Intracranial hemorrhage detection in 3D computed tomography images using a bi-directional long short-term memory network-based modified genetic algorithm
IntroductionIntracranial hemorrhage detection in 3D Computed Tomography (CT) brain images has gained more attention in the research community. The major issue to deal with the 3D CT brain images is scarce and hard to obtain the labelled data with better recognition results.MethodsTo overcome the afo...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2023-07-01
|
| Series: | Frontiers in Neuroscience |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fnins.2023.1200630/full |
| _version_ | 1797788165792071680 |
|---|---|
| author | Jewel Sengupta Robertas Alzbutas Przemysław Falkowski-Gilski Bożena Falkowska-Gilska |
| author_facet | Jewel Sengupta Robertas Alzbutas Przemysław Falkowski-Gilski Bożena Falkowska-Gilska |
| author_sort | Jewel Sengupta |
| collection | DOAJ |
| description | IntroductionIntracranial hemorrhage detection in 3D Computed Tomography (CT) brain images has gained more attention in the research community. The major issue to deal with the 3D CT brain images is scarce and hard to obtain the labelled data with better recognition results.MethodsTo overcome the aforementioned problem, a new model has been implemented in this research manuscript. After acquiring the images from the Radiological Society of North America (RSNA) 2019 database, the region of interest (RoI) was segmented by employing Otsu’s thresholding method. Then, feature extraction was performed utilizing Tamura features: directionality, contrast, coarseness, and Gradient Local Ternary Pattern (GLTP) descriptors to extract vectors from the segmented RoI regions. The extracted vectors were dimensionally reduced by proposing a modified genetic algorithm, where the infinite feature selection technique was incorporated with the conventional genetic algorithm to further reduce the redundancy within the regularized vectors. The selected optimal vectors were finally fed to the Bi-directional Long Short Term Memory (Bi-LSTM) network to classify intracranial hemorrhage sub-types, such as subdural, intraparenchymal, subarachnoid, epidural, and intraventricular.ResultsThe experimental investigation demonstrated that the Bi-LSTM based modified genetic algorithm obtained 99.40% sensitivity, 99.80% accuracy, and 99.48% specificity, which are higher compared to the existing machine learning models: Naïve Bayes, Random Forest, Support Vector Machine (SVM), Recurrent Neural Network (RNN), and Long Short-Term Memory (LSTM) network. |
| first_indexed | 2024-03-13T01:31:43Z |
| format | Article |
| id | doaj.art-1f94c74e374a4bcf803094b8d927a50c |
| institution | Directory Open Access Journal |
| issn | 1662-453X |
| language | English |
| last_indexed | 2024-03-13T01:31:43Z |
| publishDate | 2023-07-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Neuroscience |
| spelling | doaj.art-1f94c74e374a4bcf803094b8d927a50c2023-07-04T06:46:51ZengFrontiers Media S.A.Frontiers in Neuroscience1662-453X2023-07-011710.3389/fnins.2023.12006301200630Intracranial hemorrhage detection in 3D computed tomography images using a bi-directional long short-term memory network-based modified genetic algorithmJewel Sengupta0Robertas Alzbutas1Przemysław Falkowski-Gilski2Bożena Falkowska-Gilska3Kaunas University of Technology, Kaunas, LithuaniaKaunas University of Technology, Kaunas, LithuaniaFaculty of Electronics, Telecommunications and Informatics, Gdansk University of Technology, Gdansk, PolandSpecialist Diabetes Outpatient Clinic, Olsztyn, PolandIntroductionIntracranial hemorrhage detection in 3D Computed Tomography (CT) brain images has gained more attention in the research community. The major issue to deal with the 3D CT brain images is scarce and hard to obtain the labelled data with better recognition results.MethodsTo overcome the aforementioned problem, a new model has been implemented in this research manuscript. After acquiring the images from the Radiological Society of North America (RSNA) 2019 database, the region of interest (RoI) was segmented by employing Otsu’s thresholding method. Then, feature extraction was performed utilizing Tamura features: directionality, contrast, coarseness, and Gradient Local Ternary Pattern (GLTP) descriptors to extract vectors from the segmented RoI regions. The extracted vectors were dimensionally reduced by proposing a modified genetic algorithm, where the infinite feature selection technique was incorporated with the conventional genetic algorithm to further reduce the redundancy within the regularized vectors. The selected optimal vectors were finally fed to the Bi-directional Long Short Term Memory (Bi-LSTM) network to classify intracranial hemorrhage sub-types, such as subdural, intraparenchymal, subarachnoid, epidural, and intraventricular.ResultsThe experimental investigation demonstrated that the Bi-LSTM based modified genetic algorithm obtained 99.40% sensitivity, 99.80% accuracy, and 99.48% specificity, which are higher compared to the existing machine learning models: Naïve Bayes, Random Forest, Support Vector Machine (SVM), Recurrent Neural Network (RNN), and Long Short-Term Memory (LSTM) network.https://www.frontiersin.org/articles/10.3389/fnins.2023.1200630/fullbi-directional long short-term memory networkcomputed tomographygenetic algorithmgradient local ternary patternintracranial hemorrhage detectionTamura features |
| spellingShingle | Jewel Sengupta Robertas Alzbutas Przemysław Falkowski-Gilski Bożena Falkowska-Gilska Intracranial hemorrhage detection in 3D computed tomography images using a bi-directional long short-term memory network-based modified genetic algorithm Frontiers in Neuroscience bi-directional long short-term memory network computed tomography genetic algorithm gradient local ternary pattern intracranial hemorrhage detection Tamura features |
| title | Intracranial hemorrhage detection in 3D computed tomography images using a bi-directional long short-term memory network-based modified genetic algorithm |
| title_full | Intracranial hemorrhage detection in 3D computed tomography images using a bi-directional long short-term memory network-based modified genetic algorithm |
| title_fullStr | Intracranial hemorrhage detection in 3D computed tomography images using a bi-directional long short-term memory network-based modified genetic algorithm |
| title_full_unstemmed | Intracranial hemorrhage detection in 3D computed tomography images using a bi-directional long short-term memory network-based modified genetic algorithm |
| title_short | Intracranial hemorrhage detection in 3D computed tomography images using a bi-directional long short-term memory network-based modified genetic algorithm |
| title_sort | intracranial hemorrhage detection in 3d computed tomography images using a bi directional long short term memory network based modified genetic algorithm |
| topic | bi-directional long short-term memory network computed tomography genetic algorithm gradient local ternary pattern intracranial hemorrhage detection Tamura features |
| url | https://www.frontiersin.org/articles/10.3389/fnins.2023.1200630/full |
| work_keys_str_mv | AT jewelsengupta intracranialhemorrhagedetectionin3dcomputedtomographyimagesusingabidirectionallongshorttermmemorynetworkbasedmodifiedgeneticalgorithm AT robertasalzbutas intracranialhemorrhagedetectionin3dcomputedtomographyimagesusingabidirectionallongshorttermmemorynetworkbasedmodifiedgeneticalgorithm AT przemysławfalkowskigilski intracranialhemorrhagedetectionin3dcomputedtomographyimagesusingabidirectionallongshorttermmemorynetworkbasedmodifiedgeneticalgorithm AT bozenafalkowskagilska intracranialhemorrhagedetectionin3dcomputedtomographyimagesusingabidirectionallongshorttermmemorynetworkbasedmodifiedgeneticalgorithm |