Automated lesion segmentation with BIANCA: Impact of population-level features, classification algorithm and locally adaptive thresholding

Development and validation of a two-stage convolutional neural network algorithm for segmentation of MRI white matter hyperintensities for longitudinal studies in CADASIL

BACKGROUND

Segmentation of white matter hyperintensities (WMH) in CADASIL, one of the most severe cerebral small vessel disease of genetic origin, is challenging.

METHOD

We adapted and validated an automatic method based on a convolutional neural network (CNN) algorithm and using a large dataset of 2D and/or 3D FLAIR and T1-weighted images acquired in 132 patients, to measure the progression of WMH in this condition.

RESULTS

The volume of WMH measured using this method correlated strongly with reference data validated by experts. WMH segmentation was also clearly improved compared to the BIANCA segmentation method. Combining two successive learning models was found to be of particular interest, reducing the number of false-positive voxels and the extent of under-segmentation detected after a single-stage process. With the two-stage approach, WMH progression correlated with measures derived from the reference masks for lesions increasing with age, and with the variable WMH progression trajectories at individual level. We also confirmed the expected effect of the initial load of WMH and the influence of the type of MRI acquisition on measures of this progression.

CONCLUSION

Altogether, our findings suggest that WMH progression in CADASIL can be measured automatically with adequate confidence by a CNN segmentation algorithm.

Mapping the interplay of atrial fibrillation, brain structure, and cognitive dysfunction

INTRODUCTION

Atrial fibrillation (AF) is associated with an elevated risk of cognitive impairment and dementia. Understanding the cognitive sequelae and brain structural changes associated with AF is vital for addressing ensuing health care needs.

METHODS AND RESULTS

We examined 1335 stroke-free individuals with AF and 2683 matched controls using neuropsychological assessments and multimodal neuroimaging. The analysis revealed that individuals with AF exhibited deficits in executive function, processing speed, and reasoning, accompanied by reduced cortical thickness, elevated extracellular free-water content, and widespread white matter abnormalities, indicative of small vessel pathology. Notably, brain structural differences statistically mediated the relationship between AF and cognitive performance.

DISCUSSION

Integrating a comprehensive analysis approach with extensive clinical and magnetic resonance imaging data, our study highlights small vessel pathology as a possible unifying link among AF, cognitive decline, and abnormal brain structure. These insights can inform diagnostic approaches and motivate the ongoing implementation of effective therapeutic strategies. Highlights We investigated neuropsychological and multimodal neuroimaging data of 1335 individuals with atrial fibrillation (AF) and 2683 matched controls. Our analysis revealed AF-associated deficits in cognitive domains of attention, executive function, processing speed, and reasoning. Cognitive deficits in the AF group were accompanied by structural brain alterations including reduced cortical thickness and gray matter volume, alongside increased extracellular free-water content as well as widespread differences of white matter integrity. Structural brain changes statistically mediated the link between AF and cognitive performance, emphasizing the potential of structural imaging markers as a diagnostic tool in AF-related cognitive decline.

End-to-end volumetric segmentation of white matter hyperintensities using deep learning

BACKGROUND AND OBJECTIVES

Reliable detection of white matter hyperintensities (WMH) is crucial for studying the impact of diffuse white-matter pathology on brain health and monitoring changes in WMH load over time. However, manual annotation of 3D high-dimensional neuroimages is laborious and can be prone to biases and errors in the annotation procedure. In this study, we evaluate the performance of deep learning (DL) segmentation tools and propose a novel volumetric segmentation model incorporating self-attention via a transformer-based architecture. Ultimately, we aim to evaluate diverse factors that influence WMH segmentation, aiming for a comprehensive analysis of the state-of-the-art algorithms in a broader context.

METHODS

We trained state-of-the-art DL algorithms, and incorporated advanced attention mechanisms, using structural fluid-attenuated inversion recovery (FLAIR) image acquisitions. The anatomical MRI data utilized for model training was obtained from healthy individuals aged 62-70 years in the Live active Successful Aging (LISA) project. Given the potential sparsity of lesion volume among healthy aging individuals, we explored the impact of incorporating a weighted loss function and ensemble models. To assess the generalizability of the studied DL models, we applied the trained algorithm to an independent subset of data sourced from the MICCAI WMH challenge (MWSC). Notably, this subset had vastly different acquisition parameters compared to the LISA dataset used for training.

RESULTS

Consistently, DL approaches exhibited commendable segmentation performance, achieving the level of inter-rater agreement comparable to expert performance, ensuring superior quality segmentation outcomes. On the out of sample dataset, the ensemble models exhibited the most outstanding performance.

CONCLUSIONS

DL methods generally surpassed conventional approaches in our study. While all DL methods performed comparably, incorporating attention mechanisms could prove advantageous in future applications with a wider availability of training data. As expected, our experiments indicate that the use of ensemble-based models enables the superior generalization in out-of-distribution settings. We believe that introducing DL methods in the WHM annotation workflow in heathy aging cohorts is promising, not only for reducing the annotation time required, but also for eventually improving accuracy and robustness via incorporating the automatic segmentations in the evaluation procedure.

Identifying and reverting the adverse effects of white matter hyperintensities on cortical surface analyses

The Human Connectome Project (HCP)-style surface-based brain MRI analysis is a powerful technique that allows precise mapping of the cerebral cortex. However, the strength of its surface-based analysis has not yet been tested in the older population that often presents with white matter hyperintensities (WMHs) on T2-weighted (T2w) MRI (hypointensities on T1w MRI). We investigated T1-weighted (T1w) and T2w structural MRI in 43 healthy middle-aged to old participants. Juxtacortical WMHs were often misclassified by the default HCP pipeline as parts of the gray matter in T1w MRI, leading to incorrect estimation of the cortical surfaces and cortical metrics. To revert the adverse effects of juxtacortical WMHs, we incorporated the Brain Intensity AbNormality Classification Algorithm into the HCP pipeline (proposed pipeline). Blinded radiologists performed stereological quality control (QC) and found a decrease in the estimation errors in the proposed pipeline. The superior performance of the proposed pipeline was confirmed using an originally-developed automated surface QC based on a large database. Here we showed the detrimental effects of juxtacortical WMHs for estimating cortical surfaces and related metrics and proposed a possible solution for this problem. The present knowledge and methodology should help researchers identify adequate cortical surface biomarkers for aging and age-related neuropsychiatric disorders.

Association between periodontal disease and microstructural brain alterations in the Hamburg City Health Study

AIM

The aim of the PAROBRAIN study was to examine the association of periodontal health with microstructural white matter integrity and cerebral small vessel disease (CSVD) in the Hamburg City Health Study, a large population-based cohort with dental examination and brain magnetic resonance imaging (MRI).

MATERIALS AND METHODS

Periodontal health was determined by measuring clinical attachment loss (CAL) and plaque index. Additionally, the decayed/missing/filled teeth (DMFT) index was quantified. 3D-FLAIR and 3D-T1-weighted images were used for white matter hyperintensity (WMH) segmentation. Diffusion-weighted MRI was used to quantify peak width of skeletonized mean diffusivity (PSMD).

RESULTS

Data from 2030 participants were included in the analysis. Median age was 65 years, with 43% female participants. After adjusting for age and sex, an increase in WMH load was significantly associated with more CAL, higher plaque index and higher DMFT index. PSMD was significantly associated with the plaque index and DMFT. Additional adjustment for education and cardiovascular risk factors revealed a significant association of PSMD with plaque index (p < .001) and DMFT (p < .01), whereas effects of WMH load were attenuated (p > .05).

CONCLUSIONS

These findings suggest an adverse effect of periodontal health on CSVD and white matter integrity. Further research is necessary to examine whether early treatment of periodontal disease can prevent microstructural brain damage.

Brain imaging and neuropsychological assessment of individuals recovered from a mild to moderate SARS-CoV-2 infection

As severe acute respiratory syndrome coronavirus type 2 (SARS-CoV-2) infections have been shown to affect the central nervous system, the investigation of associated alterations of brain structure and neuropsychological sequelae is crucial to help address future health care needs. Therefore, we performed a comprehensive neuroimaging and neuropsychological assessment of 223 nonvaccinated individuals recovered from a mild to moderate SARS-CoV-2 infection (100 female/123 male, age [years], mean ± SD, 55.54 ± 7.07; median 9.7 mo after infection) in comparison with 223 matched controls (93 female/130 male, 55.74 ± 6.60) within the framework of the Hamburg City Health Study. Primary study outcomes were advanced diffusion MRI measures of white matter microstructure, cortical thickness, white matter hyperintensity load, and neuropsychological test scores. Among all 11 MRI markers tested, significant differences were found in global measures of mean diffusivity (MD) and extracellular free water which were elevated in the white matter of post-SARS-CoV-2 individuals compared to matched controls (free water: 0.148 ± 0.018 vs. 0.142 ± 0.017, P < 0.001; MD [10-3 mm2/s]: 0.747 ± 0.021 vs. 0.740 ± 0.020, P < 0.001). Group classification accuracy based on diffusion imaging markers was up to 80%. Neuropsychological test scores did not significantly differ between groups. Collectively, our findings suggest that subtle changes in white matter extracellular water content last beyond the acute infection with SARS-CoV-2. However, in our sample, a mild to moderate SARS-CoV-2 infection was not associated with neuropsychological deficits, significant changes in cortical structure, or vascular lesions several months after recovery. External validation of our findings and longitudinal follow-up investigations are needed.

Optimizing automated white matter hyperintensity segmentation in individuals with stroke

White matter hyperintensities (WMHs) are a risk factor for stroke. Consequently, many individuals who suffer a stroke have comorbid WMHs. The impact of WMHs on stroke recovery is an active area of research. Automated WMH segmentation methods are often employed as they require minimal user input and reduce risk of rater bias; however, these automated methods have not been specifically validated for use in individuals with stroke. Here, we present methodological validation of automated WMH segmentation methods in individuals with stroke. We first optimized parameters for FSL's publicly available WMH segmentation software BIANCA in two independent (multi-site) datasets. Our optimized BIANCA protocol achieved good performance within each independent dataset, when the BIANCA model was trained and tested in the same dataset or trained on mixed-sample data. BIANCA segmentation failed when generalizing a trained model to a new testing dataset. We therefore contrasted BIANCA's performance with SAMSEG, an unsupervised WMH segmentation tool available through FreeSurfer. SAMSEG does not require prior WMH masks for model training and was more robust to handling multi-site data. However, SAMSEG performance was slightly lower than BIANCA when data from a single site were tested. This manuscript will serve as a guide for the development and utilization of WMH analysis pipelines for individuals with stroke.

Association between Coffee Consumption and Brain MRI Parameters in the Hamburg City Health Study

Despite associations of regular coffee consumption with fewer neurodegenerative disorders, its association with microstructural brain alterations is unclear. To address this, we examined the association of coffee consumption with brain MRI parameters representing vascular brain damage, neurodegeneration, and microstructural integrity in 2316 participants in the population-based Hamburg City Health Study. Cortical thickness and white matter hyperintensity (WMH) load were measured on FLAIR and T1-weighted images. Microstructural white matter integrity was quantified as peak width of skeletonized mean diffusivity (PSMD) on diffusion-weighted MRI. Daily coffee consumption was assessed in five groups (<1 cup, 1-2 cups, 3-4 cups, 5-6 cups, >6 cups). In multiple linear regressions, we examined the association between brain MRI parameters and coffee consumption (reference group <1 cup). After adjustment for covariates, 3-4 cups of daily coffee were associated with lower PSMD (p = 0.028) and higher cortical thickness (p = 0.015) compared to <1 cup. Moreover, 1-2 cups per day was also associated with lower PSMD (p = 0.022). Associations with WMH load or other groups of coffee consumption were not significant (p > 0.05). The findings indicate that regular coffee consumption is positively associated with microstructural white matter integrity and cortical thickness. Further research is necessary to determine longitudinal effects of coffee on brain microstructure.

Association of Carotid Plaque and Flow Velocity With White Matter Integrity in a Middle-aged to Elderly Population

BACKGROUND AND OBJECTIVES

It is uncertain whether there is an association of carotid plaques (CPs) and flow velocities with peak width mean diffusivity (PSMD) and white matter hyperintensities (WMH) independent of shared risk factors. We aimed to study this association controlling for biomarkers of inflammation and cardiac dysfunction and typical cardiovascular risk factors and spatial distribution.

METHODS

We included participants from the population-based Hamburg City Health Study, recruiting citizens between 45 and 74 years of age. Medical history was obtained from structured interviews and extended laboratory tests, physical examinations, MRI of the head, echocardiography, and abdominal and carotid ultrasound were performed. We performed multivariable regression analysis with PSMD and periventricular, deep, and total volume of WMH (pWMH, dWMH, tWMH) as dependent variables. PSMD was calculated as the difference between the 95th and 5th percentiles of MD values on the white skeleton in standard space. Volumes of WMH were determined by the application of a manually trained k-nearest neighbor segmentation algorithm. WMH measured within a distance of 1 cm from the surface of the lateral ventricles were defined as pWMH and above 1 cm as dWMH.

RESULTS

Two thousand six hundred twenty-three participants were included. The median age was 65 years, and 56% were women. Their median tWMH was 946 mm3(IQR:419, 2,164), PSMD 2.24 mm2/s × 10-4 (IQR: 2.04, 2.47), peak systolic velocity (PSV) of internal carotid arteries 0.70m/second (IQR:0.60, 0.81), and 35% had CPs. Adjusted for age, sex, high-sensitive CRP, NT-proBNP, and commonly measured cardiovascular risk and systemic hemodynamic factors, both CPs (B = 0.15; CI: 0.04, 0.26; p = 0.006) and low PSV (B = -0.49; CI: -0.87, -0.11; p = 0.012) were significantly associated with a higher tWMH and PSMD. Low PSV (B = -0.48; CI: -0.87, -0.1; p = 0.013) was associated with pWMH and the presence of CP with pWMH (B = 0.15; CI: 0.04, 0.26; p = 0.008) and dWMH (B = 0.42; CI: 0.11, 0.74; p < 0.009).

DISCUSSION

Low PSV and CP are associated with WMH and PSMD independent of cardiovascular risk factors and biomarkers of inflammation and cardiac dysfunction. This points toward pathophysiologic pathways underlying both large and small vessel disease beyond the common cardiovascular risk profile.

TRIAL REGISTRATION INFORMATION

The trial was submitted at clinicaltrials.gov, under NCT03934957 on January 4, 2019. The first participant was enrolled in February 2016.

Equalization of Brain State Occupancy Accompanies Cognitive Impairment in Cerebral Small Vessel Disease

BACKGROUND

Cognitive impairment is a hallmark of cerebral small vessel disease (cSVD). Functional magnetic resonance imaging has highlighted connections between patterns of brain activity and variability in behavior. We aimed to characterize the associations between imaging markers of cSVD, dynamic connectivity, and cognitive impairment.

METHODS

We obtained magnetic resonance imaging and clinical data from the population-based Hamburg City Health Study. cSVD was quantified by white matter hyperintensities and peak-width of skeletonized mean diffusivity (PSMD). Resting-state blood oxygen level-dependent signals were clustered into discrete brain states, for which fractional occupancies (%) and dwell times (seconds) were computed. Cognition in multiple domains was assessed using validated tests. Regression analysis was used to quantify associations between white matter damage, spatial coactivation patterns, and cognitive function.

RESULTS

Data were available for 979 participants (ages 45-74 years, median white matter hyperintensity volume 0.96 mL). Clustering identified five brain states with the most time spent in states characterized by activation (+) or suppression (-) of the default mode network (DMN) (fractional occupancy: DMN+ = 25.1 ± 7.2%, DMN- = 25.5 ± 7.2%). Every 4.7-fold increase in white matter hyperintensity volume was associated with a 0.95-times reduction of the odds of occupying DMN+ or DMN-. Time spent in DMN-related brain states was associated with executive function.

CONCLUSIONS

Associations between white matter damage, whole-brain spatial coactivation patterns, and cognition suggest equalization of time spent in different brain states as a marker for cSVD-associated cognitive decline. Reduced gradients between brain states in association with brain damage and cognitive impairment reflect the dedifferentiation hypothesis of neurocognitive aging in a network-theoretical context.

Automatic segmentation of white matter hyperintensities in routine clinical brain MRI by 2D VB-Net: A large-scale study

White matter hyperintensities (WMH) are imaging manifestations frequently observed in various neurological disorders, yet the clinical application of WMH quantification is limited. In this study, we designed a series of dedicated WMH labeling protocols and proposed a convolutional neural network named 2D VB-Net for the segmentation of WMH and other coexisting intracranial lesions based on a large dataset of 1,045 subjects across various demographics and multiple scanners using 2D thick-slice protocols that are more commonly applied in clinical practice. Using our labeling pipeline, the Dice consistency of the WMH regions manually depicted by two observers was 0.878, which formed a solid basis for the development and evaluation of the automatic segmentation system. The proposed algorithm outperformed other state-of-the-art methods (uResNet, 3D V-Net and Visual Geometry Group network) in the segmentation of WMH and other coexisting intracranial lesions and was well validated on datasets with thick-slice magnetic resonance (MR) images and the 2017 medical image computing and computer assisted intervention WMH Segmentation Challenge dataset (with thin-slice MR images), all showing excellent effectiveness. Furthermore, our method can subclassify WMH to display the WMH distributions and is very lightweight. Additionally, in terms of correlation to visual rating scores, our algorithm showed excellent consistency with the manual delineations and was overall better than those from other competing methods. In conclusion, we developed an automatic WMH quantification framework for multiple application scenarios, exhibiting a promising future in clinical practice.

Higher-resolution quantification of white matter hypointensities by large-scale transfer learning from 2D images on the JPSC-AD cohort

White matter lesions (WML) commonly occur in older brains and are quantifiable on MRI, often used as a biomarker in Aging research. Although algorithms are regularly proposed that identify these lesions from T2‐fluid‐attenuated inversion recovery (FLAIR) sequences, none so far can estimate lesions directly from T1‐weighted images with acceptable accuracy. Since 3D T1 is a polyvalent and higher‐resolution sequence, it could be beneficial to obtain the distribution of WML directly from it. However a serious difficulty, both for algorithms and human, can be found in the ambiguities of brain signal intensity in T1 images. This manuscript shows that a cross‐domain ConvNet (Convolutional Neural Network) approach can help solve this problem. Still, this is non‐trivial, as it would appear to require a large and varied dataset (for robustness) labelled at the same high resolution (for spatial accuracy). Instead, our model was taught from two‐dimensional FLAIR images with a loss function designed to handle the super‐resolution need. And crucially, we leveraged a very large training set for this task, the recently assembled, multi‐sites Japan Prospective Studies Collaboration for Aging and Dementia (JPSC‐AD) cohort. We describe the two‐step procedure that we followed to handle such a large number of imperfectly labeled samples. A large‐scale accuracy evaluation conducted against FreeSurfer 7, and a further visual expert rating revealed that WML segmentation from our ConvNet was consistently better. Finally, we made a directly usable software program based on that trained ConvNet model, available at https://github.com/bthyreau/deep-T1-WMH.

Deep-Learning-Based Automatic Mineral Grain Segmentation and Recognition

A multitude of applications in engineering, ore processing, mineral exploration, and environmental science require grain recognition and the counting of minerals. Typically, this task is performed manually with the drawback of monopolizing both time and resources. Moreover, it requires highly trained personnel with a wealth of knowledge and equipment, such as scanning electron microscopes and optical microscopes. Advances in machine learning and deep learning make it possible to envision the automation of many complex tasks in various fields of science at an accuracy equal to human performance, thereby, avoiding placing human resources into tedious and repetitive tasks, improving time efficiency, and lowering costs. Here, we develop deep-learning algorithms to automate the recognition of minerals directly from the grains captured from optical microscopes. Building upon our previous work and applying state-of-the-art technology, we modify a superpixel segmentation method to prepare data for the deep-learning algorithms. We compare two residual network architectures (ResNet 1 and ResNet 2) for the classification and identification processes. We achieve a validation accuracy of 90.5% using the ResNet 2 architecture with 47 layers. Our approach produces an effective application of deep learning to automate mineral recognition and counting from grains while also achieving a better recognition rate than reported thus far in the literature for this process and other well-known, deep-learning-based models, including AlexNet, GoogleNet, and LeNet.

Performance of three freely available methods for extracting white matter hyperintensities: FreeSurfer, UBO Detector, and BIANCA

White matter hyperintensities (WMH) of presumed vascular origin are frequently found in MRIs of healthy older adults. WMH are also associated with aging and cognitive decline. Here, we compared and validated three algorithms for WMH extraction: FreeSurfer (T1w), UBO Detector (T1w + FLAIR), and FSL's Brain Intensity AbNormality Classification Algorithm (BIANCA; T1w + FLAIR) using a longitudinal dataset comprising MRI data of cognitively healthy older adults (baseline N = 231, age range 64-87 years). As reference we manually segmented WMH in T1w, three-dimensional (3D) FLAIR, and two-dimensional (2D) FLAIR images which were used to assess the segmentation accuracy of the different automated algorithms. Further, we assessed the relationships of WMH volumes provided by the algorithms with Fazekas scores and age. FreeSurfer underestimated the WMH volumes and scored worst in Dice Similarity Coefficient (DSC = 0.434) but its WMH volumes strongly correlated with the Fazekas scores (rs  = 0.73). BIANCA accomplished the highest DSC (0.602) in 3D FLAIR images. However, the relations with the Fazekas scores were only moderate, especially in the 2D FLAIR images (rs  = 0.41), and many outlier WMH volumes were detected when exploring within-person trajectories (2D FLAIR: ~30%). UBO Detector performed similarly to BIANCA in DSC with both modalities and reached the best DSC in 2D FLAIR (0.531) without requiring a tailored training dataset. In addition, it achieved very high associations with the Fazekas scores (2D FLAIR: rs  = 0.80). In summary, our results emphasize the importance of carefully contemplating the choice of the WMH segmentation algorithm and MR-modality.

Automated detection and segmentation of sclerotic spinal lesions on body CTs using a deep convolutional neural network

PURPOSE

To develop a deep convolutional neural network capable of detecting spinal sclerotic metastases on body CTs.

MATERIALS AND METHODS

Our study was IRB-approved and HIPAA-compliant. Cases of confirmed sclerotic bone metastases in chest, abdomen, and pelvis CTs were identified. Images were manually segmented for 3 classes: background, normal bone, and sclerotic lesion(s). If multiple lesions were present on a slice, all lesions were segmented. A total of 600 images were obtained, with a 90/10 training/testing split. Images were stored as 128 × 128 pixel grayscale and the training dataset underwent a processing pipeline of histogram equalization and data augmentation. We trained our model from scratch on Keras/TensorFlow using an 80/20 training/validation split and a U-Net architecture (64 batch size, 100 epochs, dropout 0.25, initial learning rate 0.0001, sigmoid activation). We also tested our model's true negative and false positive rate with 1104 non-pathologic images. Global sensitivity measured model detection of any lesion on a single image, local sensitivity and positive predictive value (PPV) measured model detection of each lesion on a given image, and local specificity measured the false positive rate in non-pathologic bone.

RESULTS

Dice scores were 0.83 for lesion, 0.96 for non-pathologic bone, and 0.99 for background. Global sensitivity was 95% (57/60), local sensitivity was 92% (89/97), local PPV was 97% (89/92), and local specificity was 87% (958/1104).

CONCLUSION

A deep convolutional neural network has the potential to assist in detecting sclerotic spinal metastases.

Deep Bayesian networks for uncertainty estimation and adversarial resistance of white matter hyperintensity segmentation

White matter hyperintensities (WMHs) are frequently observed on structural neuroimaging of elderly populations and are associated with cognitive decline and increased risk of dementia. Many existing WMH segmentation algorithms produce suboptimal results in populations with vascular lesions or brain atrophy, or require parameter tuning and are computationally expensive. Additionally, most algorithms do not generate a confidence estimate of segmentation quality, limiting their interpretation. MRI‐based segmentation methods are often sensitive to acquisition protocols, scanners, noise‐level, and image contrast, failing to generalize to other populations and out‐of‐distribution datasets. Given these concerns, we propose a novel Bayesian 3D convolutional neural network with a U‐Net architecture that automatically segments WMH, provides uncertainty estimates of the segmentation output for quality control, and is robust to changes in acquisition protocols. We also provide a second model to differentiate deep and periventricular WMH. Four hundred thirty‐two subjects were recruited to train the CNNs from four multisite imaging studies. A separate test set of 158 subjects was used for evaluation, including an unseen multisite study. We compared our model to two established state‐of‐the‐art techniques (BIANCA and DeepMedic), highlighting its accuracy and efficiency. Our Bayesian 3D U‐Net achieved the highest Dice similarity coefficient of 0.89 ± 0.08 and the lowest modified Hausdorff distance of 2.98 ± 4.40 mm. We further validated our models highlighting their robustness on “clinical adversarial cases” simulating data with low signal‐to‐noise ratio, low resolution, and different contrast (stemming from MRI sequences with different parameters). Our pipeline and models are available at: https://hypermapp3r.readthedocs.io.

The Effect of Training Sample Size on the Prediction of White Matter Hyperintensity Volume in a Healthy Population Using BIANCA

Introduction: White matter hyperintensities of presumed vascular origin (WMH) are an important magnetic resonance imaging marker of cerebral small vessel disease and are associated with cognitive decline, stroke, and mortality. Their relevance in healthy individuals, however, is less clear. This is partly due to the methodological challenge of accurately measuring rare and small WMH with automated segmentation programs. In this study, we tested whether WMH volumetry with FMRIB software library v6.0 (FSL; https://fsl.fmrib.ox.ac.uk/fsl/fslwiki) Brain Intensity AbNormality Classification Algorithm (BIANCA), a customizable and trainable algorithm that quantifies WMH volume based on individual data training sets, can be optimized for a normal aging population.

Methods: We evaluated the effect of varying training sample sizes on the accuracy and the robustness of the predicted white matter hyperintensity volume in a population (n = 201) with a low prevalence of confluent WMH and a substantial proportion of participants without WMH. BIANCA was trained with seven different sample sizes between 10 and 40 with increments of 5. For each sample size, 100 random samples of T1w and FLAIR images were drawn and trained with manually delineated masks. For validation, we defined an internal and external validation set and compared the mean absolute error, resulting from the difference between manually delineated and predicted WMH volumes for each set. For spatial overlap, we calculated the Dice similarity index (SI) for the external validation cohort.

Results: The study population had a median WMH volume of 0.34 ml (IQR of 1.6 ml) and included n = 28 (18%) participants without any WMH. The mean absolute error of the difference between BIANCA prediction and manually delineated masks was minimized and became more robust with an increasing number of training participants. The lowest mean absolute error of 0.05 ml (SD of 0.24 ml) was identified in the external validation set with a training sample size of 35. Compared to the volumetric overlap, the spatial overlap was poor with an average Dice similarity index of 0.14 (SD 0.16) in the external cohort, driven by subjects with very low lesion volumes.

Discussion: We found that the performance of BIANCA, particularly the robustness of predictions, could be optimized for use in populations with a low WMH load by enlargement of the training sample size. Further work is needed to evaluate and potentially improve the prediction accuracy for low lesion volumes. These findings are important for current and future population-based studies with the majority of participants being normal aging people.

Exercise Training and Cognitive Function in Kidney Disease: Protocol for a Pilot Randomized Controlled Trial

BACKGROUND

Chronic kidney disease (CKD) is extremely common in older adults and is associated with cognitive impairment. It is hypothesized that accelerated cognitive decline in CKD results from a vascular dysfunction-induced reduction in the integrity of the brain white matter.

OBJECTIVE

The aim of this study was to describe the protocol for a study to evaluate whether exercise training provides a cerebroprotective effect by improving cerebrovascular health.

METHODS

This is a randomized controlled trial investigating feasibility and effect size.

RESULTS

Participants will be randomized to either a 24-week, home-based, walking program or a usual care group. Participants will undergo evaluation of cognitive function, brain structure via magnetic reasoning imaging, physical function, physical activity, and vascular function. The primary outcome is change in cognitive function.

DISCUSSION

The findings of this study will help determine whether exercise training influences cognitive function during a therapeutic window in the disease process of cognitive impairment in older adults with CKD.

CONCLUSION

This protocol describes a study to evaluate cognition and brain structure following a home-based exercise program to an at-risk population.

Brain tissue development of neonates with Congenital Septal Defect: Study on MRI Image Evaluation of Deep Learning Algorithm

Objectives:

This article is based on deep learning algorithms and uses MRI to study the development of congenital heart septal defects in neonatal brain tissue.

Methods:

From January 2018 to December 2019, 150 cases of congenital cardiac paper septal defect were retrospectively analyzed on 50 cases of normal newborns and neonates. The four index parametersbrain MR imaging, lateral ventricle pre-angle measurement index (F/F’), body index (D/ D’), caudal nucleus index (C/C’) were analyzed. The independent sample t test is performed to compare the difference parameters between groups.

Results:

F congenital heart disease group and control group/F ‘values were 0.301 ± 0.035 and 0.296 ± 0.031; Evans index was 0.239 ± 0.052 and 0.233 ± 0.025; 2 sets of D/D’ values were 0.261 ± 0.039 and 0.234 ± 0.032; C/C ‘value was 0.138 ± 0.018 and 0.124 ± 0.015 respectively. The congenital heart disease group D/D ‘, and the value of C/C’ obtained under the ROC curve area value, respectively 0.698 and 0.750, Youden index corresponding to the maximum D/D ‘, and the value of C/C’ values were 0.28 and 0.12.

Conclusion:

Lateral ventricle D/D ‘and C/C’ is more sensitive indicator which can be evaluated with the difference between the volume of congenital heart septal defects in newborn normal neonatal brain; when the D/D ‘value> 0.28, C/C’ value> 0.12. For the diagnosis and evaluation of congenital heart septal defect neonatal brain volume abnormalities have a certain reference value.

Apathy in small vessel cerebrovascular disease is associated with deficits in effort-based decision making

Patients with small vessel cerebrovascular disease frequently suffer from apathy, a debilitating neuropsychiatric syndrome, the underlying mechanisms of which remain to be established. Here we investigated the hypothesis that apathy is associated with disrupted decision making in effort-based decision making, and that these alterations are associated with abnormalities in the white matter network connecting brain regions that underpin such decisions. Eighty-two patients with MRI evidence of small vessel disease were assessed using a behavioural paradigm as well as diffusion weighted MRI. The decision-making task involved accepting or rejecting monetary rewards in return for performing different levels of physical effort (hand grip force). Choice data and reaction times were integrated into a drift diffusion model that framed decisions to accept or reject offers as stochastic processes approaching a decision boundary with a particular drift rate. Tract-based spatial statistics were used to assess the relationship between white matter tract integrity and apathy, while accounting for depression. Overall, patients with apathy accepted significantly fewer offers on this decision-making task. Notably, while apathetic patients were less responsive to low rewards, they were also significantly averse to investing in high effort. Significant reductions in white matter integrity were observed to be specifically related to apathy, but not to depression. These included pathways connecting brain regions previously implicated in effort-based decision making in healthy people. The drift rate to decision parameter was significantly associated with both apathy and altered white matter tracts, suggesting that both brain and behavioural changes in apathy are associated with this single parameter. On the other hand, depression was associated with an increase in the decision boundary, consistent with an increase in the amount of evidence required prior to making a decision. These findings demonstrate altered effort-based decision making for reward in apathy, and also highlight dissociable mechanisms underlying apathy and depression in small vessel disease. They provide clear potential brain and behavioural targets for future therapeutic interventions, as well as modelling parameters that can be used to measure the effects of treatment at the behavioural level.

External validation of automated focal cortical dysplasia detection using morphometric analysis

OBJECTIVE

Focal cortical dysplasias (FCDs) are a common cause of drug-resistant focal epilepsy but frequently remain undetected by conventional magnetic resonance imaging (MRI) assessment. The visual detection can be facilitated by morphometric analysis of T1-weighted images, for example, using the Morphometric Analysis Program (v2018; MAP18), which was introduced in 2005, independently validated for its clinical benefits, and successfully integrated in standard presurgical workflows of numerous epilepsy centers worldwide. Here we aimed to develop an artificial neural network (ANN) classifier for robust automated detection of FCDs based on these morphometric maps and probe its generalization performance in a large, independent data set.

METHODS

In this retrospective study, we created a feed-forward ANN for FCD detection based on the morphometric output maps of MAP18. The ANN was trained and cross-validated on 113 patients (62 female, mean age ± SD =29.5 ± 13.6 years) with manually segmented FCDs and 362 healthy controls (161 female, mean age ± SD =30.2 ± 9.6 years) acquired on 13 different scanners. In addition, we validated the performance of the trained ANN on an independent, unseen data set of 60 FCD patients (28 female, mean age ± SD =30 ± 15.26 years) and 70 healthy controls (42 females, mean age ± SD = 40.0 ± 12.54 years).

RESULTS

In the cross-validation, the ANN achieved a sensitivity of 87.4% at a specificity of 85.4% on the training data set. On the independent validation data set, our method still reached a sensitivity of 81.0% at a comparably high specificity of 84.3%.

SIGNIFICANCE

Our method shows a robust automated detection of FCDs and performance generalizability, largely independent of scanning site or MR-sequence parameters. Taken together with the minimal input requirements of a standard T1 image, our approach constitutes a clinically viable and useful tool in the presurgical diagnostic routine for drug-resistant focal epilepsy.

Automatic segmentation of white matter hyperintensities from brain magnetic resonance images in the era of deep learning and big data - A systematic review

BACKGROUND

White matter hyperintensities (WMH), of presumed vascular origin, are visible and quantifiable neuroradiological markers of brain parenchymal change. These changes may range from damage secondary to inflammation and other neurological conditions, through to healthy ageing. Fully automatic WMH quantification methods are promising, but still, traditional semi-automatic methods seem to be preferred in clinical research. We systematically reviewed the literature for fully automatic methods developed in the last five years, to assess what are considered state-of-the-art techniques, as well as trends in the analysis of WMH of presumed vascular origin.

METHOD

We registered the systematic review protocol with the International Prospective Register of Systematic Reviews (PROSPERO), registration number - CRD42019132200. We conducted the search for fully automatic methods developed from 2015 to July 2020 on Medline, Science direct, IEE Explore, and Web of Science. We assessed risk of bias and applicability of the studies using QUADAS 2.

RESULTS

The search yielded 2327 papers after removing 104 duplicates. After screening titles, abstracts and full text, 37 were selected for detailed analysis. Of these, 16 proposed a supervised segmentation method, 10 proposed an unsupervised segmentation method, and 11 proposed a deep learning segmentation method. Average DSC values ranged from 0.538 to 0.91, being the highest value obtained from an unsupervised segmentation method. Only four studies validated their method in longitudinal samples, and eight performed an additional validation using clinical parameters. Only 8/37 studies made available their methods in public repositories.

CONCLUSIONS

We found no evidence that favours deep learning methods over the more established k-NN, linear regression and unsupervised methods in this task. Data and code availability, bias in study design and ground truth generation influence the wider validation and applicability of these methods in clinical research.

Artificial intelligence for brain diseases: A systematic review

Artificial intelligence (AI) is a major branch of computer science that is fruitfully used for analyzing complex medical data and extracting meaningful relationships in datasets, for several clinical aims. Specifically, in the brain care domain, several innovative approaches have achieved remarkable results and open new perspectives in terms of diagnosis, planning, and outcome prediction. In this work, we present an overview of different artificial intelligent techniques used in the brain care domain, along with a review of important clinical applications. A systematic and careful literature search in major databases such as Pubmed, Scopus, and Web of Science was carried out using "artificial intelligence" and "brain" as main keywords. Further references were integrated by cross-referencing from key articles. 155 studies out of 2696 were identified, which actually made use of AI algorithms for different purposes (diagnosis, surgical treatment, intra-operative assistance, and postoperative assessment). Artificial neural networks have risen to prominent positions among the most widely used analytical tools. Classic machine learning approaches such as support vector machine and random forest are still widely used. Task-specific algorithms are designed for solving specific problems. Brain images are one of the most used data types. AI has the possibility to improve clinicians' decision-making ability in neuroscience applications. However, major issues still need to be addressed for a better practical use of AI in the brain. To this aim, it is important to both gather comprehensive data and build explainable AI algorithms.

Three-tissue compositional analysis reveals in-vivo microstructural heterogeneity of white matter hyperintensities following stroke

White matter hyperintensities (WMHs) are frequently observed on brain scans of older individuals and are associated with cognitive impairment and vascular brain burden. Recent studies have shown that WMHs may only represent an extreme end of a diffuse pathological spectrum of white matter (WM) degeneration. The present study investigated the microstructural characteristics of WMHs using an advanced diffusion MRI modelling approach known as Single-Shell 3-Tissue Constrained Spherical Deconvolution (SS3T-CSD), which provides information on different tissue compartments within each voxel. The SS3T-CSD method may provide complementary information in the interpretation of pathological tissue through the tissue-specific microstructural compositions of WMHs. Data were obtained from stroke patients enrolled in the Cognition and Neocortical Volume After Stroke (CANVAS) study, a study examining brain volume and cognition after stroke. WMHs were segmented using an automated method, based on fluid attenuated inversion recovery (FLAIR) images. Automated tissue segmentation was used to identify normal-appearing white matter (NAWM). WMHs were classified into juxtaventricular, periventricular and deep lesions, based on their distance from the ventricles (3-10 ​mm). We aimed to compare in stroke participants the microstructural composition of the different lesion classes of WMHs and compositions of NAWM to assess the in-vivo heterogeneity of these lesions. Results showed that the 3-tissue composition significantly differed between WMHs classes and NAWM. Specifically, the 3-tissue compositions for juxtaventricular and periventricular WMHs both exhibited a relatively greater fluid-like (free water) content, which is compatible with a presence of interstitial fluid accumulation, when compared to deep WMHs. These findings provide evidence of microstructural heterogeneity of WMHs in-vivo and may support new insights for understanding the role of WMH development in vascular neurodegeneration.

Prediction of cognitive performance in old age from spatial probability maps of white matter lesions

The purposes of this study were to explore the association between cognitive performance and white matter lesions (WMLs), and to investigate whether it is possible to predict cognitive impairment using spatial maps of WMLs. These WML maps were produced for 263 elders from the OASIS-3 dataset, and a relevance vector regression (RVR) model was applied to predict neuropsychological performance based on the maps. The association between the spatial distribution of WMLs and cognitive function was examined using diffusion tensor imaging data. WML burden significantly associated with increasing age (r=0.318, p<0.001) and cognitive decline. Eight of 15 neuropsychological measures could be accurately predicted, and the mini-mental state examination (MMSE) test achieved the highest predictive accuracy (CORR=0.28, p<0.003). WMLs located in bilateral tapetum, posterior corona radiata, and thalamic radiation contributed the most prediction power. Diffusion indexes in these regions associated significantly with cognitive performance (axial diffusivity>radial diffusivity>mean diffusivity>fractional anisotropy). These results show that the combination of the extent and location of WMLs exhibit great potential to serve as a generalizable marker of multidomain neurocognitive decline in the aging population. The results may also shed light on the mechanism underlying white matter changes during the progression of cognitive decline and aging.