Environmental and Occupational Health Sciences Institute
Rutgers University Community RepositoryNot a member yet
58345 research outputs found
Sort by
Results from 95Mo(d,p) reaction measurements: informing the 96Mo level scheme with GODDESS and investigating the (d,p) reaction as a surrogate for the (n,n’) reaction with STARLiTeR
No full textApproximately half of the elements heavier than iron are formed through the rapid neutron capture process in explosions in the cosmos. Isotopes identified in neutron capture, (n,γ), sensitivity studies are often too short lived to create a solid target. By using Radioactive Ion Beam (RIB) facilities and indirect reaction methods, (n,γ) cross sections with high impact on predictions of r-process abundances can be constrained. The Surrogate Reaction Method (SRM) was validated by Ratkiewicz and colleagues for neutron transfer (d,p) reactions as a surrogate for (n,γ) reactions in normal kinematics. To further benchmark the use of (d,p) reactions as a surrogate for (n,γ) reactions on unstable isotopes, a 95Mo beam was impinged on C2D4 thin film targets. The subsequent (d,p) reaction products were measured using the particle-γ spectrometer GODDESS (Gammasphere ORRUBA: Dual Detectors for Experimental Structure Studies), where ORRUBA (Oak Ridge – Rutgers University Barrel Array), an array of silicon-strip detectors, is coupled to the gamma detector array Gammasphere.Challenges from the 2015 GODDESS commissioning campaign are discussed in regards to the limitations of the inverse kinematics 95Mo(d,p) reaction measurement data for use in a SRM analysis. Results from the γ-singles, γ − γ coincidences, and triple-γ coincidences are presented to inform the level scheme of 96Mo.The neutron induced (n,n’) reaction is important to the fields of astrophysics, nuclear forensics, and national security. It was recently determined that the (d,p) reaction could also be a potential surrogate reaction for the (n,n’) reaction. Data from the 95Mo(d,p) reaction measurement made with the (Silicon Telescope Array for Reactions with Livermore, Texas A&M University, Richmond (STARLiTeR) and performed with a deuteron beam at 12.4 MeV incident upon an enriched (98.6%) 95Mo target were analyzed working towards verifying the (d,p) reaction as a surrogate for the (n,n’) reaction. Constraining the 95Mo(n,n’) cross-section through the surrogate (d,p) reaction requires the analysis of proton-γ coincidences and γ-decay emission probabilities Ppγ of the 95Mo nucleus. Gamma-ray emission probabilities, Ppγ , for transitions in 95Mo from the (d,p) reaction populating states above the neutron separation energy of 96Mo are presented with a discussion of next steps.Ph.D.Includes bibliographical reference
Testing lepton universality with ep and μp elastic scattering in the MUSE experiment
No full textThe MUon proton Scattering Experiment (MUSE) at the PiM1 beam line of the Paul Scherrer Institute in Villigen, Switzerland, simultaneously measures the elastic scattering cross sections of electrons and muons from a liquid hydrogen target. The main objective of the experiment is to determine the charge radius of the proton by extrapolating the Q2 dependence of the electric form factor as Q2 → 0 to resolve the proton radius puzzle. The proton radius puzzle arose from the 5σ discrepancy between recent muonic hydrogen spectroscopy measurements and electron-proton scattering measurements. New studies and measurements have been done since the puzzle was established, yet the discrepancy remains unresolved. MUSE will be the first high precision experiment to measure ep and μp elastic scattering at the same time in the same apparatus, and directly compare the differences between ep and μp interactions. By comparing the cross sections, MUSE will explore different possible explanations for the proton radius puzzle, including two photon exchange and lepton universality violation.
Lepton universality, as defined by the Standard Model states that all lepton interactions are the same except for trivial mass differences. The recent muon g − 2 experiment from BNL and FermiLab raised the possibility of the violation of lepton universality, which is a potential solution to the proton radius puzzle. MUSE will test lepton universality by comparing the elastic scattering cross sections of ep and μp for both negative and positively charged leptons. In this dissertation, a study of this topic will be presented with blinded preliminary data from the 2023 beam time at momenta of ±115 MeV/c and ±160 MeV/c. The discussion will involve an overview of the proton radius puzzle, the past and recent effort on the violation of lepton universality search, and the details of the MUSE experimental setup. Analysis along with detector calibration that lead to the preliminary result will be introduced. The preliminary data shows that the ratio of the measured electron proton cross section to the muon proton cross section agrees with prediction at the level of the blinding range. While further analysis is ongoing, current results demonstrate the method and ability to test lepton universality at MUSE.Ph.D.Includes bibliographical reference
Uncovering the ciliary and mitochondrial functions of the NIMA-related kinase NEKL-4
No full textNeurodegeneration is a devastating phenomenon that may result from aging or a failure in one or multiple biological processes that regulate neuronal homeostasis. One of these processes involves the maintenance of the neuronal microtubule cytoskeleton, which is subject to multiple post-translational modifications that dictate how often and how quickly molecular motors travel throughout the cell. Dysregulation of one of these modifications- glutamylation- results in neurodegeneration. Therefore, studies into factors that regulate tubulin glutamylation is essential to understanding how neurodegeneration may be lessened or prevented. In this thesis, I characterize the ciliated neuron-exclusive kinase NEKL-4 and its discovery as a genetic interactor of CCPP-1, a deglutamylase enzyme essential for maintaining neuronal homeostasis via removal of glutamate residues from tubulin. I identify functions of this understudied kinase in the regulation of both ciliary stability and mitochondrial homeostasis. By these two routes, NEKL-4 influences the health of the ciliated neurons and acts as a disease modifier of neurodegenerative phenotypes.Ph.D.Includes bibliographical reference
Microplastic concentration, characterisation, and size distribution in surface waters of the Delaware Bay estuary
No full textMicro and nano plastic (MP and NP) pollution widely contaminates the Earth’s water, food, air, and organisms. With no foreseeable end to plastic use and pollution- and large uncertainties regarding the toxicity and risk plastic poses to humans and the environment- there is increased public, scientific, and governmental attention and concern for this issue. However, the characteristics and abundance of plastics are not well studied or characterized in many environments. Estuaries are one such understudied environment that are crucial to the understanding of oceanic MPs. Furthermore, estuaries provide unique environments that can greatly alter the fate, transport, size distribution, and abundance of plastic pollution. This study was conducted in an attempt to quantify and characterize micro (<5 mm) and macroplastic pollution in the Delaware Bay estuary, to understand the fate, transport, abundance, and size distribution of such plastics. Samples (n=31) were collected in frontal and non-frontal zones of the Delaware Bay during two sampling campaigns (2019 and 2022) using nets of various mesh sizes. Microplastics were extracted from the collected particles using wet peroxide oxidation (WPO) and density separation (NaCl) techniques. After density separation, floating particles were collected on 0.5 mm mesh sieves and dried prior to analysis. Chemical analysis was done via Fourier transform infrared (FTIR) spectroscopy for all particles collected from the 31 samples on the 0.5 mm sieve (n=1015). For the two samples with too many particles to analyze, subsampling was performed. Across all samples, 327 microplastics and 11 macroplastics were analyzed. Particles from samples collected from 5/3/22-5/5/22 (n=864) also had color, morphology, and two dimensions recorded. Polypropylene and polyethylene were by far the most abundant plastic type, comprising 95% of total plastics (i.e., including both microplastics collected on 0.5mm sieve and macroplastics) from quantitative samples. Total plastic concentrations (MP and macroplastic particles/m3 of ocean water) ranged from below detection (defined here as one particle per largest sample volume: one particle /120.12 m3) to 4.25 plastics per m3. The size distribution of plastics showed some similarities to size distributions observed in other water bodies. A correlation between total particles and total plastic particles was observed. The results and data from this study may be used as input into fate and transport and or fragmentation models and input for risk assessment, to spread awareness, and guide mitigation, remediation, and policy for plastic pollution.M.S.Includes bibliographical reference
Implementation of an opioid misuse screening tool in the emergency department
No full textPurpose of Project: The misuse of opioids has become an emerging health concern throughout the United States. While numerous screening tools exist, a standardized tool is lacking to identify patients for potential opioid misuse within emergency departments (ED) where patients are prescribed opioids for acute pain management, in many cases, for the first time. This quality improvement project aims to improve the identification of adult patients at risk for opioid misuse before prescribing opiates, enabling providers to reconsider non-opiate regimens and facilitate referrals for pain management.
Methodology: Forty-six participants were recruited to complete the twelve-item Screeners and Opioid Assessment for Patients with Pain (SOAPP-R-12) questionnaire measuring potential opioid misuse risk. Those who screened positive were referred to pain management on discharge. A follow-up phone call was made within one week to determine where a pain management appointment was scheduled. Inclusion criteria were adults over 21 years experiencing acute pain within the last four weeks.
Results: SOAPP-R-12 scores and scheduled pain management follow-up appointments were analyzed using descriptive statistics. Results showed that 73.3% of those screened positive on SOAPP-R-12 scheduled a follow-up appointment. The average SOAPP-R-12 score was 9.8, indicating a high-risk screening result for potential opioid misuse.
Implications for Practice: Results of this project indicated that utilizing a screening tool in the ED improved provider identification of patients at risk for potential opioid misuse. Further research is needed to determine the effects of implementing this tool across various EDs nationwide to potentially enhance referrals for pain management to ensure patient safety.D.N.P.Includes bibliographical reference
Implementation of a palliative care screening tool in a long-term acute care hospital (LTACH)
No full textPurpose of Project: The purpose of the project was to conduct a plan-do-study-act (PDSA) quality improvement cycle implementing use of the Center to Advance Palliative Care (CAPC) screening criteria for palliative care referral in a 25-bed LTACH for all patients admitted over a 10-week period.
Methods: During a 5-week period, the nurse-driven screening tool was used for all patients following staff education. The immediately preceding 5-week period was examined without use of the tool for all patients admitted. Both pre- and post-intervention groups were compared and analyzed for statistical significance relative to categorical variables and along data points, including use of the tool, criteria met, and whether a palliative care consultation was indicated, ordered, and completed.
Results: Pre-intervention, 8 of 16 patients received an appropriate palliative care consultation, while 10 of 14 received a palliative care consultation post-intervention. There was a 25% increase in palliative care consultations ordered with tool use. Missed appropriate consultations decreased by 50%. While not statistically significant, the results were clinically significant since an increased awareness of end-of-life care and conversations at the LTACH were achieved.
Implications for Practice: Additional future PDSA quality improvement cycles evaluating the use of the tool over longer timeframes and conducted throughout the tri-campus LTACH system with a wider patient population are recommended to develop further and establish a plan for a clear nurse-driven, multidisciplinary, collaborative LTACH palliative care program.D.N.P.Includes bibliographical reference
Expedited partner treatment for sexually transmitted infections: chlamydia and gonorrhea
No full textPurpose of Project: The purpose of this quality improvement project was to determine the rates of EPT and its correlation to re-infection rates, specifically for patients who test positive for Chlamydia and Gonorrhea.
Methods: A secondary analysis via a retrospective chart review was conducted for patients who tested positive for chlamydia and gonorrhea during January 2023 and February 2023.
Results: A total of 33 patient charts were evaluated. Patient demographics were collected including age, race, insurance status, and educational background. A total of 33 patients tested positive for chlamydia and gonorrhea. Of the 33 patients, 66.67% (n=22) patients were offered EPT; of those 22.73% (N=5) patients were reinfected. Of the 33 patients, 33.33% (n=11) patients were not offered EPT; of those 9.09% (n=1) patients were reinfected. Regardless of whether EPT was provided, 42.42% (n=14) of patients did not have follow-up cultures, and therefore re-infection status could not be determined for those patients.
Implications for Practice: Implications of these results involve practice change initiatives including the implementation of clinical practice guidelines, increased patient education, annual provider training, and reduced health-care economic burdens. Overall, expedited partner treatment and follow-up cultures are important in the reduction of reinfection rates.D.N.P.Includes bibliographical reference
Thermal analysis of lead minerals found in Newark, New Jersey drinking water distribution systems
No full textPrevalent lead (Pb) minerals from drinking water systems were analyzed in the heating range from 25 °C to 800 °C in nitrogen (N2(g)) and air atmospheres and these analyses were utilized to identify mineral constituents of three mixed phase lead pipe scale samples obtained from the United States Environmental Protection Agency (USEPA) sourced from a local Newark, New Jersey distribution system. X-ray diffraction (XRD) and solid phase Fourier-transform infrared spectroscopy with universal attenuated total reflection (UATR-FTIR) were utilized for mineral identification, while simultaneous thermal analysis coupled with evolved gas analysis and Fourier-transform infrared spectroscopy (STA-EGA-FTIR) was used to evaluate mass loss, energy changes and corresponding gases evolved from mass loss events. Results from XRD and FTIR analyses confirmed the USEPA samples to be a mixture of mainly Pb-bearing minerals dominated by Pb carbonates and Pb oxides. Thermogravimetric (TG) mass loss, differential thermogravimetric analysis (DTG), and differential scanning calorimetry (DSC) curves were most influenced by the carbonate content of the samples. Peaks in the DTG, DSC and the Gram Schmidt (GS) absorbance for the samples aligned with peaks for hydrocerussite (Pb3(CO3)2(OH)2) and cerussite (PbCO3) in the 200 – 400 °C range, with the intensity of the peaks correlated with the carbonate content of the sample. The GS curves were analyzed at temperatures of the DTG peaks to determine specific gases evolving based on FTIR data. The FTIR spectra indicated that the primary gas evolved at all temperatures was carbon dioxide (CO2(g)). Data from thermal analysis and gas evolution were compared to the carbonate content determined by XRD analysis to evaluate thermal analysis as a viable tool for mixed mineral sample identification for these and other pipe scale samples. The method was shown to be effective at characterizing Pb pipe scale mineral assemblages with >50% carbonate minerals, to confirm gaseous evolution of CO2(g) with STA-EGA-FTIR, and additional corresponding characterizations are necessary to confirm accurate mineralogical composition.M.S.Includes bibliographical reference
Three essays on outlier detection in auditing: applying machine learning to accounting data for full-population testing
No full textFull-population testing in auditing has drawn much attention today due to technological advancements. A branch of machine learning literature, unsupervised outlier detection, appears to be a suitable tool for auditors to discover abnormalities in large accounting data such as general ledger (GL). It identifies outliers that deviate from the majority of observations in a dataset without utilizing any prior knowledge. The algorithms learn normal patterns from the data and calculate an outlier score for each observation based on its deviation from the normal patterns. A large score indicates a higher probability of being an outlier. Applications in various domains, such as credit card fraud detection, prove useful insights can be generated through the identified outliers. However, the application of outlier detection in auditing has not been thoroughly explored in existing literature. Since unsupervised outlier detection is not guided by prior knowledge, investigation of the identified outliers provides a chance for auditors to discover new potential problems. However, “there ain’t no such thing as a free lunch” – the credibility of a detection algorithm is hard to evaluate without prior knowledge about the actual outliers. When various algorithm candidates flag different observations as outliers, it is almost impossible for auditors, who have limited investigation capacity, to manually examine all the outliers. Therefore, to establish outlier detection as a viable technique in auditing, it is imperative to effectively address the evaluation difficulty.
Additionally, insights obtained through the investigation of outliers are valuable and have the potential to benefit future audits. From this perspective, auditors can employ supervised classifiers for outlier detection purposes. Specifically, investigated outliers from previous years can be collected to form a labeled data pool and utilized to train supervised classifiers for outlier prediction in the current year. As the investigated outliers accumulate year by year, the classifiers learn from more experience, which may improve their prediction accuracy. However, some factors, such as the quality of the initial training set, may negatively affect the prediction accuracy. Hence, more research is required to examine the impact of relevant factors in this specific auditing application scenario on overall prediction accuracy so that auditors can gain some insights into the reliability of such a methodology.
To explore how to apply machine learning to accounting data for full-population testing, this dissertation 1) proposes a framework to identify various types of outliers in general ledger data and tests whether new insights are gained through the identified outliers, 2) studies properties of outlier score distributions and proposes an internal validation solution, and 3) studies whether investigated transactions in previous periods can be used to train supervised classifiers so that they can identify more real outliers in the current period.
The first essay (Chapter 2) proposes a Multilevel Outlier Detection Framework (MODF) that identifies outliers at three levels: transaction level, account level, and combination-by-variable level, respectively, with each level representing a specific type of outliers in general ledger (GL). The framework is tested with both real and synthetic datasets to test its effectiveness in gaining new insights. For the real GL dataset, an external audit team from one of the world’s largest audit firms partnered with our research team to provide both the data and the feedback on the outlier detection results. The synthetic GL dataset used is from a case study created by Ernst Young Academic Resource Center (EYARC), a clean dataset with no fraud. To test whether outlier detection can capture errors that are not obvious to auditors, we create seeded errors in a way that the variable values fall in the same ranges as the original observations in the dataset, but the dependency structure between the variables is changed. Since the dependency structure is not directly observable, such seeded errors are difficult to identify by auditors’ judgment. The performance of the framework on the two datasets indicates that outlier detection can be utilized to gain new insights regarding the risk of misstatements in GL data.
The second essay (Chapter 3) focuses on the main challenge of applying outlier detection in auditing – algorithm evaluation with no ground truth. In particular, the properties of outlier score distributions are studied for internal validation. When scores from an algorithm are normalized and ranked in descending order, how the scores are distributed between one and zero is referred to as a score distribution. The slope at an observation is the score difference between this observation and its immediate right neighbor. We find that if an algorithm can successfully identify real outliers, their score distributions always involve a shift in the pattern of neighboring score differences. It motivates us to develop a hypothesis that the shift signifies the transmission from the identified real outliers to the remaining observations, and the location of the shift has a positive relationship with algorithm performance. We conduct a proof of concept to test the hypothesis using six methods and their parameterizations on real and synthetic data. The results indicate that a positive relationship exists. We then propose an internal validation solution called the Index of Score Distribution Breakpoint (ISDB) to estimate the location of the shift in a given score distribution. The algorithm with the largest ISDB is selected. The evaluation results show that the algorithm selected by the ISDB methodology tends to perform well.
The third essay (Chapter 4) explores whether auditors’ previous experience with a client can be utilized to train supervised learning algorithms for outlier detection. Specifically, as auditors investigate a sample of transactions in each audit period, those investigated transactions can be used to train supervised classifiers for future prediction. We study two main factors that can affect the prediction accuracy of supervised classifiers to provide some insights into the reliability of such a methodology: data similarity across different periods and quality of the initial training set. Particularly, we design an experiment to examine the impact of the data similarity across different periods, the class balance in the initial training set, the proportion of the initial training set in the first period, and the percentage of labeled real outliers among all the real outliers in the first period on prediction accuracy of supervised classifiers. The experiment results with seven supervised classifiers suggest that the data similarity and the percentage of labeled real outliers can significantly influence the overall prediction accuracy. This indicates that auditors should construct high-quality initial training set to ensure a high prediction accuracy in subsequent periods.
Overall, this dissertation examines the potential of using machine learning to identify suspicious observations in auditing, which aims to efficiently uncover risks that auditors do not expect. The findings of this dissertation have important implications for audit practitioners and regulators.Ph.D.Includes bibliographical reference
For dignity or honor: the effects of Kantian and utilitarian duty orientations on third-party responses to abusive supervision
No full textResearch has shown that third parties’ sense of duty can explain their reactions to the abusive supervision of coworkers, where duty has been theorized as a general concept to contrast with instrumental explanations of moral behaviors. Recently, researchers started investigating different duty orientations in organizations, including one’s perceived duty to organizational members, missions, and codes. This dissertation argues that such conceptualization of duty has primarily focused on Utilitarian duty orientation (UDO). In contrast, moral duty orientations in the deontological tradition, particularly Kantian ethics, were missing. Drawn from Kant’s discussion on the duty to not disrespect others, this dissertation proposed a new construct termed Kantian duty orientation (KDO), defined as the employees' volition to refrain from disrespecting others. Based on individual differences in duty orientations, I construct a model that contrasts the effects of Kantian-based and Utilitarian-based duty orientations on the perception of observed abusive supervision and the subsequent emotions (anger and empathy) and behaviors (supervisor deviance and coworker support). Study 1 documents the development of the Kantian Duty Orientation (KDO) scale, with results supporting its initial reliability and validity. Studies 2 and 3 (N=503 and 390) tested the theoretical model by contrasting the effects of KDO and UDO. The results from both field survey studies indicate a negative relationship between the two duty orientations and observed abusive supervision, which, in turn, leads to negative relationships with emotional and behavioral responses. This dissertation highlights how moral duty shapes the complex interactions among third parties, victims, and aggressors in organizational settings. In the interest of establishing moral motives, previous third-party studies have generally assumed or created situations in which the third party has little stake in what they observe, thereby making standing up to the aggressor less challenging. Since both KDO and UDO are deeply situated in organizations, the complex relationships that the third party has with the aggressor and the victim may account for these results. This argument is consistent with the theory and research on organizational silence and the challenges of promoting voice in organizations. The dissertation’s limitations, future research directions, and theoretical and practical implications were also discussed.Ph.D.Includes bibliographical reference