New approach to detect and classify stroke in skull CT images via analysis of brain tissue densities
Introduction
Stroke is an injury that abruptly affects brain tissues. This disease is caused by a change in the blood supply to a particular region of the brain, and it results in the loss or reduction of its related functions.
Cerebral vascular accidents (CVA) affect 16 million people worldwide every year, and 6 million of these people die [1]. However, another important problem related to strokes, besides mortality, is that many survivors have chronic consequences that are complex and heterogeneous. Studies estimate that there are about 2 million people with sequels to strokes in the United States. Treatment of these people represent a cost around of 30 billion dollars per year [1], [2], [3].
In research conducted by Rangel [4] with 181 stroke patients, the sequels cause dependence for trivial activities. The research showed that 49.6% of the patients had moderate to severe dependence and 49.7% had dysphoric or depressive symptoms.
However, if treatment is performed immediately, recovery in the damaged region is possible. For this, some studies indicate that early stroke detection is performed from the analysis of cerebral microbleeds (CMBs), that is a prodromal stage of stroke [5]. Medical images obtained by Computed Tomography (CT) or Methods using Magnetic Resonance Imaging (MRI) are very important to aid diagnose this disease [6]. Although the two exams present detailed images of body parts, each method has different objectives and operating mechanisms, as well dissimilar applications [7]. MRI has higher sensitivity and it has no radiation to patient, different from TC scan exam. However, this method has limitations, due to the longer examination time and the costs, both of which hamper its use on a large scale; consequently, it is usually only found at large hospitals [8]. Otherwise, CT is faster, is less affected by noise and has a lower cost, so is available to a larger number of people. These tests allow the extent and severity of the stroke to be evaluated by evaluating the flow and blood volume [9]. Also, CT is more indicated for lung and bone analysis, in the study of abdominal structures and for initial assessment of the central nervous system [10].
Some characteristics of a CT image can be used to highlight the differences and similarities between the objects within the image. The medical specialist performs a visual segmentation of the exams to try to distinguish particular areas that display pathological changes. However, the changes that appear on CT images are subtle and in some cases imperceptible to a simple visual analysis [11]; moreover, CT scans provide information that are invisible to the human eye.
These digital images usually have approximately 4000 different gray levels. Whereas the Human Visual System (HVS) is able to distinguish only about 64 grayscale for the same condition of lightness. Thus, a computational analysis may reveal a much larger number of attributes that cannot be detected by the human eye.
Section snippets
Literature review
The use of computing systems, in conjunction with medical knowledge, have an important role in several areas of medicine. The contribution is significant in aiding medical diagnosis, which motivates researchers to invest time and effort to study Artificial Intelligence (AI). These efforts aim to develop new techniques and systems that assist medical imaging. Thus, existing techniques are improved or new methods are developed [12], [13], [14].
The study developed by Elias Restum [15] shows that
Materials and methods
In general, computer vision systems involve the steps of preprocessing, Region of Interest (ROI) segmentation, feature extraction and machine learning. Fig. 1 shows the stages to identify and classify stroke regions on the proposed study.
The stages of the proposed approach used in this study, as presented in Fig. 1, are presented in details on the sections bellow: image acquisition, ROI segmentation, feature extraction, and machine learning. All this stages are used in each evaluated methods
Analysis of brain tissue densities
The images obtained from a CT scan present a large variety of gray levels which is the measure attenuation of the X-ray beam. This attenuation, is proportional to the local tissue density and is designated the attenuation coefficient (AC) measured in HUs [46]. Materials denser than water have higher attenuation values, and occupy the positive tracks on the HU scale. These denser materials appear lighter than water on the gray tone scale. In contrast the low density tissues take the negative
Results
This section presents the results obtained by the proposed extraction method. The proposed method was validated using 420 images from a database, which were equally divided into images from healthy patients, ischemic stroke patients and hemorrhagic stroke patients. The computational performance and efficacy of the proposed method was evaluated with computer simulations using the MLP, kNN, Bayesian, SVM and OPF classifiers. The results are analyzed in two steps: first with the optimal
Discussion about statistical analysis
Considering the statistical analysis presented in 5.2.1 Statistical analysis, Table 5 shows that some extractors got better results with specific classifiers, for example: the LBP and CM obtained their best results with the classifiers kNN, MLP and OPF using the Euclidean kernel represented by OPF(E), Hu’s Moments reached its best rates with the SVM, Zernike’s Moments presented best results with the MLP, and the GLCM obtained average Acc greater than 96.24% for all classifiers, proving that is
Conclusion
The main contribution in this paper is the development of a new feature extraction technique for CT images of the brain, called ABTD, which is based on radiation attenuation patterns of brain tissues. This method in conjunction with Machine Learning techniques may assist in the detection and classification process of a possible stroke.
The work in this paper defined the optimal configuration of this method for the identification 605 and classification of brain stroke in CT images, and produced a
Acknowledgment
The authors acknowledge the financial support and encouragement from Brazilian National Council for Research and Development (CNPq).
The authors thank the Graduate Program in Computer Science from Instituto Federal do Ceará and the Department of Computer Engineering and Walter Cantídio University Hospital of Universidade Federal do Ceará for technical support in Pulmonology and images.
The first author acknowledges the sponsorship from the Federal Institute of Education, Science and Technology of
References (58)
- et al.
Automatic liver segmentation in MRI images using an iterative watershed algorithm and artificial neural network
Biomed. Signal Process. Control
(2012) - et al.
Ct enterography: concept, technique, and interpretation
Radiol. Clin. North Am.
(2009) - et al.
Automated delineation of stroke lesions using brain CT images
NeuroImage Clin.
(2014) - et al.
A new computationally efficient CAD system for pulmonary nodule detection in CT imagery
Med. Image. Anal.
(2010) - et al.
Novel and powerful 3D adaptive crisp active contour method applied in the segmentation of CT lung images
Med. Image Anal.
(2017) - et al.
Automatic detection of breast cancer mitotic cells based on the combination of textural, statistical and innovative mathematical features
Appl. Math. Model
(2015) Computer-aided diagnosis in radiology
Acad. Radiol.
(2002)- et al.
Improved early stroke detection: wavelet-based perception enhancement of computerized tomography exams
Comput. Biol. Med.
(2007) - et al.
An image feature approach for computer-aided detection of ischemic stroke
Comput. Biol. Med.
(2011) - et al.
Generation of moment invariants and their uses for character recognition
Pattern Recognit. Lett.
(1995)
A survey of moment-based techniques for unoccluded object representation and recognition
CVGIP: Gr. Models Image Process.
Texture classification using texture spectrum
Pattern Recognit.
Efficient supervised optimum-path forest classification for large datasets
Pattern Recognit.
Ct and mr imaging of intracerebral hemorrhage
Intracerebral Hemorrhage
Heart disease and stroke statistics 2010 update a report from the american heart association
Circulation
Heart disease and stroke statistics-2006 update a report from the american heart association statistics committee and stroke statistics subcommittee
Circulation
Executive summary: heart disease and stroke statistics-2012 update a report from the american heart association
Circulation
Qualidade de vida de pacientes com acidente vascular cerebral em reabilitação
Acta Paulista de Enfermagem
Voxelwise detection of cerebral microbleed in CADASIL patients by leaky rectified linear unit and early stopping
Multimed. Tools Appl.
Principles of Computerized Tomographic Imaging
Automatic liver parenchyma segmentation from abdominal CT images using support vector machines
Proceedings of the 2009 ICME International Conference on Complex Medical Engineering
3D Lung fissure segmentation in TC images based in textures
IEEE Lat. Am. Trans.
3D adaptive balloon active contour: method of segmentation of structures in three dimensions
IEEE Lat. Am. Trans.
Ferramenta de Auxílio Diagnóstico para o Tratamento de Acidentes Vasculares Cerebrais, Ph.D. thesis
Ambiente virtual colaborativo de diagnóstico a distância integrado a ferramentas de manipulação de imagens
Rotation-invariant feature extraction using a structural co-occurrence matrix
Measurement
Computer aided detection of ischemic stroke using segmentation and texture features
Measurement
Cited by (59)
The future of health diagnosis and treatment: an exploration of deep learning frameworks and innovative applications
2023, Deep Learning in Personalized Healthcare and Decision SupportLeveraging artificial intelligence in ischemic stroke imaging
2022, Journal of NeuroradiologyBrain stroke detection from computed tomography images using deep learning algorithms
2022, Applications of Artificial Intelligence in Medical ImagingDisplacement voxelization to resolve mesh-image mismatch: Application in deriving dense white matter fiber strains
2022, Computer Methods and Programs in BiomedicineDeep learning applications for IoT in health care: A systematic review
2021, Informatics in Medicine UnlockedArtificial intelligence in cerebral stroke images classification and segmentation: A comprehensive study
2024, Multimedia Tools and Applications