Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering

Abstract

Cerebral Blood Flow Velocity waveforms acquired via Transcranial Doppler (TCD) can provide evidence for cerebrovascular occlusion and stenosis. Thrombolysis in Brain Ischemia (TIBI) flow grades are widely used for this purpose, but require subjective assessment by expert evaluators to be reliable. In this work we seek to determine whether TCD morphology can be objectively assessed using an unsupervised machine learning approach to waveform categorization. TCD beat waveforms were recorded at multiple depths from the Middle Cerebral Arteries of 106 subjects; 33 with Large Vessel Occlusion (LVO). From each waveform, three morphological features were extracted, quantifying onset of maximal velocity, systolic canopy length, and the number/prominence of peaks/troughs. Spectral clustering identified groups implicit in the resultant three-dimensional feature space, with gap statistic criteria establishing the optimal cluster number. We found that gap statistic disparity was maximized at four clusters, referred to as flow types I, II, III, and IV. Types I and II were primarily composed of control subject waveforms, whereas types III and IV derived mainly from LVO patients. Cluster morphologies for types I and IV aligned clearly with Normal and Blunted TIBI flows, respectively. Types II and III represented commonly observed flow-types not delineated by TIBI, which nonetheless deviate from normal and blunted flows. We conclude that important morphological variability exists beyond that currently quantified by TIBI in populations experiencing or at-risk for acute ischemic stroke, and posit that the observed flow-types provide the foundation for objective methods of real-time automated flow type classification.

Large Vessel Occlusion Stroke Detection in the Prehospital Environment

“…However, there have been recent strides regarding complex training requirements of TCD operators and approach to results analysis. Cerebral blood flow velocity waveforms have been categorized into Thrombolysis in Brain Ischemia (TIBI) grading system which requires expert analysis, so TCD-derived morphological biomarkers such as Velocity Asymmetry Index (VAI) and Velocity Curvature Index (VCI) have been studied as user-independent LVO metrics that may lead to increased TCD use in the prehospital setting [51][52][53][54][55]. A robotically assisted ultrasound system has recently been developed to evaluate brain health in the acute setting with machine learning analysis of data [56].…”

Application of Controlled Hypotension During Surgery for Spinal Metastasis

“…TCD analysis of pulsatile cerebral blood flow velocity waveforms of intracranial arteries can provide information on various cerebrovascular changes. 56 The latest TCD includes the use of spectral and color Doppler as well as grayscale tissue imaging, allowing direct display of major intracranial arteries and the identification of arteries and their hemodynamics. 57 Cerebral blood flow resistance can be measured by the Pulsatility Index (PI).…”

A Review of the use of Transcranial Doppler Waveform Morphology for Acute Stroke Assessment

“…Additional works have focused enriching this classification with additional categorical information, utilizing data-driven spectral clustering analysis techniques to group pathological waveforms into strata that have TIBI analogs (Fig. 4) [33]. The utilization of machine learning techniques such as this allow for a dramatic reduction in the involve- ment of human error and subjectivity.…”

“…The utilization of machine learning techniques such as this allow for a dramatic reduction in the involve- ment of human error and subjectivity. The assignment of cluster labels to each waveform was also based on objectively computed CBFV waveform features (further described in Thorpe et al, 2020 [33]), thus representing a first step towards fully automated morphological categorization.…”