172 research outputs found
Author Correction: A computational framework for guiding the MOCVD-growth of wafer-scale 2D materials (npj Computational Materials, (2022), 8, 1, (240), 10.1038/s41524-022-00936-y)
No full textIn this article the author name Nuruzzaman Sakib was incorrectly written as Nurruzaman Sakib. The original article has been corrected. © The Author(s) 2022
Can government-sponsored sustainable agricultural farming practices reduce land decay through crop biodiversity conservation under production uncertainties?
No full textUnder income uncertainties, agricultural farmers might be influenced by government-sponsored programs that might lead to higher income opportunities by focusing on monoculture at the expense of crop diversification strategy. However, the latter strategy is likely to reduce production uncertainties for agricultural farmers and hence, ensuring sustainable agricultural development in the targeted area. A theoretical model is proposed to understand such possible economic trade-offs between high income-lower crop diversification and lower income-higher crop diversification outcomes resulting from government-sponsored programs and institutions
Estimating Mixing Parameters Using Ocean Buoyancy Glider Hydrography and Vehicle Dynamics: Applications to Gulf Loop Current Structure
No full textThis dissertation has combined observations from four Slocum glider missions in the deep Gulf of Mexico to quantify the turbulent kinetic energy (TKE) dissipation rates (��) and diffusive mixing processes in the basin. The Thorpe scale (TM) method is used to estimate �� and then used to construct depth �� profiles (surface to 1000 m) using the Large Eddy Method (LEM). The accuracy of the TM-LEM estimates are compared and quantified against direct estimates from a simultaneous /co-located MicroRider deployment in the Gulf of Mexico (i.e., a glider equipped with MicroRider). Survey-averaged profiles of the three methods are compared and found to be within the range of expected error, i.e., within a factor of 2. Profile to profile comparison of �� reveals that LEM overestimates when the magnitude of �� is small. The overestimation is attributed to the stratification-dependent detection limits of the LEM and is mostly observed in deep water, where �� falls close to the noise level of LEM. Spectral comparison of dissipation rates from the three methods (using histograms of occurrence) confirms that the LEM and TM are able to capture dissipation rate variability greater than 1 �� 10���9 Wkg���1; however, less than this limit, only the direct measurement of TKE dissipation rate (in regions of weak vertical density gradients) are robust. Despite this limitation, the TM -LEM-derived dissipation rate estimates are able to provide structures that are interpretable as associated with the underlying physical processes of the deep ocean. Maps showing the temporal and spatial variability of �� are able to reveal the well-defined turbu-lence structure of LCE and LC. Eddy-induced elevated �� are observed around the core of LC and LCE, but the interior of the eddy core is relatively quiescent when compared to the oceanic frontal regions of the eddy. Diapycnal mixing around the eddy cores is suppressed due to the presence of stronger stratification. Away from the eddy cores, where stratification is less, diapycnal mixing is enhanced. The analysis quantifying the relative strength of the diffusion processes, using Turner Angle and density ratio, concluded that salt-fingering is the dominant double-diffusive process in the GoM and is related to proximity to the LC and to depth of observation influence the strength of the salt-fingering in the water column. The potential for fine-structure thermohaline staircases is quantified and observations of irregular shape staircases in the deep GoM are reported for the first time. The glider-based measurements provide an economical option to estimate ocean turbulence and has the potential to fill the gaps between the direct microstructure measurements provides opportunity to obtain mixing parameters of the world ocean in the absence of direct microstructure observations
DEEP LEARNING TECHNIQUES FOR KIDNEY DISEASE DETECTION
No full textAround 10% of the world's population gets affected by Chronic Kidney Diseases or CKDs at some point in their lives and millions die each year due to not having access to affordable treatments and clinical facilities. During the last two decades, CKDs and short-term Acute Kidney Injuries (AKIs) have been steadily increasing in low- and middle-income countries due to obesity, diabetes, and other diseases. Based on recent studies, CKDs are directly correlated to kidney cancers and the diagnosis of kidney cancers through laboratory tests is time-consuming, complicated, unreliable, and costly. Kidney cancer is, in fact, one of the most common malignant CKDs worldwide. Accurate diagnosis is a critical step in the management of patients with kidney cancer and is influenced by multiple factors including tumor size or volume, cancer type or stage, etc. For malignant tumors, partial or radical surgery of the kidney might be required, but for many clinicians, the basis for making this decision is often unclear. Partial nephrectomy could result in avoidable patient death if kidney removal was needed, whereas total nephrectomy in less severe cases could resign patients to life-long dialysis or need for future transplantation without sufficient cause. Using machine learning to consider clinical data alongside computed tomography images could potentially help resolve some of these issues, by enabling a more robust classification of kidney cancers and selection of optimal surgical approaches. In this study, I used the publicly available KiTS dataset of contrast-enhanced CT images and corresponding patient metadata to differentiate four major classes of renal cell carcinoma: clear cell (ccRCC), chromophobe (chRCC), papillary (pRCC), and oncocytoma (ONC). I rationalized these data to overcome the high field of view (FoV), extract tumor regions of interest (ROIs), classify patients using deep machine learning models, and extract / post-process CT image features for combination with clinical data. Regardless of marked data imbalance, the proposed combined approach with the help of the proposed novel DenseAUXNet201 classifier achieved a high level of performance (85.66% accuracy, 84.18% precision, 85.66% recall, and 84.92% f1-score). When selecting surgical procedures for malignant tumors (RCC), the proposed method proved even more reliable (90.63% accuracy, 90.83% precision, 90.61% recall, and 90.50% f1-score). Using feature ranking, it was confirmed that tumor volume and cancer stage are the most relevant clinical features for predicting surgical procedures. Once fully mature, the approach I propose could be used to assist surgeons in performing nephrectomies by guiding the choices of optimal procedures in individual patients with kidney cancer. Adults are not the only ones who suffer from kidney complications, even fetuses might suffer from kidney problems such as hydronephrosis. Antenatal or prenatal hydronephrosis (AHN) is a common kidney complication in unborn children. While AHN is generally benign and resolves over time, in severe cases this condition can inflict serious kidney damage or even organ failure due to excessive waste accumulation. Regardless of severity, AHN must be clinically monitored for resolution, with treatment plans and medications being revised according to updated prognoses. Kidney ultrasound (US) images are one of the most common methods of monitoring AHN, but grading of this condition is highly subjective and clinicians may select inappropriate therapies or surgical interventions as a result. New approaches are required to differentiate subjects who can be managed without surgical intervention from those who require life-saving operations. An end-to-end deep machine learning framework was developed to sequentially detect ultrasound regions of interest (ROIs), segment kidneys from US images, and classify AHN severity. I propose the novel Kidney Ultrasound Segmentation Network (KUSNet) for kidney segmentation from ultrasound images, and the Prenatal Hydronephrosis Classification Network (PHCNet) for hydronephrosis severity stratification according to Society of Fetal Urology standards. At each stage, the performance of the proposed models was assessed both quantitatively and qualitatively against state-of-the-art networks in the respective fields. PHCNet achieved 87.7% accuracy, 88.1% precision, 87.7% recall, 78.0% specificity, and 87.7% f1-score when performing multiclass stratification of AHN severity. Artificial intelligence-based tools can reliably classify AHN severity to reduce inter- and intra-observer bias, thereby aiding clinicians in the rapid selection of appropriate treatments and surgeries
What works best to motivate students in a general education introductory economics course?
No full textConsidering the research gaps on student motivation of treating economics as an interesting subject matter, the learning goal of my research is to find what works best to engender positive learning experience for students dealing with serious motivational issues. My research design is based on the convergent parallel mixed methods using the quantitative pre-and-post anonymous online questionnaire surveys and the qualitative short reflection notes
Comparing the Performance of Bottom-Moored and Unmanned Surface Vehicle Towed Passive Acoustic Monitoring Platforms for Marine Mammal Detections
No full textPassive acoustic monitoring (PAM) is a more effective method of monitoring cetaceans’ distribution and abundance than conventional visual surveys. Cetaceans are highly vocally active and produce identifiable acoustic signals during echolocation and communication. Three different PAM platforms recorded data in overlapping time periods in the vicinity of the 2010 Deepwater Horizon oil spill site: bottom-moored buoys (EARS), Unmanned Surface Vehicle towed arrays (USV), and subsurface glider-mounted hydrophones. Detection rates of the EARS and USV were compared to investigate their efficiency in detecting marine mammals. Detection events were obtained using independent detectors for each platform and then compared by feeding data through a common detector. Results from both detectors and platforms were compared, and a comparable trend of detection rates was found. The purpose of this study is to aid in the development of cost-efficient PAM methodology for mitigation and environmental impact assessment purposes
The Corporate Social Responsibilities of Robi Axiata Ltd.
No full textInternship reportThe report deals with the Corporate Social Responsibilities-CSRs of Robi Axiata
Ltd. The report includes the brief discussion of CSR activities and exemplifies in
detailed by the CSRs of Robi. Robi Axiata Limited is the 2nd largest mobile network operator in Bangladesh.
With its 44.225Mn it is leading towards success. The company is the subsidiary of
the Asian telecom giant Axiata Group Berhad, based in Malaysia. Another
subsidiaries are Bharti Airtel International (Singapore) Pte Ltd. and NTT DOCOMO
Lnc. The company initiated its business in 1997 as Telecom Malaysia International &
adopted the brand name Aktel. The company rebranded itself as Robi Axiata Ltd.
in 2010 & took new initiatives to strengthen its name with new vigor in the
Bangladeshi Market.
Robi has taken some revolutionary steps as CSRs in Bangladesh and gained
popularity all over the world. Among them Gori Nijer Bhobisshot, Robir Alo, Robi
10 Minute School, Internet4U gained popularity. The success stories are covered
in detail in the report. At the bottom of the report covers the effectiveness and
contribution of these initiatives in Bangladeshi Socio-economic system.Not applicabl
Connected and Automated Vehicles in Urban Transportation Cyber-Physical Systems
No full textUnderstanding the components of Transportation Cyber-Physical Systems (TCPS), and inter-relation and interactions among these components are key factors to leverage the full potentials of Connected and Automated Vehicles (CAVs). In a connected environment, CAVs can communicate with other components of TCPS, which include other CAVs, other connected road users, and digital infrastructure. Deploying supporting infrastructure for TCPS, and developing and testing CAV-specific applications in a TCPS environment are mandatory to achieve the CAV potentials. This dissertation specifically focuses on the study of current TCPS infrastructure (Part 1), and the development and verification of CAV applications for an urban TCPS environment (Part 2).
Among the TCPS components, digital infrastructure bears sheer importance as without connected infrastructure, the Vehicle-to-Infrastructure (V2I) applications cannot be implemented. While focusing on the V2I applications in Part 1, this dissertation evaluates the current digital roadway infrastructure status. The dissertation presents a set of recommendations, based on a review of current practices and future needs.
In Part 2, To synergize the digital infrastructure deployment with CAV deployments, two V2I applications are developed for CAVs for an urban TCPS environment. At first, a real-time adaptive traffic signal control algorithm is developed, which utilizes CAV data to compute the signal timing parameters for an urban arterial in the near-congested traffic condition. The analysis reveals that the CAV-based adaptive signal control provides operational benefits to both CVs and non-CVs with limited data from 5% CVs, with 5.6% average speed increase, and 66.7% and 32.4% average maximum queue length and stopped delay reduction, respectively, on a corridor compared to the actuated coordinated scenario.
The second application includes the development of a situation-aware left-turning CAV controller module, which optimizes CAV speed based on the follower driver\u27s aggressiveness. Existing autonomous vehicle controllers do not consider the surrounding driver\u27s behavior, which may lead to road rage, and rear-end crashes. The analysis shows that the average travel time reduction for the scenarios with 600, 800 and 1000 veh/hr/lane opposite traffic stream are 61%, 23%, and 41%, respectively, for the follower vehicles, if the follower driver\u27s behavior is considered by CAVs
Consequences of public programs and private transfers on household’s investment in protection from natural disasters
No full textConsidering the issues of households’ accessibility to public programs and private inward remittances, there is a need to better understand the linkages through which households’ decision to pursue private defensive strategies (or private protection activities) might be influenced. This has significant policy implications especially for low-and-middle income countries vulnerable to natural disasters. We introduce a theoretical model of household private investment in protection against damages from a natural disaster event given the presence of public programs and the possibility of receiving inward remittances from members of the household
- …
