RESUMEN
Epilepsy is a globally distributed chronic neurological disorder that may pose a threat to life without warning. Therefore, the use of wearable devices for real-time detection and treatment of epilepsy is crucial. Additionally, personalizing disease detection algorithms for individual users is also a challenge in clinical applications. Some studies have proposed seizure detection algorithms with convolutional neural networks (CNNs) and programmable hardware architectures for speeding up the process of CNN inference. However, personalizing seizure detection algorithms could still not be performed on these hardware architectures. Consequently, this study proposes three key contributions to address the challenges: a real-time seizure detection and personalization algorithm, a programmable reduced instruction set computer-V (RISC-V) deep learning accelerator (DLA) hardware architecture (RVDLAHA), and a dedicated RISC-V DLA (RVDLA) compiler. In animal experiments with lab rats, the proposed CNN-based seizure detection algorithm obtains an accuracy of 99.5% for a 32-bit floating point and an accuracy of 99.3% for a 16-bit fixed point. Additionally, the proposed personalization algorithm increases the testing accuracy across different databases from 85.0% to 92.9%. The RVDLAHA is implemented on Xilinx PYNQ-Z2, with a power consumption of only 0.107 W at an operating frequency of 1 MHz. Each step, including raw data input, preprocessing, detection, and personalization, requires only 17.8, 1.0, 1.1, and 1.3 ms, respectively. With the hardware architecture, the seizure detection and personalization algorithm can provide on-device real-time monitoring.
RESUMEN
This work proposes a classification system for arrhythmias, aiming to enhance the efficiency of the diagnostic process for cardiologists. The proposed algorithm includes a naive preprocessing procedure for electrocardiography (ECG) data applicable to various ECG databases. Additionally, this work proposes an ultralightweight model for arrhythmia classification based on a convolutional neural network and incorporating R-peak interval features to represent long-term rhythm information, thereby improving the model's classification performance. The proposed model is trained and tested by using the MIT-BIH and NCKU-CBIC databases in accordance with the classification standards of the Association for the Advancement of Medical Instrumentation (AAMI), achieving high accuracies of 98.32% and 97.1%. This work applies the arrhythmia classification algorithm to a web-based system, thus providing a graphical interface. The cloud-based execution of automated artificial intelligence (AI) classification allows cardiologists and patients to view ECG wave conditions instantly, thereby remarkably enhancing the quality of medical examination. This work also designs a customized integrated circuit for the hardware implementation of an AI accelerator. The accelerator utilizes a parallelized processing element array architecture to perform convolution and fully connected layer operations. It introduces proposed hybrid stationary techniques, combining input and weight stationary modes to increase data reuse drastically and reduce hardware execution cycles and power consumption, ultimately achieving high-performance computing. This accelerator is implemented in the form of a chip by using the TSMC 180 nm CMOS process. It exhibits a power consumption of 122 µW, a classification latency of 6.8 ms, and an energy efficiency of 0.83 µJ/classification.
RESUMEN
This study proposes a charge-mode neural stimulator for electrical stimulation systems that utilizes a capacitor-reuse technique with a residual charge detector and achieves active charge balancing simultaneously. The design is mainly used for epilepsy suppression systems to achieve real-time symptom relief during seizures. A charge-mode stimulator is adopted in consideration of the complexity of circuit design, the high voltage tolerance of transistors, and system integration requirements in the future. The residual charge detector allows users to understand the current stimulus situation, enabling them to make optimal adjustments to the stimulation parameters. On the basis of the information on actual stimulation charge, active charge balancing can effectively prevent the accumulation of mismatched charges on electrode impedance. The capacitor- and phase-reuse techniques help realize high integration of the overall stimulator circuit in consideration of the commonality of the use of a capacitor and charging/discharging phase in the stimulation circuit and charge detector. The proposed charge-mode neural stimulator is implemented in a TSMC 0.18 µm 1P6M CMOS process with a core area of 0.2127 mm2. Measurement results demonstrate the accuracy of the stimulation's functionality and the programmable stimulus parameters. The effectiveness of the proposed charge-mode neural stimulator for epileptic seizure suppression is verified through animal experiments.
RESUMEN
Noninvasive brain stimulation (NIBS) techniques, including transcranial direct current stimulation (tDCS) and transcranial random noise stimulation (tRNS), are emerging as promising tools for enhancing cognitive functions by modulating brain activity and enhancing cognitive functions. Despite their potential, the specific and combined effects of tDCS and tRNS on brain functions, especially regarding functional connectivity, cortical inhibition, and memory performance, are not well-understood. This study aims to explore the distinct and combined impacts of tDCS and tRNS on these neural and cognitive parameters. Using a within-subject design, ten participants underwent four stimulation conditions: sham, tDCS, tRNS, and combined tDCS + tRNS. We assessed the impact on resting-state functional connectivity, cortical inhibition via Cortical Silent Period (CSP), and visuospatial memory performance using the Corsi Block-tapping Test (CBT). Our results indicate that while tDCS appears to induce brain lateralization, tRNS has more generalized and dispersive effects. Interestingly, the combined application of tDCS and tRNS did not amplify these effects but rather suggested a non-synergistic interaction, possibly due to divergent mechanistic pathways, as observed across fMRI, CSP, and CBT measures. These findings illuminate the complex interplay between tDCS and tRNS, highlighting their non-additive effects when used concurrently and underscoring the necessity for further research to optimize their application for cognitive enhancement.