Orthogonal tensor dictionary learning for accelerated dynamic MRI Abstract A direct application of the compressed sensing (CS) theory to dynamic magnetic resonance imaging (MRI) reconstruction needs vectorization or matricization of the dynamic MRI data, which is composed of a stack of 2D images and can be naturally regarded as a tensor. This 1D/2D model may destroy the inherent spatial structure property of the data. An alternative way to exploit the multidimensional structure in dynamic MRI is to employ tensor decomposition for dictionary learning, that is, learning multiple dictionaries along each dimension (mode) and sparsely representing the multidimensional data with respect to the Kronecker product of these dictionaries. In this work, we introduce a novel tensor dictionary learning method under an orthonormal constraint on the elementary matrix of the tensor dictionary for dynamic MRI reconstruction. The proposed algorithm alternates sparse coding, tensor dictionary learning, and updating reconstruction, and each corresponding subproblem is efficiently solved by a closed-form solution. Numerical experiments on phantom and synthetic data show significant improvements in reconstruction accuracy and computational efficiency obtained by the proposed scheme over the existing method that uses the 1D/2D model with overcomplete dictionary learning. Graphical abstract Fig. 1 Comparison between (a) the traditional method and (b) the proposed method based on dictionary learning for dynamic MRI reconstruction

Single channel surface electromyogram deconvolution to explore motor unit discharges Abstract Interference surface electromyogram (EMG) reflects many bioelectric properties of active motor units (MU), which are however difficult to estimate due to the asynchronous summation of their discharges. This paper introduces a deconvolution technique to estimate the cumulative firings of MUs. Tests in simulations show that the power spectral density of the estimated MU firings has a low-frequency peak corresponding to the mean firing rate of MUs in the detection volume of the recording system, weighted by the amplitudes of MU action potentials. The peak increases in amplitude and its centroid shifts to a higher frequency when MU synchronization is simulated (mainly due to the shift of discharges of large MUs). The peak is found even at high force levels, when such a contribution does not emerge from the EMG. This result is also confirmed in preliminary applications to experimental data. Moreover, the simulated cumulative firings of MUs are estimated with a correlation above 90% (considering frequency contributions up to 150 Hz), for all force levels. The method requires a single EMG channel, thus being feasible even in applied studies using simple recording systems. It may open many potential applications, e.g., in the study of the modulation of MU firing rate induced by either fatigue or pathology and in coherency analysis. Graphical Abstract Examples of application of the deconvolution (Deconv) algorithm and comparison with the cumulative firings and the cumulated weighted firings (CWF, i.e., each firing pattern is weighted by the root mean squared amplitude of the corresponding MU action potential). Portions of data are shown on the left, the power spectral densities (PSD) on the right (Welch method applied to 3 s of data, sub-epochs of 0.5 s, mean value removed from each of them, 50% of overlap). A) Simulated signal (50% of maximal voluntary contraction, MVC) with random MU firings. B) Simulated signal (50% MVC) with a level of synchronization equal to 10%. C) Experimental data from vastus medialis at 40% MVC (data decomposed by the algorithm of Holobar and Zazula, IEEE Trans. Sig. Proc. 2007; PSD of the cumulated firings almost identical to that of CWF, as few MUs were identified).

An iterative damped least-squares algorithm for simultaneously monitoring the development of hemorrhagic and secondary ischemic lesions in brain injuries Abstract Electrical impedance tomography (EIT) is a non-invasive and real-time imaging method that has the potential to be used for monitoring intracerebral hemorrhage (ICH). Recent studies have proposed that ischemia secondary to ICH occurs simultaneously in the brain. Real-time monitoring of the development of hemorrhage and risk of secondary ischemia is crucial for clinical intervention. However, few studies have explored the performance of EIT monitoring in cases where hemorrhage and secondary ischemia exist. When these lesions get close to each other, or their conductivity and volume changes differ greatly, it becomes challenging for dynamic EIT algorithms to simultaneously reconstruct subtle injuries. To address this, an iterative damped least-squares (IDLS) algorithm is proposed in this study. The quality of the IDLS algorithm was assessed using blur radius and temporal response during computer simulation and a phantom 3D head-shaped model where bidirectional disturbance targets were simulated. The results showed that the IDLS algorithm enhanced contrast and concurrently reconstructed bidirectional disturbance targets in images. Moreover, it showed superior performance in decreasing the blur radius and was time cost-effective. With further improvement, the IDLS algorithm has the potential to be used for monitoring the development of hemorrhage and risk of ischemia secondary to ICH. Graphical abstract (a) and (b) are simulation images of bidirectional disturbance targets with different change ratios of volume (Vr) and conductivity (σr) based on the damped least-squares (DLS) algorithm and iterative damped least-squared (IDLS) algorithm, respectively. (c) shows the performance metrics of blur radius and temporal response with different volume ratio (corresponding to Vr). (d) shows the performance metrics of blur radius and temporal response with different conductivity change percentage (corresponding to σr).

A deep learning algorithm for one-step contour aware nuclei segmentation of histopathology images Abstract This paper addresses the task of nuclei segmentation in high-resolution histopathology images. We propose an automatic end-to-end deep neural network algorithm for segmentation of individual nuclei. A nucleus-boundary model is introduced to predict nuclei and their boundaries simultaneously using a fully convolutional neural network. Given a color-normalized image, the model directly outputs an estimated nuclei map and a boundary map. A simple, fast, and parameter-free post-processing procedure is performed on the estimated nuclei map to produce the final segmented nuclei. An overlapped patch extraction and assembling method is also designed for seamless prediction of nuclei in large whole-slide images. We also show the effectiveness of data augmentation methods for nuclei segmentation task. Our experiments showed our method outperforms prior state-of-the-art methods. Moreover, it is efficient that one 1000×1000 image can be segmented in less than 5 s. This makes it possible to precisely segment the whole-slide image in acceptable time. The source code is available at https://github.com/easycui/nuclei_segmentation. Graphical Abstract The neural network for nuclei segmentation

Optimized needle shape reconstruction using experimentally based strain sensors positioning Abstract Needles are tools that are used daily during minimally invasive procedures. During the insertions, needles may be affected by deformations which may threaten the success of the procedure. To tackle this problem, needles with embedded strain sensors have been developed and associated with navigation systems. The localization of the needle in the tissues is then obtained in real time by reconstruction from the strain measurements, allowing the physician to optimize its gesture. As the number of strain sensors embedded is limited in number, their positions on the needle have a great impact on the accuracy of the shape reconstruction. The main contribution of this paper is a novel strain sensor positioning method to improve the reconstruction accuracy. A notable feature of our method is the use of experimental needle insertion data, which increases the relevancy of the resulting sensor optimal locations. To the best of the author’s knowledge, no experimentally based needle sensor positioning method has been presented yet. Reconstruction validations from clinical data show that the localization accuracy of the needle tip is improved by almost 40% with optimal locations compared with equidistant locations when reconstructing with two sensor triplets or more. Graphical Abstract Improvement of the reconstruction accuracy of a deformed needle shape by using experimental data to position strain sensors

Bypassing the volume conduction effect by multilayer neural network for effective connectivity estimation Abstract Differentiation of real interactions between different brain regions from spurious ones has been a challenge in neuroimaging researches. While using electroencephalographic data, those spurious interactions are mostly caused by the volume conduction (VC) effect between the recording sites. In this study, we address the problem by jointly modeling the causal relationships among brain regions and the mixing effects of volume conduction. The VC effect is formulated with a time-invariant linear equation, and the causal relationships between the brain regions are modeled with a nonlinear multivariate autoregressive process. These two models are simultaneously implemented by a multilayer neural network. The internal hidden layers represent the interactions among the regions, while the external layers are devoted for the relationship between the source activities and observed EEG measurements at the scalp. The causal interactions are estimated by the causality coefficient measure, which is based on the information (weights and parameters) embedded in the network. The proposed method is verified using various simulated data. It is then applied to the real EEG signals collected from a memory retrieval test. The results showed that the method is able to eliminate the volume conduction interferences and consequently leads to higher accuracy in identification of true causal interactions. Graphical abstract The proposed network structure used to simultaneously model the volume conduction and source interactions. By this special structure, we first move from the sensor space to the source space at the first layer. Then, within internal hidden layers, the interactions between the sources are represented in the form of a general (nonlinear) multivariate autoregressive (nMVAR) model. Finally, we return from the source space to the sensor space at the last layer of the network. The proposed method bypasses the effect of volume conduction and causes more accurate connectivity estimation.

Evaluation of divided attention using different stimulation models in event-related potentials Abstract Divided attention is defined as focusing on different tasks at once, and this is described as one of the biggest problems of today’s society. Default examinations for understanding attention are questionnaires or physiological signals, like evoked potentials and electroencephalography. Physiological records were obtained using visual, auditory, and auditory-visual stimuli combinations with 48 participants—18-25-year-old university students—to find differences between sustained and divided attention. A Fourier-based filter was used to get a 0.01–30-Hz frequency band. Fractal dimensions, entropy values, power spectral densities, and Hjorth parameters from electroencephalography and P300 components from evoked potentials were calculated as features. To decrease the size of the feature set, some features, which yield less detail level for data, were eliminated. The visual and auditory stimuli in selective attention were compared with the divided attention state, and the best accuracy was found to be 88.89% on a support vector machine with linear kernel. As a result, it was seen that divided attention could be more difficult to determine from selective attention, but successful classification could be obtained with appropriate methods. Contrary to literature, the study deals with the infrastructure of attention types by working on a completely healthy and attention-high group. Graphical abstract

Hybrid position/force output feedback second-order sliding mode control for a prototype of an active orthosis used in back-assisted mobilization Abstract This article shows the design of a robust second-order sliding mode controller to solve the trajectory tracking problem of an active orthosis for assisting back physiotherapies. The orthosis was designed in agreement with morphological dimensions and its articulations distribution followed the same designing rules. The orthosis has six articulated arms attached to an articulated column. The orthosis was fully instrumented with actuators and position sensors at each articulation. The controller implemented a class of hybrid/position controller depending on the relative force exerted by the patient and the orthosis movement. The position information provided by each articulation was supplied to a distributed super-twisting differentiator to recover the corresponding angular velocity. A set of twisting controllers was implemented to regulate the position of the robot in agreement to predefined reference trajectories. Reference trajectories were obtained from a biomechanical-based analysis. The hybrid tracking control problem solved the automation of the assisted therapy to the patient, including the force feedback. The performance of the orthosis was tested with different dummy bodies with different resistance. The robust output feedback controller successfully tracked the reference trajectories despite the material of the dummy used during the testing. The orthosis was evaluated with two volunteers using a simple reference trajectory. Graphical Abstract General structure of the active back assisted orthosis

Estimation of microvascular perfusion after esophagectomy: a quantitative model of dynamic fluorescence imaging Abstract Most common complications of esophagectomy stem from a perfusion deficiency of the gastric conduit at the anastomosis. Fluorescent tracer imaging allows intraoperative visualization of tissue perfusion. Quantitative assessment of fluorescence dynamics has the potential to identify perfusion deficiency. We developed a perfusion model to analyze the relation between fluorescence dynamics and perfusion deficiency. The model divides the gastric conduit into two well-perfused and two anastomosed sites. Hemodynamics and tracer transport were modeled. We analyzed the value of relative time-to-threshold (RTT) as a predictor of the relative remaining flow (RRF). Intensity thresholds for RTT of 20% to 50% of the maximum fluorescence intensity of the well-perfused site were tested. The relation between RTT and RRF at the anastomosed sites was evaluated over large variations of vascular conductance and volume. The ability of RTT to distinguish between sufficient and impaired perfusion was analyzed using c-statistics. We found that RTT was a valuable estimate for low RRF. The threshold of 20% of the maximum fluorescence intensity provided the best prediction of impaired perfusion on the two anastomosed sites (AUC = 0.89 and 0.86). The presented model showed that for low flows, relative time-to-threshold may be used to estimate perfusion deficiency.

Leveraging network analysis to support experts in their analyses of subjects with MCI and AD Abstract In this paper, we propose a network analysis–based approach to help experts in their analyses of subjects with mild cognitive impairment (hereafter, MCI) and Alzheimer’s disease (hereafter, AD) and to investigate the evolution of these subjects over time. The inputs of our approach are the electroencephalograms (hereafter, EEGs) of the patients to analyze, performed at a certain time and, again, 3 months later. Given an EEG of a subject, our approach constructs a network with nodes that represent the electrodes and edges that denote connections between electrodes. Then, it applies several network-based techniques allowing the investigation of subjects with MCI and AD and the analysis of their evolution over time. (i) A connection coefficient, supporting experts to distinguish patients with MCI from patients with AD; (ii) A conversion coefficient, supporting experts to verify if a subject with MCI is converting to AD; (iii) Some network motifs, i.e., network patterns very frequent in one kind of patient and absent, or very rare, in the other. Patients with AD, just by the very nature of their condition, cannot be forced to stay motionless while undergoing examinations for a long time. EEG is a non-invasive examination that can be easily done on them. Since AD and MCI, if prodromal to AD, are associated with a loss of cortical connections, the adoption of network analysis appears suitable to investigate the effects of the progression of the disease on EEG. This paper confirms the suitability of this idea Graphical Abstract Ability of our proposed model to distinguish a control subject from a patient with MCI and a patient with AD. Blue edges represent strong connections among the corresponding brain areas; red edges denote middle connections, whereas green edges indicate weak connections. In the control subject (at the top), most connections are blue. In the patient with MCI (at the middle), most connections are red and green. In the patient with AD (at the bottom), most connections are either absent or green.  .