Intelligent Deep Learning Enabled Human Activity Recognition for Improved Medical Services

Human Activity Recognition(HAR)has been made simple in recent years,thanks to recent advancements made in Artificial Intelligence(AI)techni-ques.These techniques are applied in several areas like security,surveillance,healthcare,human-robot interaction,and entertainment.Since wearable sensor-based HAR system includes in-built sensors,human activities can be categorized based on sensor values.Further,it can also be employed in other applications such as gait diagnosis,observation of children/adult’s cognitive nature,stroke-patient hospital direction,Epilepsy and Parkinson’s disease examination,etc.Recently-developed Artificial Intelligence(AI)techniques,especially Deep Learning(DL)models can be deployed to accomplish effective outcomes on HAR process.With this motivation,the current research paper focuses on designing Intelligent Hyperparameter Tuned Deep Learning-based HAR(IHPTDL-HAR)technique in healthcare environment.The proposed IHPTDL-HAR technique aims at recogniz-ing the human actions in healthcare environment and helps the patients in mana-ging their healthcare service.In addition,the presented model makes use of Hierarchical Clustering(HC)-based outlier detection technique to remove the out-liers.IHPTDL-HAR technique incorporates DL-based Deep Belief Network(DBN)model to recognize the activities of users.Moreover,Harris Hawks Opti-mization(HHO)algorithm is used for hyperparameter tuning of DBN model.Finally,a comprehensive experimental analysis was conducted upon benchmark dataset and the results were examined under different aspects.The experimental results demonstrate that the proposed IHPTDL-HAR technique is a superior per-former compared to other recent techniques under different measures.

An effective nonlinear dynamic formulation to analyze grasping capability of soft pneumatic robotic gripper

Soft pneumatic robotic grippers have found extensive applica-tions across various engineering domains,which prompts active research due to their splendid compliance,high flex-ibility,and safe human-robot interaction over conventional stiff counterparts.Previously simplified rod-based models prin-cipally focused on clarifying overall large deformation and bending postures of soft grippers from static or quasi-static perspectives,whereas it is challenging to elaborate grasping characteristics of soft grippers without considering contact interaction and nonlinear large deformation behaviors.To address this,based on absolute nodal coordinate formulation(ANCF),comprehensively allowing for structural complexity,geometric,material and boundary nonlinearities,and incorpor-ating Coulomb’friction law with a multiple-point contact method,we put forward an effective nonlinear dynamic mod-eling approach for exploring grasping capability of soft grip-per.Moreover,we solved the established dynamic equations using Generalized-αscheme,and conducted thorough numer-ical simulation analysis on a three-jaw soft pneumatic gripper(SPG)in terms of grasping configurations,displacements and contact forces.The proposed dynamic approach can accurately both describe complicated deformed configurations along with stress distribution and provide a feasible solution to simulate grasping targets,whose effectiveness and precision were analyzed theoretically and verified experimentally,which may shed new light on devising and optimizing other multi-functional SPGs.

DeepRice: A deep learning and deep feature based classification of Rice leaf disease subtypes

Rice stands as a crucial staple food globally,with its enduring sustainability hinging on the prompt detection of rice leaf diseases.Hence,efficiently detecting diseases when they have already occurred holds paramount importance for solving the cost of manual visual identification and chemical testing.In the recent past,the identification of leaf pathologies in crops predominantly relies on manual methods using specialized equipment,which proves to be time-consuming and inefficient.This study offers a remedy by harnessing Deep Learning(DL)and transfer learning techniques to accurately identify and classify rice leaf diseases.A comprehensive dataset comprising 5932 self-generated images of rice leaves was assembled along with the benchmark datasets,categorized into 9 classes irrespective of the extent of disease spread across the leaves.These classes encompass diverse states including healthy leaves,mild and severe blight,mild and severe tungro,mild and severe blast,as well as mild and severe brown spot.Following meticulous manual labelling and dataset segmentation,which was validated by horticulture experts,data augmentation strategies were implemented to amplify the number of images.The datasets were subjected to evaluation using the proposed tailored Convolutional Neural Networks models.Their performance are scrutinized in conjunction with alternative transfer learning approaches like VGG16,Xception,ResNet50,DenseNet121,Inception ResnetV2,and Inception V3.The effectiveness of the proposed custom VGG16 model was gauged by its capacity to generalize to unseen images,yielding an exceptional accuracy of 99.94%,surpassing the benchmarks set by existing state-of-the-art models.Further,the layer wise feature extraction is also visualized as the interpretable AI.

Optimal Center-of-Mass Pivot Effect of Spinal Motion in Feline Galloping

In this paper,we studied the acceleration behavior of a quadruped animal during a galloping motion.Because the development of many quadruped robotic systems has been focused on dynamic movements,it is obvious that guidance from the dynamic behavior of quadruped animals is needed for robotics engineers.To fulfill this demand,this paper deals with analysis of the galloping motions of a domestic cat,which is well known for its excellent acceleration performance among four-legged animals.Based on the planar motion capture environment,the movement data of a galloping feline was acquired and the dynamic motions were estimated using a spring-mass system.In particular,the effects of the position and angle of the center-of-mass of the cat,angular displacement of the spine,and angular velocity of the spine were analyzed and are discussed below.Through this process,it was possible to understand the dynamic movement characteristics of the cat,and to understand the relationships between,and the influences of,these parameters.From this analysis,we provide significant data applicable to the design of joint movements in quadruped robot systems.

A 3D flexible neural interface based on a microfluidic interconnection cable capable of chemical delivery

The demand for multifunctional neural interfaces has grown due to the need to provide a better understanding of biological mechanisms related to neurological diseases and neural networks.Direct intracerebral drug injection using microfluidic neural interfaces is an effective way to deliver drugs to the brain,and it expands the utility of drugs by bypassing the blood-brain barrier(BBB).In addition,uses of implantable neural interfacing devices have been challenging due to inevitable acute and chronic tissue responses around the electrodes,pointing to a critical issue still to be overcome.Although neural interfaces comprised of a collection of microneedles in an array have been used for various applications,it has been challenging to integrate microfluidic channels with them due to their characteristic three-dimensional structures,which differ from two-dimensionally fabricated shank-type neural probes.Here we present a method to provide such three-dimensional needle-type arrays with chemical delivery functionality.We fabricated a microfluidic interconnection cable(pFIC)and integrated it with a flexible penetrating microelectrode array(FPMA)that has a 3-dimensional structure comprised of silicon microneedle electrodes supported by a flexible array base.We successfully demonstrated chemical delivery through the developed device by recording neural signals acutely from in vivo brains before and after KCl injection.This suggests the potential of the developed microfluidic neural interface to contribute to neuroscience research by providing simultaneous signal recording and chemical delivery capabilities.

Epidermal radio frequency electronics for wireless power transfer

Epidermal electronic systems feature physical properties that approximate those of the skin,to enable intimate,long-lived skin interfaces for physiological measurements,human–machine interfaces and other applications that cannot be addressed by wearable hardware that is commercially available today.A primary challenge is power supply;the physical bulk,large mass and high mechanical modulus associated with conventional battery technologies can hinder efforts to achieve epidermal characteristics,and near-field power transfer schemes offer only a limited operating distance.Here we introduce an epidermal,farfield radio frequency(RF)power harvester built using a modularized collection of ultrathin antennas,rectifiers and voltage doublers.These components,separately fabricated and tested,can be integrated together via methods involving soft contact lamination.Systematic studies of the individual components and the overall performance in various dielectric environments highlight the key operational features of these systems and strategies for their optimization.The results suggest robust capabilities for battery-free RF power,with relevance to many emerging epidermal technologies.