ABSTRACT
Saccular intracranial aneurysms corresponding to their berry or sac-like shape are characterized as most vulnerable to grow and rupture relatively quicker than other known types, such as, fusiform, distal and dissection. Approximately 80% of aneurysmal Subarachnoid Hemorrhage (aSAH) are reckoned due to burst of saccular aneurysms only. Therefore, timely detection of unruptured saccular brain aneurysms can greatly help neurosurgeons to treat them well before an aSAH occurs, and thus saves precious lives. Most of the research efforts in this respect involve statistical analysis of manually collected retrospective data to predict the risk of aneurysmal rupture. Whereas, geometrical, anatomical and textural characteristics of medical imaging, such as digital subtraction angiography (DSA) and magnetic resonance angiography (MRA), is not investigated for accurate detection of unruptured saccular aneurysms and their individualized prediction of rupture likelihood. The main contribution of this work is a) 98% accurate identification of aneurysms from both DSA and MRA using Multilayer Perceptron Neural Network trained upon robust Haralick textural features of individual regions of interest (ROIs) segmented through Watershed Segmentation and Distance Transformation; b) prediction of aneurysms rupture probability, by categorizing aneurysmal ROIs into 5 classes, using geometrical, anatomical and textural features, with 76.67% accuracy using Decision Support Trees, ensembled with Bagging classifier. Our evaluation is based upon de-identified dataset of 180 images (54 MRA, and 126 DSA), obtained from Henry Ford Hospital, Bloomfield Hills, MI, USA, after IRB approval (No. 11254).
- Yoon N. K. et al. 2016. Imaging of cerebral aneurysms: a clinical perspective. Neurovascular Imaging 2:6.Google ScholarCross Ref
- Greving et al. 2014. Development of the PHASES score for prediction of risk of rupture of intracranial aneurysms: a pooled analysis of six prospective cohort studies. The Lancet Neurology 13.1: 59--66.Google ScholarCross Ref
- Siuly S. and Zhang, Y. 2016. Medical Big Data: Neurological Diseases Diagnosis through Medical Data Analysis. Data Sci. Eng. 1(2):54--64.Google Scholar
- Kassner and Thornhill R.E. 2010. Texture Analysis: A Review of Neurologic MR Imaging Applications. Published April 15 asGoogle Scholar
- Moghaddam M. J. and Soltanian-Zadeh H. 2010. Medical Image Segmentation Using Artificial Neural Networks, Artificial Neural Networks - Methodological Advances and Biomedical Applications. Prof. Kenji Suzuki (Ed.), ISBN: 978-953-307-243-2, InTech.Google Scholar
- Kostopoulos S. et al. 2007. A Hybrid pixel-based classification method for blood vessel segmentation and aneurysm detection on CTA. Elsevier, Computers & Graphics 31 (2007) 493--500. Google ScholarDigital Library
- Yohichi I. et al. 2018. Detection rates and sites of unruptured intracranial aneurysms according to sex and age: an analysis of MR angiography -- based brain examinations of 4070 healthy Japanese adults. J Neurosurg (April 06, 2018).Google Scholar
- Park S. and Lee D. H. et al. 2018. Incidental Saccular Aneurysms on Head MR Angiography: 5 Years' Experience at a Single Large-Volume Center (Mar 13, 2018).Google Scholar
- Millán R. D. et al. 2007. Morphological Characterization of Intracranial Aneurysms Using 3-D Moment Invariants. IEEE Transactions on Medical Imaging, VOL. 26, NO. 9 (September 2007).Google ScholarCross Ref
- Kostopoulos S. et al. 2007. A Hybrid pixel-based classification method for blood vessel segmentation and aneurysm detection on CTA. Elsevier, Computers & Graphics 31 (2007) 493--500. Google ScholarDigital Library
- Jerman T. et al. 2016. Blob Enhancement and Visualization for Improved Intracranial Aneurysm Detection. IEEE Transaction on Visualization and Computer Graphics, VOL. 22, No. 6. Google ScholarDigital Library
- Yang X. et al. 2011. Computer-Aided Detection of Intracranial Aneurysm in MR Angiography. Journal of Imaging, Vol 24--1: pp86--95.Google Scholar
- Hentschke C. M. et al. 2011. Automatic Cerebral Aneurysm Detection in Multimodal Angiographic Images. IEEE Nuclear Science Symposium Conference Record, MIC12.M-193.Google Scholar
- Khalid M. and Shakeel M. Anjum, et al. 2018. ISADAQ--A Framework for Intracranial Saccular Aneurysm Detection and Quantification using Morphological Analysis of Cerebral Angiograms. IEEE Access, PP(99):1--1.Google Scholar
- Rahmany et al. 2014. Detection of Intracranial Aneurysm in Angiographic Images Using Fuzzy Approaches. IEEE IPAS'14: International Image Processing Applications and Systems Conference.Google Scholar
- Nomura Y. and Masutani Y. et al. 2014. Performance improvement in computerized detection of cerebral aneurysms by retraining classifier using feedback data collected in routine reading environment. SEIDU, Journal of Biomedical Graphics and Computing. VOL. 4, NO. 4.Google ScholarCross Ref
- Farhan S. and Fahim M. A. et al. 2014. An Ensemble-of-Classifiers Based Approach for Early Diagnosis of Alzheimer's Disease: Classification Using Structural Features of Brain Images. Computational and Mathematical Models in Medicine. Vol 2014, Article ID 862307.Google Scholar
- Haralick R. M. 1979. Statistical and structural approaches to texture. In proceeding of IEEE (Volume: 67, Issue: 5, May 1979).Google Scholar
- Breiman L. 1996. Bagging predictors. Mach. Learn. 24(2):123--140. Google ScholarDigital Library
- Freund Y., and Schapire R.E. 1996. Experiments with a new boosting algorithm. Proceedings 13th International Conference on Machine Learning:148--156. Google ScholarDigital Library
- Breiman L. 2001. Random forests. Mach. Learn. 45(1):5--32 Google ScholarDigital Library
- Gama J., and Brazdil P. 2000. Cascade Generalization. Mach. Learn. 41(3):315--343. Google ScholarDigital Library
- Wolpert D., and Stacked, H. 1992. Generalization. Neural Netw. 5:241--259. Google ScholarDigital Library
Index Terms
- Saccular Brain Aneurysm Detection and Multiclassifier Rupture Prediction using Digital Subtraction and Magnetic Resonance Angiograms
Recommendations
Simulation of cerebral aneurysm growth and prediction of evolving rupture risk
Cerebral aneurysms are local expansions of blood vessel walls in the brain blood system. The rupture of an aneurysm is a very severe event associated with a high rate of mortality.When cerebral aneurysms are detected, clinicians need to decide if ...
Risk of rupture of the cerebral aneurysm in relation to traumatic brain injury using a patient-specific fluid-structure interaction model
Highlights- Patient-specific FSI model of the skull, brain, and cerebral aneurysm was established.
Abstract Background and objectiveCerebral aneurysm, which is defined as one of the weakened area in the wall of an artery in the brain, ruptures when wall tension exceeds its mechanical strength. Traumatic brain injury (TBI) by ...
A novel computational fluid dynamic method and validation for assessing distal cerebrovascular microcirculatory resistance
Highlights- This study developed novel computational strategies that allowed us to estimate the blood flow and resistance distal to the stenosis by using CFD simulation.
- Our computational strategy may serve as an effective approach to evaluate the ...
Abstract Background and objectiveThe non-invasive assessment of microcirculatory resistance could improve the treatment of cerebrovascular stenosis. This study aimed to validate a novel computational strategy for determining the reference value of ...
Comments