Aakash Kumar

Intravascular nanorobotic interventions in the neurovasculature are a promising yet challenging frontier in medical technology. This study addresses the challenge of precise control over nanorobots within the Middle Cerebral Artery (MCA), particularly under conditions of significant stenosis (80.1%) that complicate flow and navigation. Despite advances in magnetically actuated nanorobots for targeted therapies, achieving fine-grained control in complex vascular environments remains a critical issue. To address this, we propose a novel tri-coil electromagnetic system that uses three independent coils to generate dynamically modulated magnetic fields, enabling precise control of nanorobot motion. Coils 1 and 3 create evolving magnetic fields, while Coil 2 maintains a static field to shape the overall force application. A mathematical model was developed and implemented to optimize the system, demonstrating its ability to manipulate nanorobots within stenosed M2 MCA vessels. Our approach achieved high precision, allowing a lateral shift of the nanorobot trajectory with a magnetic field intensity of 40 mT. This tri-coil system offers a significant advancement in nanorobotic navigation and treatment of cerebrovascular diseases through minimally invasive techniques.

The swift advancement of convolutional neural networks (CNNs) in numerous real-world utilizations urges an elevation in computational cost along with the size of the model. In this context, many researchers steered their focus to eradicate these specific issues by compressing the original CNN models by pruning weights and filters, respectively. As filter pruning has an upper hand over the weight pruning method because filter pruning methods don’t impact sparse connectivity patterns. In this work, we suggested a Bayesian Convolutional Neural Network (BayesCNN) with Variational Inference, which prefaces probability distribution over weights. For the pruning task of Bayesian CNN, we utilized a combined version of L1-norm with capped L1-norm to help epitomize the amount of information that can be extracted through filter and control regularization. In this formation, we pruned unimportant filters directly without any test accuracy loss and achieved a slimmer model with comparative accuracy. The whole process of pruning is iterative and to validate the performance of our proposed work, we utilized several different CNN architectures on the standard classification dataset available. We have compared our results with non-Bayesian CNN models particularly, datasets such as CIFAR-10 on VGG-16, and pruned 75.8% parameters with float-point-operations (FLOPs) reduction of 51.3% without loss of accuracy and has achieved advancement in state-of-art.