15 research outputs found
An efficient IoT based prediction system for classification of water using novel adaptive incremental learning framework
No full textCreating an adaptive, accurate, and reliable model is a universal problem. Machine learning models give poor accuracy on unseen data, and therefore, the testing accuracy of the trained model is affected. This study presents a novel adaptive incremental learning framework for IoT based smart water quality classification system to predict the suitability of water for different applications. Initially, water quality data is collected using IoT sensors. After that, data cleaning is performed by removing missing values and outliers. Next, features associated with the sensed data are obtained, and unwanted features are removed. Then, the G-SMOTE technique is proposed, which hybridizes the SMOTE and the genetic algorithm to address the imbalanced data set problem. After that, the multi-class classification is performed using a modified deep learning neural network classifier which uses hyperparameter tunning technique to obtain better accuracy with minimum validation loss. Finally, the study presents a novel framework for adaptive incremental learning on unseen data. Experimental result shows that our method presents a new state-of-the-art multi-class water quality classification method with an accuracy of 99.34% and validation loss of 0.0415
Usability of Access Ramps and Securement Systems on Transit Buses for Wheeled Mobility Device Users
No full textAbstract
Date Presented 3/30/2017
This study evaluated six ramp conditions and three securement devices encountered by wheeled mobility device users when using transit buses. The findings will help occupational therapists anticipate potential barriers and suggest accommodation strategies to improve community mobility outcomes for clients.
Primary Author and Speaker: Brittany Perez
Additional Authors and Speakers: Jim Lenker
Contributing Authors: Victor Paquet, Lydia Kocher, Medha Nemade</jats:p
Revolutionizing Healthcare through Health Monitoring Applications with Wearable Biomedical Devices
No full textThe Internet of Things (IoT) has revolutionized the connectivity and communication of tangible objects, and it serves as a versatile and cost-effective solution in the healthcare sector, particularly in regions with limited healthcare infrastructure. This research explores the application of sensors such as LM35, AD8232, and MAX30100 for the detection of vital health indicators, including body temperature, pulse rate, electrocardiogram (ECG), and oxygen saturation levels, with data transmission through IoT cloud, offering real-time parameter access via an Android application for non-invasive remote patient monitoring. The study aims to expand healthcare services to various settings, such as hospitals, commercial areas, educational institutions, workplaces, and residential neighborhoods. After the COVID-19 pandemic, IoT-enabled continuous monitoring of critical health metrics such as temperature and pulse rate has become increasingly crucial for early illness detection and efficient communication with healthcare providers. Our low-cost wearable device, which includes ECG monitoring, aims to bridge the accessibility gap for people with limited financial resources, with the primary goal of providing efficient healthcare solutions to underserved rural areas while also contributing valuable data to future medical research. Our proposed system is a low-cost, high-efficiency solution that outperforms existing systems in healthcare data collection and patient monitoring. It improves access to vital health data and shows economic benefits, indicating a significant advancement in healthcare technology
Modelling and Finite Element Simulation of a Surface Acoustic Wave Driven Linear Motor
No full textAbstractThe paper presents modeling and finite element simulation of a surface acoustic wave (SAW) linear motor. A SAW linear motor works on the principle of friction drive provided by SAW propagating on a piezoelectric stator. The SAW motor comprises of a cubical slider driven by Rayleigh wave generated on a piezoelectric substrate using an interdigital transducer (IDT) fabricated on the stator. In the study, a lithium niobate piezoelectric substrate is used as the stator on which aluminum IDTs are fabricated at the two edges and a cuboid slider is placed in the path of SAW propagation along with a preload. The characteristics such as displacement, velocity and forces acting on the slider for different amplitudes of wave excitations are studied. The slider in the SAW motor can move both in forward and reverse directions and the motor attains a saturated velocity with the continuous wave excitation
Structural, electrical and optical properties of nanostructured ZrO2 thin film deposited by SILAR method
No full textModeling and Simulation of a Piezoelectric Vibration Energy Harvester
No full textAbstractThe paper presents a lumped parameter model of a vibration energy harvester consisting of a bimorph piezoelectric cantilever with end mass. Stress on the piezoelectric material which is primarily accountable for the generation of electrical energy in the harvester is obtained along the beam length in terms of the end mass displacement. The model is applicable for parallel as well as series connection of the piezoelectric layers, and used to obtain resonant frequency, displacement of end mass and generated voltage across resistive load. Effect of load resistance on the resonant frequency and generated power is studied, and the results are verified with the finite element analysis in COMSOL Multiphysics
Simulation of Longitudinal Mode of Vibration in Piezoelectric Monolayer MoS2
No full textAbstractMonolayer molybdenum disulphide (MoS2) is one of the desired materials for the new age piezoelectric devices. The main objective of the paper is to present the finite element (FE) simulation of piezoelectric monolayer MoS2 in COMSOL Multiphysics software. A rectangular MoS2 sheet is simulated in fixed-fixed end and free-free end configuration. The eigenmode analysis is done and the eigen frequency is calculated. Due to the presence of one molecular layer and piezoelectric matrix the vibration should be in longitudinal mode. The longitudinal acoustic velocity is also calculated from the simulation and is verified analytically
An adaptive transformer-based framework for advanced brain activity mapping and intelligent neurotherapeutic decision support
No full textIntroductionIdentification and treatment of neurological disorders depend much on brain imaging and neurotherapeutic decision support. Although they are loud, do not remain in one spot, and are rather complex, electroencephalogram (EEG) signals are the principal tool used in research of brain function. This work employs an Adaptive Transformer-based technique with improved attention processes to extract temporal and spatial relationships in EEG data, effectively addressing these issues.MethodsFirst processed to eliminate noise and split them into time-series chunks, EEG data are then included into the proposed approach. Channel-wise embeddings and temporal encoding help to depict the data. Then, a transformer design including spatial attention for inter-channel interactions, multi-head self-attention for temporal aspects, and an adaptive attention mask for domain-specific modifications is used. Other openly accessible EEG datasets as well as the TUH EEG Corpus and CHB-MIT were evaluated against the model. Its performance was scored using metrics like accuracy, precision, memory, and F1-score.ResultsThe suggested method was more accurate than standard models like CNNs and LSTMs, with a score of 98.24%. The method was also shown to be able to find minor patterns in EEG data by improving precision and memory. Attention maps showed important areas of time and space, which made them easier to understand and useful in professional settings.DiscussionThe Adaptive Transformer turns out to be a useful tool for neurotherapeutic use of EEG data modeling. The approach provides greater medical assistance and knowledge on the functioning of the brain as well as answers significant issues. Future research might focus on subject-specific modifications and interaction with real-time systems.ConclusionThis study demonstrates the potential of transformer-based models in revolutionizing EEG analysis for precision brain imaging and neurotherapeutic decision-making
