TIME- AND FREQUENCY-BASED INDEPENDENT EVALUATION OF QRST CANCELLATION TECHNIQUES FOR SINGLE-LEAD ELECTROCARDIOGRAMS DURING ATRIAL FIBRILLATION.

With the increased prevalence of atrial fibrillation (AF) - a rhythm disturbance in heart's top chambers - there is growing interest in accurate non-invasive diagnosis of atrial activity to improve its therapy. A key component in non-invasive analysis of atrial activity is a successful removal of the ventricular QRST complexes from electrocardiograms (ECGs). In this study, we have developed a new approach for an objective and physiologically-based evaluation of QRST cancellation methods based on comparisons with the power spectra of the AF. Three commonly used QRST cancellation methods were evaluated; namely, average beat subtraction, singular value cancellation, and principal component analysis. These methods were evaluated in time and frequency domains using a set of synthesized ECGs preserving the atrial-specific temporal and spectral properties. It was observed that the ABS method provided the best estimation when QRST morphological variability is low, while PCA produces an overall best estimate when a large QRST morphological variability is present.

Fractal, recurrent, and dense U-Net architectures with EfficientNet encoder for medical image segmentation.

Purpose: U-Net is a deep learning technique that has made significant contributions to medical image segmentation. Although the accomplishments of deep learning algorithms in terms of image processing are evident, many challenges still need to be overcome to achieve human-like performance. One of the main challenges in building deeper U-Nets is black-box problems, such as vanishing gradients. Overcoming this problem allows us to develop neural networks with advanced network designs. Approach: We propose three U-Net variants, namely efficient R2U-Net, efficient dense U-Net, and efficient fractal U-Net, that can create highly accurate segmentation maps. The first part of our contribution makes use of EfficientNet to distribute resources in the network efficiently. The second part of our work applies the following layer connections to design the U-Net decoders: residual connections, dense connections, and fractal expansion. We apply EfficientNet as the encoder to our three decoders to design three conceivable models. Results: The aforementioned three proposed deep learning models were tested on four benchmark datasets, including the CHASE DB1 and digital retinal images for vessel extraction (DRIVE) retinal image databases and the ISIC 2018 and HAM10000 dermoscopy image databases. We obtained the highest Dice coefficient of 0.8013, 0.8808, 0.8019, and 0.9295 for CHASE DB1, ISIC 2018, DRIVE, and HAM10000, respectively, and a Jaccard (JAC) score of 0.6686, 0.7870, 0.6694, and 0.8683 for CHASE DB1, ISIC 2018, DRIVE, and HAM10000, respectively. Statistical analysis revealed that the proposed deep learning models achieved better segmentation results compared with the state-of-the-art models. Conclusions: U-Net is quite an adaptable deep learning framework and can be integrated with other deep learning techniques. The use of recurrent feedback connections, dense convolution, residual skip connections, and fractal convolutional expansions allow for the design of improved deeper U-Net models. With the addition of EfficientNet, we can now leverage the performance of an optimally scaled classifier for U-Net encoders.

Effects of fibrosis morphology on reentrant ventricular tachycardia inducibility and simulation fidelity in patient-derived models.

Myocardial fibrosis detected via delayed-enhanced magnetic resonance imaging (MRI) has been shown to be a strong indicator for ventricular tachycardia (VT) inducibility. However, little is known regarding how inducibility is affected by the details of the fibrosis extent, morphology, and border zone configuration. The objective of this article is to systematically study the arrhythmogenic effects of fibrosis geometry and extent, specifically on VT inducibility and maintenance. We present a set of methods for constructing patient-specific computational models of human ventricles using in vivo MRI data for patients suffering from hypertension, hypercholesterolemia, and chronic myocardial infarction. Additional synthesized models with morphologically varied extents of fibrosis and gray zone (GZ) distribution were derived to study the alterations in the arrhythmia induction and reentry patterns. Detailed electrophysiological simulations demonstrated that (1) VT morphology was highly dependent on the extent of fibrosis, which acts as a structural substrate, (2) reentry tended to be anchored to the fibrosis edges and showed transmural conduction of activations through narrow channels formed within fibrosis, and (3) increasing the extent of GZ within fibrosis tended to destabilize the structural reentry sites and aggravate the VT as compared to fibrotic regions of the same size and shape but with lower or no GZ. The approach and findings represent a significant step toward patient-specific cardiac modeling as a reliable tool for VT prediction and management of the patient. Sensitivities to approximation nuances in the modeling of structural pathology by image-based reconstruction techniques are also implicated.

Fast, accurate, and fully automatic segmentation of the right ventricle in short-axis cardiac MRI.

This paper presents a fully automatic method to segment the right ventricle (RV) from short-axis cardiac MRI. A combination of a novel window-constrained accumulator thresholding technique, binary difference of Gaussian (DoG) filters, optimal thresholding, and morphology are utilized to drive the segmentation. A priori segmentation window constraints are incorporated to guide and refine the process, as well as to ensure appropriate area confinement of the segmentation. Training and testing were performed using a combined 48 patient datasets supplied by the organizers of the MICCAI 2012 right ventricle segmentation challenge, allowing for unbiased evaluations and benchmark comparisons. Marked improvements in speed and accuracy over the top existing methods are demonstrated.

Accurate reconstruction of 3D cardiac geometry from coarsely-sliced MRI.

We present a comprehensive validation analysis to assess the geometric impact of using coarsely-sliced short-axis images to reconstruct patient-specific cardiac geometry. The methods utilize high-resolution diffusion tensor MRI (DTMRI) datasets as reference geometries from which synthesized coarsely-sliced datasets simulating in vivo MRI were produced. 3D models are reconstructed from the coarse data using variational implicit surfaces through a commonly used modeling tool, CardioViz3D. The resulting geometries were then compared to the reference DTMRI models from which they were derived to analyze how well the synthesized geometries approximate the reference anatomy. Averaged over seven hearts, 95% spatial overlap, less than 3% volume variability, and normal-to-surface distance of 0.32 mm was observed between the synthesized myocardial geometries reconstructed from 8 mm sliced images and the reference data. The results provide strong supportive evidence to validate the hypothesis that coarsely-sliced MRI may be used to accurately reconstruct geometric ventricular models. Furthermore, the use of DTMRI for validation of in vivo MRI presents a novel benchmark procedure for studies which aim to substantiate their modeling and simulation methods using coarsely-sliced cardiac data. In addition, the paper outlines a suggested original procedure for deriving image-based ventricular models using the CardioViz3D software.

Long-term frequency gradients during persistent atrial fibrillation in sheep are associated with stable sources in the left atrium.

BACKGROUND: Dominant frequencies (DFs) of activation are higher in the atria of patients with persistent than paroxysmal atrial fibrillation (AF), and left atrial (LA)-to-right atrial (RA) DF gradients have been identified in both. However, whether such gradients are maintained as long-term persistent AF is established remains unexplored. We aimed at determining in vivo the time course in atrial DF values from paroxysmal to persistent AF in sheep and testing the hypothesis that an LA-to-RA DF difference is associated with LA drivers in persistent AF. METHODS AND RESULTS: AF was induced using RA tachypacing (n=8). Electrograms were obtained weekly from an RA lead and an implantable loop recorder implanted near the LA. DFs were determined for 5-second-long electrograms (QRST subtracted) during AF in vivo and in ex vivo optical mapping. Underlying structural changes were compared with weight-matched controls (n=4). After the first AF episode, DF increased gradually during a 2-week period (7±0.21 to 9.92±0.31 Hz; n=6; P<0.05). During 9 to 24 weeks of AF, the DF values on the implantable loop recorder were higher than the RA (10.6±0.08 versus 9.3±0.1 Hz, respectively; n=7; P<0.0001). Subsequent optical mapping confirmed a DF gradient from posterior LA-to-RA (9.1±1.0 to 6.9±0.9 Hz; P<0.05) and demonstrated patterns of activation compatible with drifting rotors in the posterior LA. Persistent AF sheep showed significant enlargement of the posterior LA compared with controls. CONCLUSIONS: In the sheep, transition from paroxysmal to persistent AF shows continuous LA-to-RA DF gradients in vivo together with enlargement of the posterior LA, which harbors the highest frequency domains and patterns of activation compatible with drifting rotors.

An enhanced MEMS error modeling approach based on Nu-Support Vector Regression.

Micro Electro Mechanical System (MEMS)-based inertial sensors have made possible the development of a civilian land vehicle navigation system by offering a low-cost solution. However, the accurate modeling of the MEMS sensor errors is one of the most challenging tasks in the design of low-cost navigation systems. These sensors exhibit significant errors like biases, drift, noises; which are negligible for higher grade units. Different conventional techniques utilizing the Gauss Markov model and neural network method have been previously utilized to model the errors. However, Gauss Markov model works unsatisfactorily in the case of MEMS units due to the presence of high inherent sensor errors. On the other hand, modeling the random drift utilizing Neural Network (NN) is time consuming, thereby affecting its real-time implementation. We overcome these existing drawbacks by developing an enhanced Support Vector Machine (SVM) based error model. Unlike NN, SVMs do not suffer from local minimisation or over-fitting problems and delivers a reliable global solution. Experimental results proved that the proposed SVM approach reduced the noise standard deviation by 10-35% for gyroscopes and 61-76% for accelerometers. Further, positional error drifts under static conditions improved by 41% and 80% in comparison to NN and GM approaches.

Identification of CpG islands in DNA sequences using statistically optimal null filters.

: CpG dinucleotide clusters also referred to as CpG islands (CGIs) are usually located in the promoter regions of genes in a deoxyribonucleic acid (DNA) sequence. CGIs play a crucial role in gene expression and cell differentiation, as such, they are normally used as gene markers. The earlier CGI identification methods used the rich CpG dinucleotide content in CGIs, as a characteristic measure to identify the locations of CGIs. The fact, that the probability of nucleotide G following nucleotide C in a CGI is greater as compared to a non-CGI, is employed by some of the recent methods. These methods use the difference in transition probabilities between subsequent nucleotides to distinguish between a CGI from a non-CGI. These transition probabilities vary with the data being analyzed and several of them have been reported in the literature sometimes leading to contradictory results. In this article, we propose a new and efficient scheme for identification of CGIs using statistically optimal null filters. We formulate a new CGI identification characteristic to reliably and efficiently identify CGIs in a given DNA sequence which is devoid of any ambiguities. Our proposed scheme combines maximum signal-to-noise ratio and least squares optimization criteria to estimate the CGI identification characteristic in the DNA sequence. The proposed scheme is tested on a number of DNA sequences taken from human chromosomes 21 and 22, and proved to be highly reliable as well as efficient in identifying the CGIs.

A Systolic Array-Based FPGA Parallel Architecture for the BLAST Algorithm.

A design of systolic array-based Field Programmable Gate Array (FPGA) parallel architecture for Basic Local Alignment Search Tool (BLAST) Algorithm is proposed. BLAST is a heuristic biological sequence alignment algorithm which has been used by bioinformatics experts. In contrast to other designs that detect at most one hit in one-clock-cycle, our design applies a Multiple Hits Detection Module which is a pipelining systolic array to search multiple hits in a single-clock-cycle. Further, we designed a Hits Combination Block which combines overlapping hits from systolic array into one hit. These implementations completed the first and second step of BLAST architecture and achieved significant speedup comparing with previously published architectures.

Identification of CpG islands in DNA sequences using matched filters.

CpG islands (CGIs), rich in CG dinucleotides, are usually located in the promoter regions of genes in DNA sequences and are used as gene markers. Identification of CGIs plays an important role in the analysis of DNA sequences. In this paper, we propose a new digital signal processing (DSP) based approach using matched filters for the identification of CGIs. We also formulate a new/reliable CGI identification characteristic replacing the several existing probability transition tables for CGIs and non-CGIs. The peaks in matched filter output, obtained by correlating the CGI characteristic with the DNA sequence to be analyzed, accurately and reliably identify CGIs. This approach is tested on a number of DNA sequences and is proved to be capable of identifying CpG islands efficiently and reliably.

Microarray image segmentation using Chan-Vese active contour model and level set method.

Microarray technology is considered to be one of the major breakthroughs in bioinformatics for profiling gene-expressions of thousands of genes, simultaneously. Analysis of a microarray image plays an important role in the accurate depiction of gene-expression. Segmentation, the process of separating the foreground from the background, of a microarray image, is one of the key issues in microarray image analysis. Level sets have tremendous potential in the segmentation of images. In this paper, a new approach for segmentation of the microarray images is proposed. In this work, Chan-Vese approximation of the Mumford-Shah model and the level set method are employed for image segmentation. Illustrative examples of the proposed method are presented highlighting its effectiveness.

A new approach to analysis and modeling of esophageal manometry data in humans.

In this paper, we propose a new approach to the analysis and modeling of esophageal manometry (EGM) data to assist the diagnosis of esophageal motility disorders in humans. The proposed approach combines three techniques, namely, wavelet decomposition (WD), nonlinear pulse detection technique (NPDT), and statistical pulse modeling. Specifically, WD is applied to the filtering of the EGM data, which is contaminated with electrocardiography (ECG) artifacts. A new NPDT is applied to the denoised data leading to identification and extraction of diagnostically important information, i.e., esophageal pulses from the respiration artifacts. Such information is used to generate a statistical model that can classify the EGM patterns. The proposed approach is computationally effortless, thus making it suitable for real-time application. Experimental results using measured EGM data of 20 patients, including ten abnormal cases is presented. Comparison of our results with those from existing techniques illustrates the advantages of the proposed approach in terms of accuracy and efficiency.

A two-stage neural network based technique for protein secondary structure prediction.

Protein secondary structure prediction is one of the most important research areas in bioinformatics. In this paper, we propose a two-stage protein secondary structure prediction technique, implemented using neural network models. The first neural network stage of the proposed technique associates the input protein sequence to a bin containing its corresponding homologues. The second stage predicts the secondary structure of the input sequence utilizing a neural prediction model specific to the bin obtained from stage one. The strategy of binning allows for simplified and accurate neural models. This technique is implemented on the RS126 dataset and its prediction accuracy is compared with that of the standard PHD approach.