AUTHOR=Sengupta Jewel , Alzbutas Robertas , Falkowski-Gilski Przemysław , Falkowska-Gilska Bożena TITLE=Intracranial hemorrhage detection in 3D computed tomography images using a bi-directional long short-term memory network-based modified genetic algorithm JOURNAL=Frontiers in Neuroscience VOLUME=Volume 17 - 2023 YEAR=2023 URL=https://www.frontiersin.org/journals/neuroscience/articles/10.3389/fnins.2023.1200630 DOI=10.3389/fnins.2023.1200630 ISSN=1662-453X ABSTRACT=Intracranial hemorrhage recognition in 3D Computerized Tomography (CT) brain images gained more attention among the researcher’s community. The major issue to deal with the 3D CT brain images is scarce and hard to obtain labelled data with better recognition results. In order to overcome this issue, a new model is implemented in this research manuscript. After acquiring the images from Radiological Society of North America (RSNA) 2019 database, the Region of Interest (RoI) is segmented by employing Otsu thresholding method. Next, the feature extraction is performed utilizing Tamura features: directionality, contrast, and coarseness and Gradient Local Ternary Pattern (GLTP) descriptors for extracting vectors from segmented RoI regions. The extracted vectors are dimensionally reduced by proposing a modified genetic algorithm, where the infinite feature selection technique is incorporated with the conventional genetic algorithm for further reducing the redundancy within the regularized vectors. The selected optimal vectors are finally fed to the Bi-directional Long Short Term Memory (Bi-LSTM) network for classifying intracranial hemorrhage sub-types, like: subdural, intraparenchymal, subarachnoid, epidural and intraventricular. The 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 related 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.