South Dakota State University
Public Research Access Institutional Repository and Information ExchangeNot a member yet
30309 research outputs found
Sort by
Implications of assuming common within-source distributions and their effect on evidence interpretation
No full textWithin the forensic science community, there is a need for a statistically rigorous determination of whether an exclusionary difference exists; this determination is integral to the Kirkian Two-Stage approach to evidence interpretation (Parker 1966). If the known source is not excluded as the source of the questioned object in the first stage, then the examiner must determine the rate at which competing alternative sources are excluded. Current methods are typically constructed to ensure the same false exclusion rate for each source. For example, in ASTM glass standards E2927-16e1 and E2330-19, an exclusionary difference occurs if any of the standardized differences between the measured trace element concentrations is greater than a fixed threshold of four. However, if the algorithm’s score function has a distribution that varies by source, then the corresponding thresholds will need to vary as well.
In this work, we review strategies for identifying when the within-source distribution of scores varies between sources; methods for estimating thresholds; remedial approaches such as pooling subsets of sources together; and implications of this type of variability among the sources to the Kirkian and likelihood ratio (LR) approaches. We illustrate these methods with example data from traces such as glass and improvised explosive device components.
Although the focus is on the Two-Stage approach, this work is also important for LR-based methods due to the need to estimate a likelihood function from just a few observations from a specified source. The discussed remedial methods also apply to the LR paradigm for evidence interpretation
Evaluating the Effectiveness of OPTN Regulations in Organ Transplantation
No full textOrgan transplantation is often the only therapy available to patients with end-stage organ failure. In 2024, over 48,000 transplants were performed in the US, while another 67,000 candidates joined the national waitlist. The Organ Procurement & Transplantation Network (OPTN) manages the national transplant system and oversees the operation of transplant programs (i.e., hospitals that perform transplants). The OPTN has historically monitored transplant program performance using 1-year post-transplant survival metrics. Recently, the OPTN introduced a new set of criteria with additional evaluation metrics to improve performance monitoring. Our study investigates the effectiveness and possible limitations of these regulations using a simulated environment. As a first step, we aim to quantify the prevalence of ‘false flagging’, i.e., the likelihood that a well-performing transplant program is incorrectly flagged and vice-versa. We develop a Monte Carlo simulation framework comprising three phases: (i) preliminary phase, where we estimate patients’ pre- and post-transplant mortality; (ii) simulation phase, where we assign patients to virtual transplant programs and generate simulated patient outcomes; and (iii) evaluation phase, where we assess the accuracy of the new metrics. Because true program-level risks are unobservable, we artificially designate certain programs as high-risk and simulate patient outcomes using preliminary regression models. The models are calibrated using national data on adult first-time kidney transplant candidates waitlisted between 2012 and 2019. Our findings aim to substantiate the OPTN’s efforts to refine performance metrics, prioritize patient safety, and promote increased utilization of deceased-donor organs.
KEYWORDS: organ transplantation, Monte Carlo simulation, health policy, data science in healthcare
Deep Learning for Forensic Identification of Source
No full textWe used contrastive neural networks to learn a similarity score within the framework of the forensic common-but-unknown source problem. Similarity scores are often used for the interpretation of forensic evidence. Utilizing the NBIDE dataset of 144 spent cartridge casings, this work tested the ability of contrastive neural networks to learn useful similarity scores. The results obtained by contrastive learning were directly compared to a standard forensic statistics algorithm, Congruent Matching Cells (CMC). When trained on the E3 dataset of 2967 spent cartridge casings, contrastive networks outperformed the CMC algorithm. Generally held principles in deep learning would suggest that a larger training dataset would yield even more effective similarity scores. We also considered the effects of varying the neural network architecture; specifically, altering the network\u27s width or depth. This work was in part motivated by the potential to use similarity scores learned via contrastive networks for standard evidence interpretation methods such as likelihood ratios
Effects of Essential Oils Alone or in Combination with Monensin in an In Vitro Model of Ruminal Acidosis
Full text linkThe objective of this study was to evaluate the effect of essential oils and/or monensin on the concentration of volatile fatty acids (VFA), ammonia, and pH in a ruminal acidosis in vitro model. Red Angus steers (n = 4; BW = 435 ± 9 kg) with ruminal and duodenal cannulas were allocated in a 4 × 4 Latin square design. Treatments were 1) Fed no essential oils (RB) or monensin sodium (MON:CON); 2) Fed RB at a rate of 14 g daily with no Mon (RB+); 3) Fed no RB and fed Mon at 400 mg daily (MON+); 4) Fed RB and Mon [RB+MON+] at the same levels of inclusion. The basal diet (TMR) consisted of dry rolled corn (67%), grass hay (10%), DDGS (20%), and vitamin and mineral premix (3%). Feed additives (RB or MON) were dosed in the rumen through the cannulas in gelatin capsules to guarantee the steers were receiving the targeted dose. Each period lasted 28-d, consisting of a 14-d adaptation and 14-d collection period, where in-vitro ruminal acidosis challenges were performed on d 14 and 26. Briefly, rumen fluid from each steer was placed into three 250-ml glass flasks (30 mL of rumen fluid per flask) containing McDougall buffer (120 ml per flask), thus resulting in a triplicate sample per treatment (rumen fluid:buffer ratio of 1:4 per flask). In addition to the buffer, each flask contained 7.5 g of substrate (reconstituted TMR). Flasks were incubated for 36 h at 39 ºC, subsamples (4 mL) were taken every 4 hours and pH was measured. Samples were stored at -20 ºC for further analysis of VFA and ammonia. Data were analyzed as repeated measures using the Glimmix procedure of SAS with fixed effects of treatment, day, time, and their interactions. Animal and period were considered random effects. Significance was set at P ≤ 0.05. A treatment × day interaction (P ≤ 0.001) was observed for concentrations of all VFA. Acetate decreased from d 14 to d 26 for CON, with similar levels observed for the other treatments. Concentrations of propionate increased from d 14 to d 26 for CON and RBMON and decreased for MON. Butyrate concentrations decreased from d 14 to d 26 for RB+ and RB+MON+ treatments. Isobutyrate increased from d 14 to d 26 for MON and decreased for RB+MON+. Concentrations of valerate decreased from d 14 to d 26 for MON, RB+, and RB+MON+, while the opposite was observed for CON. Isovalerate concentrations decreased from d 14 to d 26 for CON and increased for MON. Acetate:Propionate ratio remained similar between d 14 and d 26 for MON, RB, and RBMON, and decreased for CON from d 14 to d 26. Sampling time affected concentrations of all VFA, ammonia, and pH (P \u3c 0.001). In vitro ruminal pH decreased (P \u3c 0.001) from d 14 to d 26, and as expected pH decreased (P \u3c 0.001) from 0 to 36h. No differences in pH were observed between treatments, with an average of 5.69. Discrepancies between the present in vitro findings and those reported in a parallel in vivo study suggest that the absence of a treatment effect may be attributed to limitations inherent to the in vitro model, such as gas accumulation, rapid pH decline, and the absence of ruminal wall interactions
Testing the Effect of Source and Environment on Native Seed Germination to Inform Seed-Based Restoration of Grasslands
Full text linkGrasslands are one of the most endangered ecosystems on the planet, and they provide a variety of important ecosystem services that people rely on. Losses of grasslands can be mitigated through restoration. Seed-based restoration is a regularly used practice due to its practicality and cost-effectiveness. Planting non-local seed that may not be suitable for its destination environment can have consequences in plant establishment and subsequent restoration success. This experiment tested the response (final germination and germination rate) of seed from 11 perennial tallgrass species sourced from 3 different locations to different germination temperature conditions. Seeds from three sources of each species were exposed to different temperature treatments (cool, intermediate, and warm) and germination was tracked over 42 days. Dependent variables of final germination and germination rate were significantly impacted by the source by temperature interaction for seven and six species respectively. Final germination and germination rate for all other species were significantly impacted by main effects of source or temperature, but not the interaction. The majority (nine out of 11) of the species tested displayed at least one type of interaction between source and environment on germination indicating potential for local adaptation. If the potential for local adaptation leads to a home site advantage where non-local seeds perform more poorly than local seed, restoration practitioners may need to select seed sources carefully. However, restoration practitioners might be able to take advantage of species such as Helenium autumnale and Sorghastrum nutans, because they did not exhibit signs of potential local adaptation (no significant source by environment interaction) by planting them at wider ranges. This study has implications for helping land managers make informed restoration decisions
Minimizing Fasting Requirements to Maximize Patient Satisfaction Prior to a Scheduled Cardiac Catheterization
Full text linkBackground: Patients are often kept NPO (nil per os), or “nothing by mouth,” longer than necessary before cardiac catheterization procedures. Local Problem: Patients scheduled for cardiac catheterization procedures receive instructions to begin fasting at midnight on the day of their procedure regardless of their scheduled procedural start time. This directive results in patient dissatisfaction as patients are fasting longer than current guideline recommendation. Methods: The facility updated its order set to align with the American Society of Anesthesiologists (ASA) fasting guidelines for scheduled cardiac catheterizations. To evaluate patient satisfaction, a 6-item survey was administered to 169 individuals undergoing cardiac catheterization. The pre-intervention group (n=116) and postintervention group (n=53) were independent samples. An independent T-test was used to analyze differences in patient satisfaction. Results: A statistically significant difference was observed in both patient satisfaction (p= 0.046) and reported fasting duration (p= 0.0248). The pre-intervention group reported an average fasting burden score of 16.92 and an average fasting time of 12.59 hours, compared to the post-intervention group, which had an average fasting burden score of 15.62 and fasting duration of 11.08 hours. No cases of aspiration were reported with either fasting protocol. Conclusion: The results demonstrate that following ASA guidelines can improve patient satisfaction without increasing the risk of aspiration
Dairy and Food Science Student Newsletter, July 1, 2025
Full text linkhttps://openprairie.sdstate.edu/dairy_student-news/1009/thumbnail.jp
Capstone II Project : MLB
Full text linkClustering player statistics provides new perspectives on identifying elite Major League Baseball (MLB) players. This paper explores the use of k-means clustering combined with principal component analysis (PCA) to segment players based on various performance metrics such as batting average, on-base percentage, wins above replacement, and many others. The analysis examines over a century of MLB data, dating from 1900 to 2023, to uncover meaningful groups and highlight key differences between elite and non-elite players. Results demonstrate how modern statistical techniques can go beyond traditional metrics to provide insights into player performance. These findings have implications for player evaluation, team strategy, and advancing the use of data in sports analytics
