Search for: All records

This study emulates associative learning in rodents by using a neuromorphic robot navigating an open-field arena. The goal is to investigate how biologically inspired neural models can reproduce animal-like learning behaviors in real-world robotic systems. We constructed a neuromorphic robot by deploying computational models of spatial and sensory neurons onto a mobile platform. Different coding schemes—rate coding for vibration signals and population coding for visual signals—were implemented. The associative learning model employs 19 spiking neurons and follows Hebbian plasticity principles to associate visual cues with favorable or unfavorable locations. Our robot successfully replicated classical rodent associative learning behavior by memorizing causal relationships between environmental cues and spatial outcomes. The robot’s self-learning capability emerged from repeated exposure and synaptic weight adaptation, without the need for labeled training data. Experiments confirmed functional learning behavior across multiple trials. This work provides a novel embodied platform for memory and learning research beyond traditional animal models. By embedding biologically inspired learning mechanisms into a real robot, we demonstrate how spatial memory can be formed and expressed through sensorimotor interactions. The model’s compact structure (19 neurons) illustrates a minimal yet functional learning network, and the study outlines principles for synaptic weight and threshold design, guiding future development of more complex neuromorphic systems. 

Parkinson's Disease (PD) impacts millions globally, causing debilitating motor symptoms. While Closed-Loop Deep Brain Stimulation (CL-DBS) has emerged as a promising treatment, existing systems often suffer from high energy consumption, making them impractical for wearable or implantable devices. This research introduces an innovative neuromorphic approach to enhance CL-DBS performance, utilizing Leaky Integrate-and-Fire (LIF) neuron-based controllers to adaptively modulate stimulation signals based on symptom severity. Two controllers, the on-off LIF and dual LIF models, are proposed, achieving significant reductions in power consumption by 19% and 56%, respectively, while enhancing suppression efficiency by 4.7% and 6.77%. Additionally, this work addresses the scarcity of datasets for PD symptoms by developing a novel dataset featuring neural activity from the subthalamic nucleus (STN), incorporating beta oscillations as key physiological biomarkers. This dataset aims to support further advancements in neuromorphic CL-DBS systems and is openly shared with the research community. By combining energy-efficient neuromorphic controllers with a comprehensive dataset, this study not only advances the technological feasibility of CL-DBS systems for PD treatment but also provides a foundation for personalized and adaptive neuromodulation therapies, paving the way for improved quality of life for individuals with Parkinson's Disease. 

Deep neural networks (DNNs) have achieved remarkable success in various cognitive tasks through training on extensive labeled datasets. However, the heavy reliance on these datasets poses challenges for DNNs in scenarios with energy constraints in particular scenarios, such as on the moon. On the contrary, animals exhibit a self-learning capability by interacting with their surroundings and memorizing concurrent events without annotated data—a process known as associative learning. A classic example of associative learning is when a rat memorizes desired and undesired stimuli while exploring a T-maze. The successful implementation of associative learning aims to replicate the self-learning mechanisms observed in animals, addressing challenges in data-constrained environments. While current implementations of associative learning are predominantly small scale and offline, this work pioneers associative learning in a robot equipped with a neuromorphic chip, specifically for online learning in a T-maze. The system successfully replicates classic associative learning observed in rodents, using neuromorphic robots as substitutes for rodents. The neuromorphic robot autonomously learns the cause-and-effect relationship between audio and visual stimuli. 

Abstract PurposeTo introduce a novel deep model‐based architecture (DMBA), SPICER, that uses pairs of noisy and undersampled k‐space measurements of the same object to jointly train a model for MRI reconstruction and automatic coil sensitivity estimation. MethodsSPICER consists of two modules to simultaneously reconstructs accurate MR images and estimates high‐quality coil sensitivity maps (CSMs). The first module, CSM estimation module, uses a convolutional neural network (CNN) to estimate CSMs from the raw measurements. The second module, DMBA‐based MRI reconstruction module, forms reconstructed images from the input measurements and the estimated CSMs using both the physical measurement model and learned CNN prior. With the benefit of our self‐supervised learning strategy, SPICER can be efficiently trained without any fully sampled reference data. ResultsWe validate SPICER on both open‐access datasets and experimentally collected data, showing that it can achieve state‐of‐the‐art performance in highly accelerated data acquisition settings (up to ). Our results also highlight the importance of different modules of SPICER—including the DMBA, the CSM estimation, and the SPICER training loss—on the final performance of the method. Moreover, SPICER can estimate better CSMs than pre‐estimation methods especially when the ACS data is limited. ConclusionDespite being trained on noisy undersampled data, SPICER can reconstruct high‐quality images and CSMs in highly undersampled settings, which outperforms other self‐supervised learning methods and matches the performance of the well‐known E2E‐VarNet trained on fully sampled ground‐truth data. 

Deformable image registration (DIR) is an active research topic in biomedical imaging. There is a growing interest in developing DIR methods based on deep learning (DL). A traditional DL approach to DIR is based on training a convolutional neural network (CNN) to estimate the registration field between two input images. While conceptually simple, this approach comes with a limitation that it exclusively relies on a pre-trained CNN without explicitly enforcing fidelity between the registered image and the reference. We present plug-and-play image registration network (PIRATE) as a new DIR method that addresses this issue by integrating an explicit data-fidelity penalty and a CNN prior. PIRATE pre-trains a CNN denoiser on the registration field and "plugs" it into an iterative method as a regularizer. We additionally present PIRATE+ that fine-tunes the CNN prior in PIRATE using deep equilibrium models (DEQ). PIRATE+ interprets the fixed-point iteration of PIRATE as a network with effectively infinite layers and then trains the resulting network end-to-end, enabling it to learn more task-specific information and boosting its performance. Our numerical results on OASIS and CANDI datasets show that our methods achieve state-of-the-art performance on DIR. 

Spiking Neural Networks (SNNs) are energy-efficient artificial neural network models that can carry out data-intensive applications. Energy consumption, latency, and memory bottleneck are some of the major issues that arise in machine learning applications due to their data-demanding nature. Memristor-enabled Computing-In-Memory (CIM) architectures have been able to tackle the memory wall issue, eliminating the energy and time-consuming movement of data. In this work we develop a scalable CIM-based SNN architecture with our fabricated two-layer memristor crossbar array. In addition to having an enhanced heat dissipation capability, our memristor exhibits substantial enhancement of 10% to 66% in design area, power and latency compared to state-of-the-art memristors. This design incorporates an inter-spike interval (ISI) encoding scheme due to its high information density to convert the incoming input signals into spikes. Furthermore, we include a time-to-first-spike (TTFS) based output processing stage for its energy-efficiency to carry out the final classification. With the combination of ISI, CIM and TTFS, this network has a competitive inference speed of 2μs/image and can successfully classify handwritten digits with 2.9mW of power and 2.51pJ energy per spike. The proposed architecture with the ISI encoding scheme can achieve ∼10% higher accuracy than those of other encoding schemes in the MNIST dataset. 

IntroductionParkinson’s disease (PD) is a neurodegenerative disorder affecting millions of patients. Closed-Loop Deep Brain Stimulation (CL-DBS) is a therapy that can alleviate the symptoms of PD. The CL-DBS system consists of an electrode sending electrical stimulation signals to a specific region of the brain and a battery-powered stimulator implanted in the chest. The electrical stimuli in CL-DBS systems need to be adjusted in real-time in accordance with the state of PD symptoms. Therefore, fast and precise monitoring of PD symptoms is a critical function for CL-DBS systems. However, the current CL-DBS techniques suffer from high computational demands for real-time PD symptom monitoring, which are not feasible for implanted and wearable medical devices. MethodsIn this paper, we present an energy-efficient neuromorphic PD symptom detector using memristive three-dimensional integrated circuits (3D-ICs). The excessive oscillation at beta frequencies (13–35 Hz) at the subthalamic nucleus (STN) is used as a biomarker of PD symptoms. ResultsSimulation results demonstrate that our neuromorphic PD detector, implemented with an 8-layer spiking Long Short-Term Memory (S-LSTM), excels in recognizing PD symptoms, achieving a training accuracy of 99.74% and a validation accuracy of 99.52% for a 75%–25% data split. Furthermore, we evaluated the improvement of our neuromorphic CL-DBS detector using NeuroSIM. The chip area, latency, energy, and power consumption of our CL-DBS detector were reduced by 47.4%, 66.63%, 65.6%, and 67.5%, respectively, for monolithic 3D-ICs. Similarly, for heterogeneous 3D-ICs, employing memristive synapses to replace traditional Static Random Access Memory (SRAM) resulted in reductions of 44.8%, 64.75%, 65.28%, and 67.7% in chip area, latency, and power usage. DiscussionThis study introduces a novel approach for PD symptom evaluation by directly utilizing spiking signals from neural activities in the time domain. This method significantly reduces the time and energy required for signal conversion compared to traditional frequency domain approaches. The study pioneers the use of neuromorphic computing and memristors in designing CL-DBS systems, surpassing SRAM-based designs in chip design area, latency, and energy efficiency. Lastly, the proposed neuromorphic PD detector demonstrates high resilience to timing variations in brain neural signals, as confirmed by robustness analysis.