The artificial intelligence (AI) system has achieved expert-level performance in electrocardiogram (ECG) signal analysis. However, in underdeveloped countries or regions where the healthcare information system is imperfect, only paper ECGs can be provided. Analysis of real-world ECG images (photos or scans of paper ECGs) remains challenging due to complex environments or interference. In this study, we present an AI system developed to detect and screen cardiac abnormalities (CAs) from real-world ECG images. The system was evaluated on a large dataset of 52,357 patients from multiple regions and populations across the world. On the detection task, the AI system obtained area under the receiver operating curve (AUC) of 0.996 (hold-out test), 0.994 (external test 1), 0.984 (external test 2), and 0.979 (external test 3), respectively. Meanwhile, the detection results of AI system showed a strong correlation with the diagnosis of cardiologists (cardiologist 1 (R=0.794, p<1e-3), cardiologist 2 (R=0.812, p<1e-3)). On the screening task, the AI system achieved AUCs of 0.894 (hold-out test) and 0.850 (external test). The screening performance of the AI system was better than that of the cardiologists (AI system (0.846) vs. cardiologist 1 (0.520) vs. cardiologist 2 (0.480)). Our study demonstrates the feasibility of an accurate, objective, easy-to-use, fast, and low-cost AI system for CA detection and screening. The system has the potential to be used by healthcare professionals, caregivers, and general users to assess CAs based on real-world ECG images.
Current neuroprosthetics rely on stable, high quality recordings from chronically implanted microelectrode arrays (MEAs) in neural tissue. While chronic electrophysiological recordings and electrode failure modes have been reported from rodent and larger non-human primate (NHP) models, chronic recordings from the marmoset model have not been previously described. The common marmoset is a New World primate that is easier to breed and handle compared to larger NHPs and has a similarly organized brain, making it a potentially useful smaller NHP model for neuroscience studies. This study reports recording stability and signal quality of MEAs chronically implanted in behaving marmosets. Six adult male marmosets, trained for reaching tasks, were implanted with either a 16-channel tungsten microwire array (five animals) or a Pt-Ir floating MEA (one animal) in the hand-arm region of the primary motor cortex (M1) and another MEA in the striatum targeting the nucleus accumbens (NAcc). Signal stability and quality was quantified as a function of array yield (active electrodes that recorded action potentials), neuronal yield (isolated single units during a recording session), and signal-to-noise ratio (SNR). Out of 11 implanted MEAs, nine provided functional recordings for at least three months, with two arrays functional for 10 months. In general, implants had high yield, which remained stable for up to several months. However, mechanical failure attributed to MEA connector was the most common failure mode. In the longest implants, signal degradation occurred, which was characterized by gradual decline in array yield, reduced number of isolated single units, and changes in waveform shape of action potentials. This work demonstrates the feasibility of longterm recordings from MEAs implanted in cortical and deep brain structures in the marmoset model. The ability to chronically record cortical signals for neural prosthetics applications in the common marmoset extends the potential of this model in neural interface research.
Here we demonstrate how a marmoset monkey can use a reinforcement learning (RL) Brain-Machine Interface (BMI) to effectively control the movements of a robot arm for a reaching task. In this work, an actor-critic RL algorithm used neural ensemble activity in the monkey's motor cortext to control the robot movements during a two-target decision task. This novel approach to decoding offers unique advantages for BMI control applications. Compared to supervised learning decoding methods, the actor-critic RL algorithm does not require an explicit set of training data to create a static control model, but rather it incrementally adapts the model parameters according to its current performance, in this case requiring only a very basic feedback signal. We show how this algorithm achieved high performance when mapping the monkey's neural states (94%) to robot actions, and only needed to experience a few trials before obtaining accurate real-time control of the robot arm. Since RL methods responsively adapt and adjust their parameters, they can provide a method to create BMIs that are robust against perturbations caused by changes in either the neural input space or the output actions they generate under different task requirements or goals.
New reinforcement based paradigms for building adaptive decoders for Brain-Machine Interfaces involve using feedback directly from the brain. In this work, we investigated neuromodulation in the Nucleus Accumbens (reward center) during a multi-target reaching task and investigated how to extract a reinforcing or non-reinforcing signal that could be used to adapt a BMI decoder. One of the challenges in brain-driven adaptation is how to translate biological neuromodulation into a single binary signal from the distributed representation of the neural population, which may encode many aspects of reward. To extract these signals, feature analysis and clustering were used to identify timing and coding properties of a user's neuromodulation related to reward perception. First, Principal Component Analysis (PCA) of reward related neural signals was used to extract variance in the firing and the optimum time correlation between the neural signal and the reward phase of the task. Next, k-means clustering was used to separate data into two classes.
Objective.The common marmoset has been increasingly used in neural interfacing studies due to its smaller size, easier handling, and faster breeding compared to Old World non-human primate (NHP) species. While assessment of cortical anatomy in marmosets has shown strikingly similar layout to macaques, comprehensive assessment of electrophysiological properties underlying forelimb reaching movements in this bridge species does not exist. The objective of this study is to characterize electrophysiological properties of signals recorded from the marmoset primary motor cortex (M1) during a reach task and compare with larger NHP models such that this smaller NHP model can be used in behavioral neural interfacing studies.Approach and main results.Neuronal firing rates and local field potentials (LFPs) were chronically recorded from M1 in three adult, male marmosets. Firing rates, mu + beta and high gamma frequency bands of LFPs were evaluated for modulation with respect to movement. Firing rate and regularity of neurons of the marmoset M1 were similar to that reported in macaques with a subset of neurons showing selectivity to movement direction. Movement phases (rest vs move) was classified from both neural spiking and LFPs. Microelectrode arrays provide the ability to sample small regions of the motor cortex to drive brain-machine interfaces (BMIs). The results demonstrate that marmosets are a robust bridge species for behavioral neuroscience studies with motor cortical electrophysiological signals recorded from microelectrode arrays that are similar to Old World NHPs.Significance. As marmosets represent an interesting step between rodent and macaque models, successful demonstration that neuron modulation in marmoset motor cortex is analogous to reports in macaques illustrates the utility of marmosets as a viable species for BMI studies.
In clinics, doctors rely on electrocardiograms (ECGs) to assess severe cardiac disorders. Owing to the development of technology and the increase in health awareness, ECG signals are currently obtained by using medical and commercial devices. Deep neural networks (DNNs) can be used to analyze these signals because of their high accuracy rate. However, researchers have found that adversarial attacks can significantly reduce the accuracy of DNNs. Studies have been conducted to defend ECG-based DNNs against traditional adversarial attacks, such as projected gradient descent (PGD), and smooth adversarial perturbation (SAP) which targets ECG classification; however, to the best of our knowledge, no study has completely explored the defense against adversarial attacks targeting ECG classification. Thus, we did different experiments to explore the effects of defense methods against white-box adversarial attack and black-box adversarial attack targeting ECG classification, and we found that some common defense methods performed well against these attacks. Besides, we proposed a new defense method called Adversarial Distillation Training (ADT) which comes from defensive distillation and can effectively improve the generalization performance of DNNs. The results show that our method performed more effectively against adversarial attacks targeting on ECG classification than the other baseline methods, namely, adversarial training, defensive distillation, Jacob regularization, and noise-to-signal ratio regularization. Furthermore, we found that our method performed better against PGD attacks with low noise levels, which means that our method has stronger robustness.
Brain-machine interface (BMI) systems give users direct neural control of robotic, communication, or functional electrical stimulation systems. As BMI systems begin transitioning from laboratory settings into activities of daily living, an important goal is to develop neural decoding algorithms that can be calibrated with a minimal burden on the user, provide stable control for long periods of time, and can be responsive to fluctuations in the decoder's neural input space (e.g. neurons appearing or being lost amongst electrode recordings). These are significant challenges for static neural decoding algorithms that assume stationary input/output relationships. Here we use an actor-critic reinforcement learning architecture to provide an adaptive BMI controller that can successfully adapt to dramatic neural reorganizations, can maintain its performance over long time periods, and which does not require the user to produce specific kinetic or kinematic activities to calibrate the BMI. Two marmoset monkeys used the Reinforcement Learning BMI (RLBMI) to successfully control a robotic arm during a two-target reaching task. The RLBMI was initialized using random initial conditions, and it quickly learned to control the robot from brain states using only a binary evaluative feedback regarding whether previously chosen robot actions were good or bad. The RLBMI was able to maintain control over the system throughout sessions spanning multiple weeks. Furthermore, the RLBMI was able to quickly adapt and maintain control of the robot despite dramatic perturbations to the neural inputs, including a series of tests in which the neuron input space was deliberately halved or doubled.
Atrial Fibrillation (AF) is a common cardiac arrhythmia. Many AF patients experience complications such as stroke and other cardiovascular issues. Early detection of AF is crucial. Existing algorithms can only distinguish ``AF rhythm in AF patients'' from ``sinus rhythm in normal individuals'' . However, AF patients do not always exhibit AF rhythm, posing a challenge for diagnosis when the AF rhythm is absent. To address this, this paper proposes a novel artificial intelligence (AI) algorithm to distinguish ``sinus rhythm in AF patients'' and ``sinus rhythm in normal individuals'' in beat-level. We introduce beat-level risk interpreters, trend risk interpreters, addressing the interpretability issues of deep learning models and the difficulty in explaining AF risk trends. Additionally, the beat-level information fusion decision is presented to enhance model accuracy. The experimental results demonstrate that the average AUC for single beats used as testing data from CPSC 2021 dataset is 0.7314. By employing 150 beats for information fusion decision algorithm, the average AUC can reach 0.7591. Compared to previous segment-level algorithms, we utilized beats as input, reducing data dimensionality and making the model more lightweight, facilitating deployment on portable medical devices. Furthermore, we draw new and interesting findings through average beat analysis and subgroup analysis, considering varying risk levels.
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)
