An efficient semi-supervised quality control system trained using physics-based MRI-artefact generators and adversarial training

Large medical imaging data sets are becoming increasingly available. A common challenge in these data sets is to ensure that each sample meets minimum quality requirements devoid of significant artefacts. Despite a wide range of existing automatic methods having been developed to identify imperfections and artefacts in medical imaging, they mostly rely on data-hungry methods. In particular, the scarcity of artefact-containing scans available for training has been a major obstacle in the development and implementation of machine learning in clinical research. To tackle this problem, we propose a novel framework having four main components: (1) a set of artefact generators inspired by magnetic resonance physics to corrupt brain MRI scans and augment a training dataset, (2) a set of abstract and engineered features to represent images compactly, (3) a feature selection process that depends on the class of artefact to improve classification performance, and (4) a set of Support Vector Machine (SVM) classifiers trained to identify artefacts. Our novel contributions are threefold: first, we use the novel physics-based artefact generators to generate synthetic brain MRI scans with controlled artefacts as a data augmentation technique. This will avoid the labour-intensive collection and labelling process of scans with rare artefacts. Second, we propose a large pool of abstract and engineered image features developed to identify 9 different artefacts for structural MRI. Finally, we use an artefact-based feature selection block that, for each class of artefacts, finds the set of features that provide the best classification performance. We performed validation experiments on a large data set of scans with artificially-generated artefacts, and in a multiple sclerosis clinical trial where real artefacts were identified by experts, showing that the proposed pipeline outperforms traditional methods. In particular, our data augmentation increases performance by up to 12.5 percentage points on the accuracy, F1, F2, precision and recall. At the same time, the computation cost of our pipeline remains low - less than a second to process a single scan - with the potential for real-time deployment. Our artefact simulators obtained using adversarial learning enable the training of a quality control system for brain MRI that otherwise would have required a much larger number of scans in both supervised and unsupervised settings. We believe that systems for quality control will enable a wide range of high-throughput clinical applications based on the use of automatic image-processing pipelines.

How do Antecedent Semantics Influence Pronoun Interpretation? Evidence from Eye Movements

Pronoun interpretation is often described as relying on a comprehender's mental model of discourse. For example, in some psycholinguistic accounts, interpreting pronouns involves a process of retrieval, whereby a pronoun is resolved by accessing information from its linguistic antecedent. However, linguistic antecedents are neither necessary nor sufficient for interpreting a pronoun, and even when an antecedent has been introduced in earlier discourse, there is little evidence for the retrieval of linguistic form. The current study extends our understanding of pronoun interpretation by examining whether the semantics of antecedent expressions are retrieved from representations of past discourse. Participants were instructed to move displayed objects in a Visual World eye-tracking task. In some cases, the semantics of the antecedent were no longer viable after an instruction was completed (e.g., "Move the house on the left to area 12," where the result was that a different house is now the leftmost one). In this case, retrieving antecedent semantics at the point of hearing a subsequent pronoun ("Now, move it…") should entail a processing penalty. Instead, the results showed that antecedent semantics have no direct effect on interpretation, raising additional questions about the role that retrieval might play in pronoun interpretation.

Real-time frequency-independent single-Lead and single-beat myocardial infarction detection

This study proposes a novel real-time frequency-independent myocardial infarction detector for Lead II electrocardiograms. The underlying Deep-LSTM network is trained using the PTB-XL database, the largest to date publicly available electrocardiography dataset, and is tested over the same and the older PTB database. By testing the model over distinct datasets, collected under different conditions and from different patients, a more realistic measure of the performance can be gauged from the deployed system. The detector is trained over 3589 myocardial infarction (MI) patients and 7115 healthy controls (HC) while it is evaluated on 1076 MIs and 1840 HCs. The proposed algorithm, achieved an accuracy of 77.12%, recall/sensitivity of 75.85%, and a specificity of 83.02% over the entire PTB database; 85.07%, 81.54%, 87.31% over the PTB-XL validation set (fold 9), and 84.17%, 78.37%, 87.55% over the PTB-XL test set (fold 10). The model also achieves stable performance metrics over the frequency range of 202 Hz to 2.8 kHz. The processing time is dependent on the sampling frequency, ranging from 130 ms at 202 Hz to 1.8 s at 2.8 kHz. Such outcome is within the time required for real-time processing (less than 300 ms for fast heartbeats), between 202 Hz and 500 Hz making the algorithm practically real-time. Therefore, the proposed MI detector could be readily deployed onto existing wearable and/or portable devices and test instruments; potentially having significant societal and clinical impact in the lives of patients at risk for myocardial infarction.

Fast occlusion-based point cloud exploration

Large-scale unstructured point cloud scenes can be quickly visualized without prior reconstruction by utilizing levels-of-detail structures to load an appropriate subset from out-of-core storage for rendering the current view. However, as soon as we need structures within the point cloud, e.g., for interactions between objects, the construction of state-of-the-art data structures requires O(NlogN) time for N points, which is not feasible in real time for millions of points that are possibly updated in each frame. Therefore, we propose to use a surface representation structure which trades off the (here negligible) disadvantage of single-frame use for both output-dominated and near-linear construction time in practice, exploiting the inherent 2D property of sampled surfaces in 3D. This structure tightly encompasses the assumed surface of unstructured points in a set of bounding depth intervals for each cell of a discrete 2D grid. The sorted depth samples in the structure permit fast surface queries, and on top of that an occlusion graph for the scene comes almost for free. This graph enables novel real-time user operations such as revealing partially occluded objects, or scrolling through layers of occluding objects, e.g., walls in a building. As an example application we showcase a 3D scene exploration framework that enables fast, more sophisticated interactions with point clouds rendered in real time.

SUPPLEMENTARY INFORMATION

The online version supplementary material available at 10.1007/s00371-021-02243-x.

Hardware architecture for real-time EEG-based functional brain connectivity parameter extraction

In this work, we proposed a novel architecture for real-time quantitative characterization of functional brain connectivity networks derived from Electroencephalogram (EEG). It consists of two main parts - calculation of Phase Lag Index (PLI) to form the functional connectivity networks and the extraction of a set of graph-theoretic parameters to quantitatively characterize these networks. The architecture was developed for a 19-channel EEG system. The system can calculate all the functional connectivity parameters in a total time of 131µs, utilizes 71% logic resources, and shows 51.84 mW dynamic power consumption at 22.16 MHz operation frequency when implemented in a Stratix IV EP4SGX230K FPGA. Our analysis also showed that the system occupies an area equivalent to approximately 937K 2-input NAND gates, with an estimated power consumption of 39.3 mW at 0.9 V supply using a 90 nm CMOS Application Specific Integrated Circuit (ASIC) technology.

What Are You Waiting For? Real-Time Integration of Cues for Fricatives Suggests Encapsulated Auditory Memory

Speech unfolds over time, and the cues for even a single phoneme are rarely available simultaneously. Consequently, to recognize a single phoneme, listeners must integrate material over several hundred milliseconds. Prior work contrasts two accounts: (a) a memory buffer account in which listeners accumulate auditory information in memory and only access higher level representations (i.e., lexical representations) when sufficient information has arrived; and (b) an immediate integration scheme in which lexical representations can be partially activated on the basis of early cues and then updated when more information arises. These studies have uniformly shown evidence for immediate integration for a variety of phonetic distinctions. We attempted to extend this to fricatives, a class of speech sounds which requires not only temporal integration of asynchronous cues (the frication, followed by the formant transitions 150-350 ms later), but also integration across different frequency bands and compensation for contextual factors like coarticulation. Eye movements in the visual world paradigm showed clear evidence for a memory buffer. Results were replicated in five experiments, ruling out methodological factors and tying the release of the buffer to the onset of the vowel. These findings support a general auditory account for speech by suggesting that the acoustic nature of particular speech sounds may have large effects on how they are processed. It also has major implications for theories of auditory and speech perception by raising the possibility of an encapsulated memory buffer in early auditory processing.

A neural network approach for real-time particle/cell characterization in microfluidic impedance cytometry

Microfluidic applications such as active particle sorting or selective enrichment require particle classification techniques that are capable of working in real time. In this paper, we explore the use of neural networks for fast label-free particle characterization during microfluidic impedance cytometry. A recurrent neural network is designed to process data from a novel impedance chip layout for enabling real-time multiparametric analysis of the measured impedance data streams. As demonstrated with both synthetic and experimental datasets, the trained network is able to characterize with good accuracy size, velocity, and cross-sectional position of beads, red blood cells, and yeasts, with a unitary prediction time of 0.4 ms. The proposed approach can be extended to other device designs and cell types for electrical parameter extraction. This combination of microfluidic impedance cytometry and machine learning can serve as a stepping stone to real-time single-cell analysis and sorting. Graphical Abstract.

Real-time algorithm for Poissonian noise reduction in low-dose fluoroscopy: performance evaluation

BACKGROUND

Quantum noise intrinsically limits the quality of fluoroscopic images. The lower is the X-ray dose the higher is the noise. Fluoroscopy video processing can enhance image quality and allows further patient's dose lowering. This study aims to assess the performances achieved by a Noise Variance Conditioned Average (NVCA) spatio-temporal filter for real-time denoising of fluoroscopic sequences. The filter is specifically designed for quantum noise suppression and edge preservation. It is an average filter that excludes neighborhood pixel values exceeding noise statistic limits, by means of a threshold which depends on the local noise standard deviation, to preserve the image spatial resolution. The performances were evaluated in terms of contrast-to-noise-ratio (CNR) increment, image blurring (full width of the half maximum of the line spread function) and computational time. The NVCA filter performances were compared to those achieved by simple moving average filters and the state-of-the-art video denoising block matching-4D (VBM4D) algorithm. The influence of the NVCA filter size and threshold on the final image quality was evaluated too.

RESULTS

For NVCA filter mask size of 5 × 5 × 5 pixels (the third dimension represents the temporal extent of the filter) and a threshold level equal to 2 times the local noise standard deviation, the NVCA filter achieved a 10% increase of the CNR with respect to the unfiltered sequence, while the VBM4D achieved a 14% increase. In the case of NVCA, the edge blurring did not depend on the speed of the moving objects; on the other hand, the spatial resolution worsened of about 2.2 times by doubling the objects speed with VBM4D. The NVCA mask size and the local noise-threshold level are critical for final image quality. The computational time of the NVCA filter was found to be just few percentages of that required for the VBM4D filter.

CONCLUSIONS

The NVCA filter obtained a better image quality compared to simple moving average filters, and a lower but comparable quality when compared with the VBM4D filter. The NVCA filter showed to preserve edge sharpness, in particular in the case of moving objects (performing even better than VBM4D). The simplicity of the NVCA filter and its low computational burden make this filter suitable for real-time video processing and its hardware implementation is ready to be included in future fluoroscopy devices, offering further lowering of patient's X-ray dose.

How do infants and adults process communicative events in real time?

Speech allows humans to communicate and to navigate the social world. By 12 months, infants recognize that speech elicits appropriate responses from others. However, it is unclear how infants process dynamic communicative scenes and how their processing abilities compare with those of adults. Do infants, like adults, process communicative events while the event is occurring or only after being presented with the outcome? We examined 12-month-olds' and adults' eye movements as they watched a Communicator grasp one (target) of two objects. During the test event, the Communicator could no longer reach the objects, so she spoke or coughed to a Listener, who selected either object. Infants' and adults' patterns of looking to the actors and objects revealed that both groups immediately evaluated the Communicator's speech, but not her cough, as communicative and recognized that the Listener should select the target object only when the Communicator spoke. Furthermore, infants and adults shifted their attention between the actors and the objects in very similar ways. This suggests that 12-month-olds can quickly process communicative events as they occur with adult-like accuracy. However, differences in looking reveal that 12-month-olds process slower than adults. This early developing processing ability may allow infants to learn language and acquire knowledge from communicative interactions.

Intelligent Perioperative System: Towards Real-time Big Data Analytics in Surgery Risk Assessment

Surgery risk assessment is an effective tool for physicians to manage the treatment of patients, but most current research projects fall short in providing a comprehensive platform to evaluate the patients' surgery risk in terms of different complications. The recent evolution of big data analysis techniques makes it possible to develop a real-time platform to dynamically analyze the surgery risk from large-scale patients information. In this paper, we propose the Intelligent Perioperative System (IPS), a real-time system that assesses the risk of postoperative complications (PC) and dynamically interacts with physicians to improve the predictive results. In order to process large volume patients data in real-time, we design the system by integrating several big data computing and storage frameworks with the high through-output streaming data processing components. We also implement a system prototype along with the visualization results to show the feasibility of system design.

Real-time radionuclide identification in γ-emitter mixtures based on spiking neural network

Portal radiation monitors dedicated to the prevention of illegal traffic of nuclear materials at international borders need to deliver as fast as possible a radionuclide identification of a potential radiological threat. Spectrometry techniques applied to identify the radionuclides contributing to γ-emitter mixtures are usually performed using off-line spectrum analysis. As an alternative to these usual methods, a real-time processing based on an artificial neural network and Bayes' rule is proposed for fast radionuclide identification. The validation of this real-time approach was carried out using γ-emitter spectra ((241)Am, (133)Ba, (207)Bi, (60)Co, (137)Cs) obtained with a high-efficiency well-type NaI(Tl). The first tests showed that the proposed algorithm enables a fast identification of each γ-emitting radionuclide using the information given by the whole spectrum. Based on an iterative process, the on-line analysis only needs low-statistics spectra without energy calibration to identify the nature of a radiological threat.

New real-time heartbeat detection method using the angle of a single-lead electrocardiogram

This study presents a new real-time heartbeat detection algorithm using the geometric angle between two consecutive samples of single-lead electrocardiogram (ECG) signals. The angle was adopted as a new index representing the slope of ECG signal. The method consists of three steps: elimination of high-frequency noise, calculation of the angle of ECG signal, and detection of R-waves using a simple adaptive thresholding technique. The MIT-BIH arrhythmia database, QT database, European ST-T database, T-wave alternans database and synthesized ECG signals were used to evaluate the performance of the proposed algorithm and compare with the results of other methods suggested in literature. The proposed method shows a high detection rate-99.95% of the sensitivity, 99.95% of the positive predictivity, and 0.10% of the fail detection rate on the four databases. The result shows that the proposed method can yield better or comparable performance than other literature despite the relatively simple process. The proposed algorithm needs only a single-lead ECG, and involves a simple and quick calculation. Moreover, it does not require post-processing to enhance the detection. Thus, it can be effectively applied to various real-time healthcare and medical devices.

Efficient sequential Bayesian inference method for real-time detection and sorting of overlapped neural spikes

Overlapping of extracellularly recorded neural spike waveforms causes the original spike waveforms to become hidden and merged, confounding the real-time detection and sorting of these spikes. Methods proposed for solving this problem include using a multi-trode or placing a restriction on the complexity of overlaps. In this paper, we propose a rapid sequential method for the robust detection and sorting of arbitrarily overlapped spikes recorded with arbitrary types of electrodes. In our method, the probabilities of possible spike trains, including those that are overlapping, are evaluated by sequential Bayesian inference based on probabilistic models of spike-train generation and extracellular voltage recording. To reduce the high computational cost inherent in an exhaustive evaluation, candidates with low probabilities are considered as impossible candidates and are abolished at each sampling time to limit the number of candidates in the next evaluation. In addition, the data from a few subsequent sampling times are considered and used to calculate the "look-ahead probability", resulting in improved calculation efficiency due to a more rapid elimination of candidates. These sufficiently reduce computational time to enable real-time calculation without impairing performance. We assessed the performance of our method using simulated neural signals and actual neural signals recorded in primary cortical neurons cultured on a multi-electrode array. Our results demonstrated that our computational method could be applied in real-time with a delay of less than 10 ms. The estimation accuracy was higher than that of a conventional spike sorting method, particularly for signals with multiple overlapping spikes.