Data

Background and Purpose- We evaluated deep learning algorithms' segmentation of acute ischemic lesions on heterogeneous multi-center clinical diffusion-weighted magnetic resonance imaging (MRI) data sets and explored the potential role of this tool for phenotyping acute ischemic stroke. Methods- Ischemic stroke data sets from the MRI-GENIE (MRI-Genetics Interface Exploration) repository consisting of 12 international genetic research centers were retrospectively analyzed using an automated deep learning segmentation algorithm consisting of an ensemble of 3-dimensional convolutional neural networks. Three ensembles were trained using data from the following: (1) 267 patients from an independent single-center cohort, (2) 267 patients from MRI-GENIE, and (3) mixture of (1) and (2). The algorithms' performances were compared against manual outlines from a separate 383 patient subset from MRI-GENIE. Univariable and multivariable logistic regression with respect to demographics, stroke subtypes, and vascular risk factors were performed to identify phenotypes associated with large acute diffusion-weighted MRI volumes and greater stroke severity in 2770 MRI-GENIE patients. Stroke topography was investigated. Results- The ensemble consisting of a mixture of MRI-GENIE and single-center convolutional neural networks performed best. Subset analysis comparing automated and manual lesion volumes in 383 patients found excellent correlation (ρ=0.92; P<0.0001). Median (interquartile range) diffusion-weighted MRI lesion volumes from 2770 patients were 3.7 cm3 (0.9-16.6 cm3). Patients with small artery occlusion stroke subtype had smaller lesion volumes ( P<0.0001) and different topography compared with other stroke subtypes. Conclusions- Automated accurate clinical diffusion-weighted MRI lesion segmentation using deep learning algorithms trained with multi-center and diverse data is feasible. Both lesion volume and topography can provide insight into stroke subtypes with sufficient sample size from big heterogeneous multi-center clinical imaging phenotype data sets.