1,721,058 research outputs found
Hyperdimensional Computing with Local Binary Patterns: One-shot Learning for Seizure Onset Detection and Identification of Ictogenic Brain Regions from Short-time iEEG Recordings.
Full text linkOBJECTIVE
We develop a fast learning algorithm combining symbolic dynamics and brain-inspired hyperdimensional computing for both seizure onset detection and identification of ictogenic (seizure generating) brain regions from intracranial electroencephalography (iEEG).
METHODS
Our algorithm first transforms iEEG time series from each electrode into symbolic local binary pattern codes from which a holographic distributed representation of the brain state of interest is constructed across all the electrodes and over time in a hyperdimensional space. The representation is used to quickly learn from few seizures, detect their onset, and identify the spatial brain regions that generated them.
RESULTS
We assess our algorithm on our dataset that contains 99 short-time iEEG recordings from 16 drug-resistant epilepsy patients being implanted with 36 to 100 electrodes. For the majority of the patients (10 out of 16), our algorithm quickly learns from one or two seizures and perfectly (100%) generalizes on novel seizures using k-fold cross-validation. For the remaining six patients, the algorithm requires three to six seizures for learning. Our algorithm surpasses the state-of-the-art including deep learning algorithms by achieving higher specificity (94.84% vs. 94.77%) and macroaveraging accuracy (95.42% vs. 94.96%), and 74x lower memory footprint, but slightly higher average latency in detection (15.9 s vs. 14.7 s). Moreover, the algorithm can reliably identify (with a p-value < 0.01) the relevant electrodes covering an ictogenic brain region at two levels of granularity: cerebral hemispheres and lobes.
CONCLUSION AND SIGNIFICANCE
Our algorithm provides: (1) a unified method for both learning and classification tasks with end-to-end binary operations; (2) one-shot learning from seizure examples; (3) linear computational scalability for increasing number of electrodes; (4) generation of transparent codes that enables post-translational supports for clinical decision making
Laelaps: An Energy-Efficient Seizure Detection Algorithm from Long-term Human iEEG Recordings without False Alarms
Full text linkWe propose Laelaps, an energy-efficient and fast learning algorithm with no false alarms for epileptic seizure detection from long-term intracranial electroencephalography (iEEG) signals. Laelaps uses end-to-end binary operations by exploiting symbolic dynamics and brain-inspired hyperdimensional computing. Laelaps's results surpass those yielded by state-of-the-art (SoA) methods [1], [2], [3], including deep learning, on a new very large dataset containing 116 seizures of 18 drug-resistant epilepsy patients in 2656 hours of recordings each patient implanted with 24 to 128 iEEG electrodes. Laelaps trains 18 patient-specific models by using only 24 seizures: 12 models are trained with one seizure per patient, the others with two seizures. The trained models detect 79 out of 92 unseen seizures without any false alarms across all the patients as a big step forward in practical seizure detection. Importantly, a simple implementation of Laelaps on the Nvidia Tegra X2 embedded device achieves 1.7x 3.9 x faster execution and 1.4 x 2.9x lower energy consumption compared to the best result from the SoA methods. Our source code and anonymized iEEG dataset are freely available at http://ieeg-swez.ethz.ch
PLiNIO: A User-Friendly Library of Gradient-based Methods for Complexity-aware DNN Optimization
Full text linkAccurate yet efficient Deep Neural Networks (DNNs) are in high demand, especially for applications that require their execution on constrained edge devices. Finding such DNNs in a reasonable time for new applications requires automated optimization pipelines since the huge space of hyper-parameter combinations is impossible to explore extensively by hand. In this work, we propose PLiNIO, an open-source library implementing a comprehensive set of state-of-the-art DNN design automation techniques, all based on lightweight gradient-based optimization, under a unified and user-friendly interface. With experiments on several edge-relevant tasks, we show that combining the various optimizations available in PLiNIO leads to rich sets of solutions that Pareto-dominate the considered baselines in terms of accuracy vs model size. Noteworthy, PLiNIO achieves up to 94.34% memory reduction for a <1% accuracy drop compared to a baseline architecture
One-shot Learning for iEEG Seizure Detection Using End-to-end Binary Operations: Local Binary Patterns with Hyperdimensional Computing
Full text linkThis paper presents an efficient binarized algorithm for both learning and classification of human epileptic seizures from intracranial electroencephalography (iEEG). The algorithm combines local binary patterns with brain-inspired hyperdimensional computing to enable end-to-end learning and inference with binary operations. The algorithm first transforms iEEG time series from each electrode into local binary pattern codes. Then atomic high-dimensional binary vectors are used to construct composite representations of seizures across all electrodes. For the majority of our patients (10 out of 16), the algorithm quickly learns from one or two seizures (i.e., one-/few-shot learning) and perfectly generalizes on 27 further seizures. For other patients, the algorithm requires three to six seizures for learning. Overall, our algorithm surpasses the state-of-the-art methods [1] for detecting 65 novel seizures with higher specificity and sensitivity, and lower memory footprint
Enhancing Structural Health Monitoring with Vehicle Identification and Tracking
Full text linkTraffic load monitoring and structural health monitoring (SHM) have been gaining increasing attention over the last decade. However, most of the current installations treat the two monitoring types as separated problems, thereby using dedicated installed sensors, such as smart cameras for traffic load or accelerometers for Structural Health Monitoring (SHM). This paper presents a new framework aimed at leveraging the data collected by a SHM system for a second use, namely, monitoring vehicles passing on the structure being monitored (a viaduct). Our framework first processes the raw three-axial acceleration signals through a series of transformations and extracts its energy. Then, an anomaly detection algorithm is used to detect peaks from 90 installed sensors, and a linear regression together with a simple threshold filters out false detection by estimating the speed of the vehicles. Initial results in conditions of moderate traffic load are promising, demonstrating the detection of vehicles and realistic characterization of their speed. Moreover, a k−means clustering analysis distinguishes two groups of peaks with statistically different features such as amplitude and damping duration that could be likely associated with heavy vehicles and cars, respectively
Profiling Inflammatory Extracellular Vesicles in Plasma and Cerebrospinal Fluid: An Optimized Diagnostic Model for Parkinson's Disease
Full text linkExtracellular vesicles (EVs) play a central role in intercellular communication, which is relevant for inflammatory and immune processes implicated in neurodegenerative disorders, such as Parkinson's Disease (PD). We characterized and compared distinctive cerebrospinal fluid (CSF)-derived EVs in PD and atypical parkinsonisms (AP), aiming to integrate a diagnostic model based on immune profiling of plasma-derived EVs via artificial intelligence. Plasma- and CSF-derived EVs were isolated from patients with PD, multiple system atrophy (MSA), AP with tauopathies (AP-Tau), and healthy controls. Expression levels of 37 EV surface markers were measured by a flow cytometric bead-based platform and a diagnostic model based on expression of EV surface markers was built by supervised learning algorithms. The PD group showed higher amount of CSF-derived EVs than other groups. Among the 17 EV surface markers differentially expressed in plasma, eight were expressed also in CSF of a subgroup of PD, 10 in MSA, and 6 in AP-Tau. A two-level random forest model was built using EV markers co-expressed in plasma and CSF. The model discriminated PD from non-PD patients with high sensitivity (96.6%) and accuracy (92.6%). EV surface marker characterization bolsters the relevance of inflammation in PD and it underscores the role of EVs as pathways/biomarkers for protein aggregation-related neurodegenerative diseases
ExaMon-X: a Predictive Maintenance Framework for Automatic Monitoring in Industrial IoT Systems
Full text linkIn recent years, the Industrial Internet of Things (IIoT) has led to significant steps forward in many industries, thanks to the exploitation of several technologies, ranging from Big Data processing to Artificial Intelligence (AI). Among the various IIoT scenarios, large-scale data centers can reap significant benefits from adopting Big Data analytics and AI-boosted approaches since these technologies can allow effective predictive maintenance. However, most of the off-the-shelf currently available solutions are not ideally suited to the HPC context, e.g., they do not sufficiently take into account the very heterogeneous data sources and the privacy issues which hinder the adoption of the cloud solution, or they do not fully
exploit the computing capabilities available in loco in a supercomputing facility. In this paper, we tackle this issue, and we propose an IIoT holistic and vertical framework for predictive maintenance in supercomputers. The framework is based on a big lightweight data monitoring infrastructure, specialized databases suited for heterogeneous data, and a set of high-level AI-based functionalities tailored to HPC actors’ specific needs. We present the deployment and assess the usage of this framework in several in-production HPC systems
Accelerating Depthwise Separable Convolutions on Ultra-Low-Power Devices
No full textDepthwise separable convolutions are a fundamental component in efficient Deep Neural Networks, as they reduce the number of parameters and operations compared to traditional convolutions while maintaining comparable accuracy. However, their low data reuse opportunities make deploying them notoriously difficult. In this work, we perform an extensive exploration of alternatives to fuse the depthwise and pointwise kernels that constitute the separable convolutional block. Our approach aims to minimize time-consuming memory transfers by combining different data layouts. When targeting a commercial ultra-low-power device with a three-level memory hierarchy, the GreenWaves GAP8 SoC, we reduce the latency of end-to-end network execution by up to 11.40%. Furthermore, our kernels reduce activation data movements between L2 and L1 memories by up to 52.97%
Improving PPG-based Heart-Rate Monitoring with Synthetically Generated Data
No full textImproving the quality of heart-rate monitoring is the basis for a full-time assessment of people’s daily care. Recent state-of-the-art heart-rate monitoring algorithms exploit PPG and inertial data to efficiently estimate subjects’ beats-per-minute (BPM) directly on wearable devices. Despite the easy-recording of these signals (e.g., through commercial smartwatches), which makes this approach appealing, new challenges are arising. The first problem is fitting these algorithms into low-power memory-constrained MCUs. Further, the PPG signal usually has a low signal-to-noise ratio due to the presence of motion artifacts (MAs) arising from movements of subjects’ arms. In this work, we propose using synthetically generated data to improve the accuracy of PPG-based heart-rate tracking using deep neural networks without increasing the algorithm’s complexity. Using the TEMPONet network as baseline, we show that the HR tracking Mean Absolute Error (MAE) can be reduced from 5.28 to 4.86 BPM on PPGDalia dataset. Noteworthy, to do so, we only increase the training time, keeping the inference step unchanged. Consequently, the new and more accurate network can still fit the small memory of the GAP8 MCU, occupying 429 KB when quantized to 8bits
Q-PPG: Energy-Efficient PPG-Based Heart Rate Monitoring on Wearable Devices
Full text linkHearth Rate (HR) monitoring is increasingly performed in wrist-worn devices using low-cost photoplethysmography (PPG) sensors. However, Motion Artifacts (MAs) caused by movements of the subject's arm affect the performance of PPG-based HR tracking. This is typically addressed coupling the PPG signal with acceleration measurements from an inertial sensor. Unfortunately, most standard approaches of this kind rely on hand-tuned parameters, which impair their generalization capabilities and their applicability to real data in the field. In contrast, methods based on deep learning, despite their better generalization, are considered to be too complex to deploy on wearable devices.In this work, we tackle these limitations, proposing a design space exploration methodology to automatically generate a rich family of deep Temporal Convolutional Networks (TCNs) for HR monitoring, all derived from a single "seed" model. Our flow involves a cascade of two Neural Architecture Search (NAS) tools and a hardware-friendly quantizer, whose combination yields both highly accurate and extremely lightweight models. When tested on the PPG-Dalia dataset, our most accurate model sets a new state-of-the-art in Mean Absolute Error. Furthermore, we deploy our TCNs on an embedded platform featuring a STM32WB55 microcontroller, demonstrating their suitability for real-time execution. Our most accurate quantized network achieves 4.41 Beats Per Minute (BPM) of Mean Absolute Error (MAE), with an energy consumption of 47.65 mJ and a memory footprint of 412 kB. At the same time, the smallest network that obtains a MAE < 8 BPM, among those generated by our flow, has a memory footprint of 1.9 kB and consumes just 1.79 mJ per inference
- …
