Engineering Commons^™

Human Activity Recognition Using Wearable Sensors: A Deep Learning Approach, Jialun Xue

Theses

In the past decades, Human Activity Recognition (HAR) grabbed considerable research attentions from a wide range of pattern recognition and human–computer interaction researchers due to its prominent applications such as smart home health care. The wealth of information requires efficient classification and analysis methods. Deep learning represents a promising technique for large-scale data analytics. There are various ways of using different sensors for human activity recognition in a smartly controlled environment. Among them, physical human activity recognition through wearable sensors provides valuable information about an individual’s degree of functional ability and lifestyle. There is abundant research that works upon real …

Preparation And Characterization Of Reactive Composites Combining Boron With Mixed Oxides, Purvam Mehulkumar Gandhi

Theses

Thirteen liquids have been explored as Process Control Agents (PCAs) for high-energy mechanical milling of two metal oxide powders, CuO and Bi₂O₃, with the practical purpose to reduce the sizes of powder particles and refine their crystal structure. Multiple liquid properties have been considered to analyze their effect on evolution of both particle and crystallite sizes during milling for each oxide. Experimental data were analyzed using the δ-AICc (Akaike Information criterion) correlation function applied to linear relations between liquid properties and powder characteristics. Multiple linear correlation models were ranked based on the highest log-likelihood and the …

Incident Duration Time Prediction Using A Supervised Topic Modeling Method, Jihyun Park

Theses

Precisely predicting the duration time of an incident is one of the most prominent components to implement proactive management strategies for traffic congestions caused by an incident. This thesis presents a novel method to predict incident duration time in a timely manner by using an emerging supervised topic modeling method. Based on Natural Language Processing (NLP) techniques, this thesis performs semantic text analyses with text-based incident dataset to train the model. The model is trained with actual 1,466 incident records collected by Korea Expressway Corporation from 2016-2019 by applying a Labeled Latent Dirichlet Allocation(L-LDA) approach. For the training, this thesis …

Treated Hfo2 Based Rram Devices With Ru, Tan, Tin As Top Electrode For In-Memory Computing Hardware, Yuvraj Dineshkumar Patel

Theses

The scalability and power efficiency of the conventional CMOS technology is steadily coming to a halt due to increasing problems and challenges in fabrication technology. Many non-volatile memory devices have emerged recently to meet the scaling challenges. Memory devices such as RRAMs or ReRAM (Resistive Random-Access Memory) have proved to be a promising candidate for analog in memory computing applications related to inference and learning in artificial intelligence. A RRAM cell has a MIM (Metal insulator metal) structure that exhibits reversible resistive switching on application of positive or negative voltage. But detailed studies on the power consumption, repeatability and retention …

Cycloidal-Drive Joint Design For Wearable Exoskeletons, Abhi J. Rawal

Theses

This thesis's scope was to construct a highly back drivable and compact joint design that can be used for wearable exoskeletons specifically designed to assist with rehabilitation. A compact cycloidal gear transmission was designed to satisfy the compactness requirement while ensuring the minimum torque is required for backdriving the mechanism. The joint was designed based on the compact cycloidal drive specifications, and the backdrivability test was performed to measure the minimum torque required (1.20 Nm) to backdrive. The joint design incorporates various important features such as rotation lock feature to prevent hyperextension, a housing design that offers the benefit of …

Comparison Of Longitudinal Changes In Resting State Functional Magnetic Resonance Imaging Between Alzheimer’S And Healthy Controls, Berk Can Yilmaz

Theses

Resting State Functional Magnetic Resonance Imaging (rs-fMRI) is a technique that is widely used for analyzing brain function using different approaches and methods. This study involves rs-fMRI analysis of Blood Oxygenation Level Dependent (BOLD) signals acquired from Alzheimer’s disease (AD) Patients and Healthy Controls (HC). Each subject in the study had both functional and anatomical images with at least one rs-fMRI scan with their Anatomical (T1) scans. Previous rs-fMRI studies have demonstrated that AD shows differences in Amplitude of Low Frequency (<0.1 Hz) Fluctuations (ALFF), and Regional Homogeneity (ReHo) measures according to HCs.

The aim of the study is to investigate individual and group level differences using ReHo and mALFF related …

A Model For Thermal Diffusion And Local Tissue Damage From The Measurement Of Temperature In Electrosurgical Incision, Marek Vira

Theses

The extremely high temperatures applied over extremely short time intervals that are characteristic of electrosurgery result in a unique tissue damage pattern. Cesarean delivery and hysterectomy, the two most frequently performed procedures in obstetrics and gynecology (OB/GYN), commonly employ electrosurgical incision. While it is controversial, it has been suggested that tissue damage produced by electrosurgery could increase surgical site infection rates. While recommendations for the settings in the use of the electrosurgical unit do exist, there is no current technique for real time assessment of the viability of tissue around the site of the electrosurgical incision. Current methods for analyzing …

Deep Learning For Quantitative Motion Tracking Based On Optical Coherence Tomography, Peter Abdelmalak

Theses

Optical coherence tomography (OCT) is a cross-sectional imaging modality based on low coherence light interferometry. OCT has been widely used in diagnostic ophthalmology and has found applications in other biomedical fields such as cancer detection and surgical guidance.

In the Laboratory of Biophotonics Imaging and Sensing at New Jersey Institute of Technology, we developed a unique needle OCT imager based on a single fiber probe for breast cancer imaging. The needle OCT imager with sub-millimeter diameter can be inserted into tissue for minimally invasive in situ breast imaging. OCT imaging provides spatial resolution similar to histology and has the potential …

Model-Based Deep Siamese Autoencoder For Clustering Single Cell Rna-Seq Data, Zixia Meng

Theses

In the biological field, the smallest unit of organisms in most biological systems is the single cell, and the classification of cells is an everlasting problem. A central task for analysis of single-cell RNA-seq data is to identify and characterize novel cell types. Currently, there are several classical methods, such as K-means algorithm, spectral clustering, and Gaussian Mixture Models (GMMs), which are widely used to cluster the cells. Furthermore, typical dimensional reduction methods such as PCA, t-SNE, and ZIDA have been introduced to overcome “the curse of dimensionality”. A more recent method scDeepCluster has demonstrated improved and promising performances in …

Recovery Of Valuable Metals From Spent Lithium-Ion Batteries Using Organic Acids: Assessment Of Technoeconomic Feasibility, Leqi Lin

Theses

Lithium ion batteries (LIBs) are used in diverse electronic products with anticipated over 500 thousand tonnes of the waste LIBs globally in 2020. To protect the environment and also recover valuable materials such as lithium (Li) and cobalt (Co), our research employed a hydrometallurgy method and demonstrated that exposure of spent LIBs to Organic Aqua Regia (OAR) could leach Li and Co without the pre-separation of cathode from Al foil using organic solvents such as Dimethylformamide (DMF) and N-Methyl-2-pyrrolidone (NMP). The leaching efficiency of 99% and 94% for Li and Co were obtained with a leaching rate of 0.021, 0.167 …

Enhanced Removal Of Perfluorooctanoic Acid (Pfoa) Via Microwave-Fenton-Reactive Membrane Filtration, Fangzhou Liu

Theses

Perfluorooctanoic Acid (PFOA), one of the common per- and poly fluorinated alkylated substances (PFASs), is increasingly detected in the environment due to the diverse industrial applications and high resistance to degradation processes. This study evaluated degradation of PFOA in microwave-assistant catalytic membrane filtration, a process that integrates microwave catalytic reactions into a ceramic membrane filtration. First, water permeation of the pristine and catalyst-coated membranes were examined under the influence of microwave irradiation to analyse the impacts of the coating layer and water temperature increase on permeate flux, which were well interpreted by three models. Then, the PFOA removal was first …

An Information Theoretic Approach To Assess Perceived Audio Quality Using Eeg With Reduced Number Of Electrodes, Sansit Das

Theses

Electroencephalograph(EEG) is a process mainly used in medical and research fields to study the electrical activities in a brain. In this technique, 128 or 256 electrodes are attached to the scalp and the electrical activities of the human brain is recorded with the help of a software. In the global scenario, the EEG responses are studied and analysed to acknowledge any disorders in the brain, such as epilepsy or head injury.

Recent studies performed by researchers, have focused on analysing these electrical activities to access perceived audio quality from users by using information theoretic approaches, such as mutual information. Experiments …

Spectral 3d Reconstruction Based On Macroscopic Oct Imaging, Xingyu Zhou

Theses

Various optical technologies have been utilized to improve art conservation by art conservators, such as laser triangulation, stereophotogrammetry, structured light, laser scanner and time of flight sensors. These methods have been deployed to capture the 3D or surface topography information of sculptures and architectures. Optical coherence tomography (OCT) has introduced new imaging methods to study the surface features and subsurface structures of delicate cultural heritage objects. However, despite its higher spatial resolution, the field of view (FOV) of OCT severely limits the size of the scanning area and does not allow macroscopic examination. To solve this issue, we develop and …