Novel imaging markers for altered cerebrovascular morphology in aging, stroke, and Alzheimer's disease

Deshpande, Aditi; Elliott, Jordan; Kari, Nitya; Jiang, Bin; Michel, Patrik; Toosizadeh, Nima; Fahadan, Pouya Tahsili; Kidwell, Chelsea S.; Wintermark, Max; Laksari, Kaveh

doi:10.1111/jon.13023

Cited by 7 publications

(17 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Of 100 AIS patients, the occlusion sites were the Internal Carotid Artery (ICA) in 18, tandem ICA-MCA in 17, and M1 and M2 segments of the MCA in 46 and 19, respectively. We utilized our previously developed probabilistic cerebrovascular atlas of spatially co-registered cerebral vessel maps of 175 healthy adults (34). The atlas was labeled to denote the five major vascular territories: (1) the ICA, (2 and 3) the left and right MCA, (4) Anterior Cerebral Artery (ACA), and (5) the posterior cerebral artery (PCA) and basilar artery (BA), as illustrated in Fig.…”

Section: Methodsmentioning

confidence: 99%

“…We implemented a two-step approach for automatic stroke detection based on previous preliminary studies that showed a significant difference in cerebrovascular features between stroke patients and healthy subjects (32,34). In the first step, the presence of occlusion was determined by comparing the vessel density of the patient-specific vascular network with the cerebrovascular atlas using the total vessel length, volume, and the number of branches.…”

Section: Stroke Detection and Occlusion Localizationmentioning

confidence: 99%

“…Cerebrovascular geometric features have been shown to correlate with aging and pathologic states (32,34,(45)(46)(47). To utilize the predictive ability of the rich imaging-based vascular information, we incorporated novel patient-specific cerebrovascular geometric features extracted by the segmentation algorithm described above (32) as well as automatically estimated CI into our ML model as additional predictors of the 90-day mRS. Table 2 in the Supplementary Material lists all the features used as predictors to train the model.…”

Section: Functional Outcome Predictionmentioning

confidence: 99%

“…Vascular morphology differs significantly between healthy and stroke subjects (32,34). Implementing complex cerebrovascular features and quantitative measurement of deviation from the average healthy atlas forms the basis of our stroke detection algorithm.…”

Section: Stroke Detection and Occlusion Localizationmentioning

confidence: 99%

“…Using our validated segmentation algorithm as ground truth (32), we train a convolutional neural network (CNN) for instantaneous, accurate, and automatic segmentation of the vascular network from the raw CTA or MRA scans that aids in subsequent extraction of the complex geometric features of the cerebral vessel network, namely – total length, average diameter, branching pattern, total volume, vessel tortuosity, and fractal dimension. We then used these features in conjunction with our previously developed statistical cerebrovascular atlas (34) to automatically detect the presence and site of the vessel occlusion, and calculate the CI for each stroke patient. The findings were validated against ground truth clinical assessment and grading of each subject’s CI.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

End to end stroke triage using cerebrovascular morphology and machine learning

Deshpande

Elliott

Jiang

et al. 2023

Preprint

View full text Add to dashboard Cite

BackgroundRapid and accurate triage of acute ischemic stroke (AIS) is essential for early revascularization and improved patient outcomes. Response to acute reperfusion therapies varies significantly based on patient-specific cerebrovascular anatomy that governs cerebral blood flow. We present an end-to-end machine learning approach for automatic stroke triage.MethodsEmploying a validated convolutional neural network (CNN) segmentation model for image processing, we extract each patient’s cerebrovasculature and its morphological features from baseline non-invasive angiography scans. These features are used to detect occlusion’s presence and the site automatically, and for the first time, to estimate collateral circulation without manual intervention. We then use the extracted cerebrovascular features along with commonly used clinical and imaging parameters to predict the 90-day functional outcome for each patient.ResultsThe CNN model achieved a segmentation accuracy of 94%. The automatic stroke detection algorithm had a sensitivity and specificity of 92% and 94%, respectively. The models for occlusion site detection and automatic collateral grading reached 96% and 87.2% accuracy, respectively. Incorporating the automatically extracted cerebrovascular features significantly improved the 90-day outcome prediction accuracy from 0.63 to 0.83.ConclusionsThe fast, automatic, and comprehensive model presented here can improve stroke diagnosis, aid collateral assessment, and enhance prognostication for treatment decisions, using cerebrovascular morphology.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Stroke Detection and Occlusion Localizationmentioning

confidence: 99%

Section: Functional Outcome Predictionmentioning

confidence: 99%

Section: Stroke Detection and Occlusion Localizationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

End to end stroke triage using cerebrovascular morphology and machine learning

Deshpande

Elliott

Jiang

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

Aging modulates homeostatic leukocyte trafficking to the peritoneal cavity in a sex-specific manner

Hopkin

Pezhman

Begum

et al. 2023

Journal of Leukocyte Biology

View full text Add to dashboard Cite

Aging is associated with exacerbated systemic inflammation (inflammaging) and the progressive loss of immune system function (immunosenescence). Leukocyte migration is necessary for effective immunity; however, dysregulated trafficking of leukocytes into tissue contributes to inflammaging and the development of age-related inflammatory diseases. Aging modulates leukocyte trafficking under inflammatory conditions; however, whether aging modulates leukocyte trafficking under homeostatic conditions remains to be elucidated. Although immune responses are evidently sexually dimorphic, limited studies have investigated the effect of sex on age-related changes to leukocyte trafficking processes. Here, we investigated age-related and sex-specific changes to the leukocyte populations within the peritoneal cavity of young (3-mo), middle-aged (18-mo) and old (21-mo) male and female wild-type mice in the steady state. We found an age-related increase in the number of leukocytes within the peritoneal cavity of female mice, predominantly B cells, which may reflect increased trafficking through this tissue with age. This was accompanied by an increased inflammatory environment within the aged cavity, including increased levels of chemoattractants, including B cell chemoattractants CXCL13 and CCL21, soluble adhesion molecules, and proinflammatory cytokines, which was more pronounced in aged female mice. Intravital microscopy techniques revealed altered vascular structure and increased vascular permeability within the peritoneal membrane of aged female mice, which may support increased leukocyte trafficking to the cavity with age. Together, these data indicate that aging affects homeostatic leukocyte trafficking processes in a sex-specific fashion.

show abstract

Reconstructing blood flow in data-poor regimes: a vasculature network kernel for Gaussian process regression

Ashtiani,

Sarabian,

Laksari

et al. 2024

J. R. Soc. Interface.

View full text Add to dashboard Cite

Blood flow reconstruction in the vasculature is important for many clinical applications. However, in clinical settings, the available data are often quite limited. For instance, transcranial Doppler ultrasound is a non-invasive clinical tool that is commonly used in clinical settings to measure blood velocity waveforms at several locations. This amount of data is grossly insufficient for training machine learning surrogate models, such as deep neural networks or Gaussian process regression. In this work, we propose a Gaussian process regression approach based on empirical kernels constructed by data generated from physics-based simulations—enabling near-real-time reconstruction of blood flow in data-poor regimes. We introduce a novel methodology to reconstruct the kernel within the vascular network. The proposed kernel encodes both spatiotemporal and vessel-to-vessel correlations, thus enabling blood flow reconstruction in vessels that lack direct measurements. We demonstrate that any prediction made with the proposed kernel satisfies the conservation of mass principle. The kernel is constructed by running stochastic one-dimensional blood flow simulations, where the stochasticity captures the epistemic uncertainties, such as lack of knowledge about boundary conditions and uncertainties in vasculature geometries. We demonstrate the performance of the model on three test cases, namely, a simple Y-shaped bifurcation, abdominal aorta and the circle of Willis in the brain.

show abstract

Novel imaging markers for altered cerebrovascular morphology in aging, stroke, and Alzheimer's disease

Cited by 7 publications

References 48 publications

End to end stroke triage using cerebrovascular morphology and machine learning

End to end stroke triage using cerebrovascular morphology and machine learning

Aging modulates homeostatic leukocyte trafficking to the peritoneal cavity in a sex-specific manner

Reconstructing blood flow in data-poor regimes: a vasculature network kernel for Gaussian process regression

Contact Info

Product

Resources

About