Search

• A shallow model, LGI-rPPG-Net, is proposed to produce rPPG signals that are highly correlated with finger PPG. • The proposed model managed to produce a better estimation of heart rate from rPPG. • The shallow architecture of the model made it suitable for real-time deployment. • Robust error analysis for signal synthesis and heart rate estimation. A method to accurately estimate physiological signals from video streams at a minimal cost is invaluable. The importance of such a technique in pre-clinical health monitoring cannot be understated. Remote photoplethysmography (rPPG) can be used as a substitute for finger photoplethysmography (PPG) when such sensors are not recommended, such as for burn victims, premature babies, and patients with sensitive skin. Good quality rPPG signal that is highly correlated to finger PPG can be used to estimate many vital health signs. In this work, a shallow encoder-decoder architecture, LGI-rPPG-Net is proposed. The proposed model aims to produce highly correlated rPPG signals which can be substituted for finger PPG. In the reconstruction of rPPG, the model achieved a very good Pearson's Correlation Coefficient (PCC), Root Mean Squared Error (RMSE), and dynamic time warping distance of 0.862, 0.148, and 0.699, respectively. This highly correlated rPPG was compared to finger PPG by calculating heart rate from rPPG and finger PPG. The model achieved a PCC of 0.984 and RMSE, and MAE of 2.91, 1.51 beats per minute (BPM), respectively. LGI-rPPG-Net model with video streaming to predict rPPG can thus be used as a replacement for finger PPG where in-contact collection is not feasible. [ABSTRACT FROM AUTHOR]

Emotion perception in stroke patients is affected since there is abnormality in the brain. Here, researchers focused on the impact of left brain damage and right brain damage towards emotion recognition. Due to the impaired emotion recognition, it is a challenge for stroke patients to express themselves in daily communication. Hence, it is inspiring to see the possibility to predict patient’s emotional state so as to prevent recurrent stroke. In this work, electroencephalograph (EEG) of 19 left brain damage patients (LBD), 19 right brain damage patients (RBD) and 19 normal control (NC) are collected as database. During data collection, six emotions (sad, disgust, fear, anger, happy and surprise) are induced by using audio visual stimuli. After normalization, EEG signals are filtered by using Butterworth 6th order band-pass filter at the cut-off frequencies of 0.5 Hz and 49 Hz. Then, wavelet packet transform (WPT) technique is implemented to localize five frequency bands: alpha (8 Hz–13 Hz), beta (13 Hz–30 Hz), gamma (30 Hz–49 Hz), alpha-to-gamma (8 Hz–49 Hz), beta-to-gamma (13 Hz–49 Hz). In WPT, four wavelet families are chosen: daubechies 4 (db4), daubechies 6 (db6), coiflet 5 (coif5) and symmlet 8 (sym8). Hurst exponents are extracted from each band and wavelet family and are classified by using K-nearest Neighbour (KNN) and Probabilistic Neural Network (PNN). Two classifications are done: comparison between three groups and comparison between six emotions. The results showed that all the H values are anti-correlated (0 < H < 0.5). From classification, the best frequency band is beta band, where sad emotion recorded the accuracy of 82.32% for LBD group. Meanwhile, both sad and fear emotion recorded 0.89 sensitivity score in LBD and RBD respectively. Due to its overall poor performance, RBD is found to have greater impairment in emotion recognition. [ABSTRACT FROM AUTHOR]

• Proposed methodology is based on flexible analytic wavelet transform (FAWT) for detection of Parkinson's disease where EEG data is collected from two different centers. • The ninteen entropy-based features are extracted from sub-bands obtained after FAWT. • Relevant features are ranked using analysis of variance (ANOVA) to achieve accurate classification of Parkinson's disease using k-nearest neighbor classifier. • The effectiveness of proposed approach is evaluated accurately in real-time with optimum computation time. Parkinson's disease (PD), a neurodegenerative disorder characterized by rest tremors, muscular rigidity, and bradykinesia, has become a global health concern. Currently, a neurologist determines the diagnosis of Parkinson's disease by taking into account several factors. An automated decision-making system would enhance patient care and improve the outcomes for the patient. Biomarkers, such as electroencephalograms (EEGs), can aid in the diagnosis process as they have proven useful in detecting abnormalities in the brain. This study presents a novel algorithm for the automated diagnosis of Parkinson's disease from EEG signals using a flexible analytic wavelet transform (FAWT). First, these acquired EEG signals are preprocessed before decomposition into five frequency sub-bands based on the FAWT method. Several entropy parameters are computed from the decomposed sub-bands and ranked based on their significance level in detecting PD through analysis of variance (ANOVA). Various classifiers are used to identify appropriate feature sets, including support vector machines (SVM), logistics, random forests (RF), radial basis functions (RBF), and k-nearest neighbors (KNN). The proposed approach is evaluated using data collected from two centers in Malaysia (Dataset-I) and the United States (Dataset-II). In dataset-I, the KNN classifier produces accuracy, specificity, sensitivity, and area under the curve of 99%, 99.45%, 99.12%, and 0.991, respectively, while in dataset-II, these values are 95.85%, 95.88%, 96.14%, and 0.959. The proposed system would be extremely useful for neurologists during their diagnostic process, as well as for current clinical practices. [ABSTRACT FROM AUTHOR]

• We proposed a fully automated Computer-Aided Diagnosis system for Diabetic Retinopathy (DR). Two specific problems have been addressed, namely, DR detection and DR grading for severity assessment. • Novel deep learning framework called DRNet has been developed for DR detection and DR grading. • Maximum mean accuracy, sensitivity and specificity of 99.72%, 99.86%, and 99.62% achieved for DR detection. • Achieved a maximum mean accuracy of 99.18%, sensitivity of 97.41%, and specificity of 99.56% in DR grading. Diabetes Retinopathy (DR) is a major microvascular complication of diabetes. Computer-Aided Diagnosis (CAD) tools for DR management are primarily developed using Artificial Intelligence (AI) methods, such as machine and deep learning algorithms. DR diagnostic tools have been developed in recent years using deep learning models. Thus, these models require large amounts of data for training. Consequently, these huge amounts of data are not balanced due to fewer cases in the dataset. To solve the problems associated with training models with small datasets, such as overfitting and poor approximation, this paper proposes a paradigm called Few-Shot Learning (FSL) which uses a relatively small amount of training data to train the models effectively. This paper proposes a novel prototype network, a type of FSL classification network capable of grading and detecting DR based on attention. The DRNet framework uses episodic learning to train its model on few-shot classification tasks. We developed a DRNet based on the APTOS2019 dataset for diabetic detection and grading. In the proposed network, aggregated transformations and gradient activations of classes are leveraged to design the attention mechanism to capture image representations. As a result, the system achieves 99.73 % accuracy, 99.82 % sensitivity, 99.63 % specificity in DR detection, 98.18 % accuracy, 97.41% sensitivity, and 99.55% specificity in DR grading. An analysis of objective performance metrics and model interpretation shows that the proposed model can detect DR more efficiently and grade the severity more accurately when using unseen fundus images than existing state-of-the-art methods. Therefore, this tool could help provide a second opinion to an ophthalmologist about the severity level of DR. [ABSTRACT FROM AUTHOR]

Objective Non-motor symptoms in Parkinson's disease (PD) involving cognition and emotion have been progressively receiving more attention in recent times. Electroencephalogram (EEG) signals, being an activity of central nervous system, can reflect the underlying true emotional state of a person. This paper presents a computational framework for classifying PD patients compared to healthy controls (HC) using emotional information from the brain's electrical activity. Approach Emotional EEG data were obtained from 20 PD patients and 20 healthy age-, gender- and education level-matched controls by inducing the six basic emotions of happiness, sadness, fear, anger, surprise and disgust using multimodal (audio and visual) stimuli. In addition, participants were asked to report their subjective affect. Because of the nonlinear and dynamic nature of EEG signals, we utilized higher order spectral features (specifically, bispectrum) for analysis. Two different classifiers namely K-Nearest Neighbor (KNN) and Support Vector Machine (SVM) were used to investigate the performance of the HOS based features to classify each of the six emotional states of PD patients compared to HC. Ten-fold cross-validation method was used for testing the reliability of the classifier results. Main results From the experimental results with our EEG data set, we found that (a) classification performance of bispectrum features across ALL frequency bands is better than individual frequency bands in both the groups using SVM classifier; (b) higher frequency band plays a more important role in emotion activities than lower frequency band; and (c) PD patients showed emotional impairments compared to HC, as demonstrated by a lower classification performance, particularly for negative emotions (sadness, fear, anger and disgust). Significance These results demonstrate the effectiveness of applying EEG features with machine learning techniques to classify the each emotional state difference of PD patients compared to HC, and offer a promising approach for detection of emotional impairments associated with other neurological disorders. [ABSTRACT FROM AUTHOR]

The evolution the novel corona virus disease (COVID-19) as a pandemic has inflicted several thousand deaths per day endangering the lives of millions of people across the globe. In addition to thermal scanning mechanisms, chest imaging examinations provide valuable insights to the detection of this virus, diagnosis and prognosis of the infections. Though Chest CT and Chest X-ray imaging are common in the clinical protocols of COVID-19 management, the latter is highly preferred, attributed to its simple image acquisition procedure and mobility of the imaging mechanism. However, Chest X-ray images are found to be less sensitive compared to Chest CT images in detecting infections in the early stages. In this paper, we propose a deep learning based framework to enhance the diagnostic values of these images for improved clinical outcomes. It is realized as a variant of the conventional SqueezeNet classifier with segmentation capabilities, which is trained with deep features extracted from the Chest X-ray images of a standard dataset for binary and multi class classification. The binary classifier achieves an accuracy of 99.53% in the discrimination of COVID-19 and Non COVID-19 images. Similarly, the multi class classifier performs classification of COVID-19, Viral Pneumonia and Normal cases with an accuracy of 99.79%. This model called the COVID-19 Super pixel SqueezNet (COVID-SSNet) performs super pixel segmentation of the activation maps to extract the regions of interest which carry perceptual image features and constructs an overlay of the Chest X-ray images with these regions. The proposed classifier model adds significant value to the Chest X-rays for an integral examination of the image features and the image regions influencing the classifier decisions to expedite the COVID-19 treatment regimen. • Innovative chest X-ray based decision support system. • Deep learning-based framework to enhance these images' diagnostic values. • Use of SqueezeNet for the detection process. • Use of Deep Transfer learning for the detection of COVID-19. [ABSTRACT FROM AUTHOR]