Radiomics approach to quantify shape irregularity from crowd-based qualitative assessment of intracranial aneurysms

Abstract

The morphological assessment of anatomical structures is clinically relevant, but often falls short of quantitative or standardised criteria. Whilst human observers are able to assess morphological characteristics qualitatively, the development of robust shape features remains challenging. In this study, we employ psychometric and radiomic methods to develop quantitative models of the perceived irregularity of intracranial aneurysms (IAs). First, we collect morphological characteristics (e.g. irregularity, asymmetry) in imaging-derived data and aggregated the data using rank-based analysis. Second, we compute regression models relating quantitative shape features to the aggregated qualitative ratings (ordinal or binary). We apply our method for quantifying perceived shape irregularity to a dataset of 134 IAs using a pool of 179 different shape indices. Ratings given by 39 participants show good agreement with the aggregated ratings (Spearman rank correlation Sp=0.84). The best-performing regression model based on quantitative shape features predicts the perceived irregularity with R2:0.84±0.05.

Intracranial Aneurysm Rupture Risk Estimation Utilizing Vessel-Graphs and Machine Learning

“…Initially, radiomics has been employed for oncological applications and an emerging technique in the cardiovascular field, especially with MRI. This observation has been confirmed by studies [15][16][17][18][19], which are extracting shape-based radiomics features and classifying diverse heart diseases.…”

“…The radiomics feature classes we extract are: First Order, Shape (3D), Gray Level Co-occurrence Matrix (GLCM), Gray Level Size Zone Matrix (GLSZM), Gray Level Run Length Matrix (GLRLM), Neighbouring Gray Tone Difference Matrix (NGTDM), and Gray Level Dependence Matrix (GLDM). The shape-based features have shown to be useful for assessing cerebral aneurysm ruptures [15][16][17][18][19]. We hypothesize that the radiomics texture features approximate flow characteristics within the aneurysm and the surrounding vessel due to the injected contrast agent.…”

CADA Challenge: Rupture Risk Assessment Using Computational Fluid Dynamics

“…The shape of aneurysms can be well quantified by data driven approaches, see for example the radiomics approach recently proposed by Juchler et al [27]. Future works may use a combination of CFD and shape analysis techniques to arrive at robust and fast analyses of aneurysm rupture risk.…”

Shape irregularity of the intracranial aneurysm lumen exhibits diagnostic value

“…Our method is generally robust with respect to missing or outlying rating data. [13] Missing clinical data was also handled by exclusion, under the assumption that the misses occurred at random and independent of the property under examination. The numbers of valid cases per property are reported in Table 2.…”

“…We have previously developed a quantitative model for lumen irregularity that matches the human perception of shape. [13] We employed a psychometric method to measure the perceived irregularity of IA domes from human raters assessing highly resolved representations of IA lumens. We reproduced aggregated shape assessments accurately by using a multivariate model of quantitative shape features that can be computed automatically from image data.…”