Motor Imagery Multi-Tasks Classification for BCIs Using the NVIDIA Jetson TX2 Board and the EEGNet Network.

Nowadays, Brain-Computer Interfaces (BCIs) still captivate large interest because of multiple advantages offered in numerous domains, explicitly assisting people with motor disabilities in communicating with the surrounding environment. However, challenges of portability, instantaneous processing time, and accurate data processing remain for numerous BCI system setups. This work implements an embedded multi-tasks classifier based on motor imagery using the EEGNet network integrated into the NVIDIA Jetson TX2 card. Therefore, two strategies are developed to select the most discriminant channels. The former uses the accuracy based-classifier criterion, while the latter evaluates electrode mutual information to form discriminant channel subsets. Next, the EEGNet network is implemented to classify discriminant channel signals. Additionally, a cyclic learning algorithm is implemented at the software level to accelerate the model learning convergence and fully profit from the NJT2 hardware resources. Finally, motor imagery Electroencephalogram (EEG) signals provided by HaLT's public benchmark were used, in addition to the k-fold cross-validation method. Average accuracies of 83.7% and 81.3% were achieved by classifying EEG signals per subject and motor imagery task, respectively. Each task was processed with an average latency of 48.7 ms. This framework offers an alternative for online EEG-BCI systems' requirements, dealing with short processing times and reliable classification accuracy.

Vascular Structure Identification in Intraoperative 3D Contrast-Enhanced Ultrasound Data.

In this paper, a method of vascular structure identification in intraoperative 3D Contrast-Enhanced Ultrasound (CEUS) data is presented. Ultrasound imaging is commonly used in brain tumor surgery to investigate in real time the current status of cerebral structures. The use of an ultrasound contrast agent enables to highlight tumor tissue, but also surrounding blood vessels. However, these structures can be used as landmarks to estimate and correct the brain shift. This work proposes an alternative method for extracting small vascular segments close to the tumor as landmark. The patient image dataset involved in brain tumor operations includes preoperative contrast T1MR (cT1MR) data and 3D intraoperative contrast enhanced ultrasound data acquired before (3D-iCEUS(start) and after (3D-iCEUS(end) tumor resection. Based on rigid registration techniques, a preselected vascular segment in cT1MR is searched in 3D-iCEUS(start) and 3D-iCEUS(end) data. The method was validated by using three similarity measures (Normalized Gradient Field, Normalized Mutual Information and Normalized Cross Correlation). Tests were performed on data obtained from ten patients overcoming a brain tumor operation and it succeeded in nine cases. Despite the small size of the vascular structures, the artifacts in the ultrasound images and the brain tissue deformations, blood vessels were successfully identified.

Improving convolutional neural network learning based on a hierarchical bezier generative model for stenosis detection in X-ray images.

BACKGROUND AND OBJECTIVE: Automatic detection of stenosis on X-ray Coronary Angiography (XCA) images may help diagnose early coronary artery disease. Stenosis is manifested by a buildup of plaque in the arteries, decreasing the blood flow to the heart, increasing the risk of a heart attack. Convolutional Neural Networks (CNNs) have been successfully applied to identify pathological, regular, and featured tissues on rich and diverse medical image datasets. Nevertheless, CNNs find operative and performing limitations while working with small and poorly diversified databases. Transfer learning from large natural image datasets (such as ImageNet) has become a de-facto method to improve neural networks performance in the medical image domain. METHODS: This paper proposes a novel Hierarchical Bezier-based Generative Model (HBGM) to improve the CNNs training process to detect stenosis. Herein, artificial image patches are generated to enlarge the original database, speeding up network convergence. The artificial dataset consists of 10,000 images containing 50% stenosis and 50% non-stenosis cases. Besides, a reliable Fréchet Inception Distance (FID) is used to evaluate the generated data quantitatively. Therefore, by using the proposed framework, the network is pre-trained with the artificial datasets and subsequently fine-tuned using the real XCA training dataset. The real dataset consists of 250 XCA image patches, selecting 125 images for stenosis and the remainder for non-stenosis cases. Furthermore, a Convolutional Block Attention Module (CBAM) was included in the network architecture as a self-attention mechanism to improve the efficiency of the network. RESULTS: The results showed that the pre-trained networks using the proposed generative model outperformed the results concerning training from scratch. Particularly, an accuracy, precision, sensitivity, and F1-score of 0.8934, 0.9031, 0.8746, 0.8880, 0.9111, respectively, were achieved. The generated artificial dataset obtains a mean FID of 84.0886, with more realistic visual XCA images. CONCLUSIONS: Different ResNet architectures for stenosis detection have been evaluated, including attention modules into the network. Numerical results demonstrated that by using the HBGM is obtained a higher performance than training from scratch, even outperforming the ImageNet pre-trained models.

Segmentation of brain tumour in 3D Intraoperative Ultrasound imaging.

BACKGROUND: Intraoperative ultrasound (iUS), using a navigation system and preoperative magnetic resonance imaging (pMRI), supports the surgeon intraoperatively in identifying tumour margins. Therefore, visual tumour enhancement can be supported by efficient segmentation methods. METHODS: A semi-automatic and two registration-based segmentation methods are evaluated to extract brain tumours from 3D-iUS data. The registration-based methods estimated the brain deformation after craniotomy based on pMRI and 3D-iUS data. Both approaches use the normalised gradient field and linear correlation of linear combinations metrics. Proposed methods were evaluated on 66 B-mode and contrast-mode 3D-iUS data with metastasis and glioblastoma. RESULTS: The semi-automatic segmentation achieved superior results with dice similarity index (DSI) values between [85.34, 86.79]% and contour mean distance values between [1.05, 1.11] mm for both modalities and tumour classes. CONCLUSIONS: Better segmentation results were obtained for metastasis detection than glioblastoma, preferring 3D-intraoperative B-mode over 3D-intraoperative contrast-mode.

Multi-Stroke handwriting character recognition based on sEMG using convolutional-recurrent neural networks.

Despite the increasing use of technology, handwriting has remained to date as an efficient means of communication. Certainly, handwriting is a critical motor skill for childrens cognitive development and academic success. This article presents a new methodology based on electromyographic signals to recognize multi-user free-style multi-stroke handwriting characters. The approach proposes using powerful Deep Learning (DL) architectures for feature extraction and sequence recognition, such as convolutional and recurrent neural networks. This framework was thoroughly evaluated, obtaining an accuracy of 94.85%. The development of handwriting devices can be potentially applied in the creation of artificial intelligence applications to enhance communication and assist people with disabilities.

Coronary artery segmentation in X-ray angiograms using gabor filters and differential evolution.

Segmentation of coronary arteries in X-ray angiograms represents an essential task for computer-aided diagnosis, since it can help cardiologists in diagnosing and monitoring vascular abnormalities. Due to the main disadvantages of the X-ray angiograms are the nonuniform illumination, and the weak contrast between blood vessels and image background, different vessel enhancement methods have been introduced. In this paper, a novel method for blood vessel enhancement based on Gabor filters tuned using the optimization strategy of Differential evolution (DE) is proposed. Because the Gabor filters are governed by three different parameters, the optimal selection of those parameters is highly desirable in order to maximize the vessel detection rate while reducing the computational cost of the training stage. To obtain the optimal set of parameters for the Gabor filters, the area (Az) under the receiver operating characteristics curve is used as objective function. In the experimental results, the proposed method achieves an Az=0.9388 in a training set of 40 images, and for a test set of 40 images it obtains the highest performance with an Az=0.9538 compared with six state-of-the-art vessel detection methods. Finally, the proposed method achieves an accuracy of 0.9423 for vessel segmentation using the test set. In addition, the experimental results have also shown that the proposed method can be highly suitable for clinical decision support in terms of computational time and vessel segmentation performance.

Patient-specific model-based segmentation of brain tumors in 3D intraoperative ultrasound images.

PURPOSE: Intraoperative ultrasound (iUS) imaging is commonly used to support brain tumor operation. The tumor segmentation in the iUS images is a difficult task and still under improvement because of the low signal-to-noise ratio. The success of automatic methods is also limited due to the high noise sensibility. Therefore, an alternative brain tumor segmentation method in 3D-iUS data using a tumor model obtained from magnetic resonance (MR) data for local MR-iUS registration is presented in this paper. The aim is to enhance the visualization of the brain tumor contours in iUS. METHODS: A multistep approach is proposed. First, a region of interest (ROI) based on the specific patient tumor model is defined. Second, hyperechogenic structures, mainly tumor tissues, are extracted from the ROI of both modalities by using automatic thresholding techniques. Third, the registration is performed over the extracted binary sub-volumes using a similarity measure based on gradient values, and rigid and affine transformations. Finally, the tumor model is aligned with the 3D-iUS data, and its contours are represented. RESULTS: Experiments were successfully conducted on a dataset of 33 patients. The method was evaluated by comparing the tumor segmentation with expert manual delineations using two binary metrics: contour mean distance and Dice index. The proposed segmentation method using local and binary registration was compared with two grayscale-based approaches. The outcomes showed that our approach reached better results in terms of computational time and accuracy than the comparative methods. CONCLUSION: The proposed approach requires limited interaction and reduced computation time, making it relevant for intraoperative use. Experimental results and evaluations were performed offline. The developed tool could be useful for brain tumor resection supporting neurosurgeons to improve tumor border visualization in the iUS volumes.

Automatic selection of localized region-based active contour models using image content analysis applied to brain tumor segmentation.

Brain tumor segmentation is a routine process in a clinical setting and provides useful information for diagnosis and treatment planning. Manual segmentation, performed by physicians or radiologists, is a time-consuming task due to the large quantity of medical data generated presently. Hence, automatic segmentation methods are needed, and several approaches have been introduced in recent years including the Localized Region-based Active Contour Model (LRACM). There are many popular LRACM, but each of them presents strong and weak points. In this paper, the automatic selection of LRACM based on image content and its application on brain tumor segmentation is presented. Thereby, a framework to select one of three LRACM, i.e., Local Gaussian Distribution Fitting (LGDF), localized Chan-Vese (C-V) and Localized Active Contour Model with Background Intensity Compensation (LACM-BIC), is proposed. Twelve visual features are extracted to properly select the method that may process a given input image. The system is based on a supervised approach. Applied specifically to Magnetic Resonance Imaging (MRI) images, the experiments showed that the proposed system is able to correctly select the suitable LRACM to handle a specific image. Consequently, the selection framework achieves better accuracy performance than the three LRACM separately.

Fast Parabola Detection Using Estimation of Distribution Algorithms.

This paper presents a new method based on Estimation of Distribution Algorithms (EDAs) to detect parabolic shapes in synthetic and medical images. The method computes a virtual parabola using three random boundary pixels to calculate the constant values of the generic parabola equation. The resulting parabola is evaluated by matching it with the parabolic shape in the input image by using the Hadamard product as fitness function. This proposed method is evaluated in terms of computational time and compared with two implementations of the generalized Hough transform and RANSAC method for parabola detection. Experimental results show that the proposed method outperforms the comparative methods in terms of execution time about 93.61% on synthetic images and 89% on retinal fundus and human plantar arch images. In addition, experimental results have also shown that the proposed method can be highly suitable for different medical applications.

Segmentation of Coronary Angiograms Using Gabor Filters and Boltzmann Univariate Marginal Distribution Algorithm.

This paper presents a novel method for improving the training step of the single-scale Gabor filters by using the Boltzmann univariate marginal distribution algorithm (BUMDA) in X-ray angiograms. Since the single-scale Gabor filters (SSG) are governed by three parameters, the optimal selection of the SSG parameters is highly desirable in order to maximize the detection performance of coronary arteries while reducing the computational time. To obtain the best set of parameters for the SSG, the area (Az ) under the receiver operating characteristic curve is used as fitness function. Moreover, to classify vessel and nonvessel pixels from the Gabor filter response, the interclass variance thresholding method has been adopted. The experimental results using the proposed method obtained the highest detection rate with Az = 0.9502 over a training set of 40 images and Az = 0.9583 with a test set of 40 images. In addition, the experimental results of vessel segmentation provided an accuracy of 0.944 with the test set of angiograms.