Abstract:Functional ultrasound imaging (fUSI) is a cutting-edge technology that measures changes in cerebral blood volume (CBV) by detecting backscattered echoes from red blood cells moving within its field of view (FOV). It offers high spatiotemporal resolution and sensitivity, allowing for detailed visualization of cerebral blood flow dynamics. While fUSI has been utilized in preclinical drug development studies to explore the mechanisms of action of various drugs targeting the central nervous system, many of these studies have primarily focused on predetermined regions of interest (ROIs). This focus may overlook relevant brain activity outside these specific areas, which could influence the results. To address this limitation, we combined convolutional neural networks (CNNs) with fUSI to comprehensively understand the pharmacokinetic process of Dizocilpine, also known as MK-801, a drug that blocks the N-Methyl-D-aspartate (NMDA) receptor in the central nervous system. CNN and class activation mapping (CAM) revealed the spatiotemporal effects of MK-801, which originated in the cortex and propagated to the hippocampus, demonstrating the ability to detect dynamic drug effects over time. Additionally, CNN and CAM assessed the impact of anesthesia on the spatiotemporal hemodynamics of the brain, revealing no distinct patterns between early and late stages. The integration of fUSI and CNN provides a powerful tool to gain insights into the spatiotemporal dynamics of drug action in the brain. This combination enables a comprehensive and unbiased assessment of drug effects on brain function, potentially accelerating the development of new therapies in neuropharmacological studies.
Abstract:A common technique for ameliorating the computational costs of running large neural models is sparsification, or the removal of neural connections during training. Sparse models are capable of maintaining the high accuracy of state of the art models, while functioning at the cost of more parsimonious models. The structures which underlie sparse architectures are, however, poorly understood and not consistent between differently trained models and sparsification schemes. In this paper, we propose a new technique for sparsification of recurrent neural nets (RNNs), called moduli regularization, in combination with magnitude pruning. Moduli regularization leverages the dynamical system induced by the recurrent structure to induce a geometric relationship between neurons in the hidden state of the RNN. By making our regularizing term explicitly geometric, we provide the first, to our knowledge, a priori description of the desired sparse architecture of our neural net. We verify the effectiveness of our scheme for navigation and natural language processing RNNs. Navigation is a structurally geometric task, for which there are known moduli spaces, and we show that regularization can be used to reach 90% sparsity while maintaining model performance only when coefficients are chosen in accordance with a suitable moduli space. Natural language processing, however, has no known moduli space in which computations are performed. Nevertheless, we show that moduli regularization induces more stable recurrent neural nets with a variety of moduli regularizers, and achieves high fidelity models at 98% sparsity.