23671 research outputs found
Sort by
Application of ensemble learning models in computer-aided diagnosis of skin diseases
No full textThis is an open access article under the CC BY license.Computer-aided diagnosis (CADx) systems often involve error-prone and/or manual processes. Recent studies show that various machine learning models have the potential to improve the accuracy of CADx systems. However, existing models suffer from low prediction accuracy. In this work, we present research findings to improve the effectiveness of CADx systems for detecting skin diseases by adopting optimized ensemble machine learning models. The investigation encompasses the exploration of three popular classification methods: linear discriminant analysis (LDA), support vector machine (SVM), and convolutional neural network (CNN); two customized CNN models: LeNet-5 and ResNet; and an ensemble model of CNN with SVM. The ensemble CNN-SVM model is optimized using techniques such as feature aggregation and weight adjustments. Skin lesion images from Kaggle’s Human Against Machine 10000 (HAM10000) are used to train and test all classification models. Through rigorous experiments, the results highlight the compelling efficacy of the ensemble CNN-SVM model, unveiling heightened accuracy of up to 92% (from ResNet accuracy of 88%, CNN accuracy of 85%, SVM accuracy of 83%, LeNet-5 accuracy of 77%, and LDA accuracy of 75%). The models are tested on another dataset from Kaggle’s Melanoma Skin Cancer Dataset of 10000 Images; new results follow a similar trend to those using the HAM10000 dataset. The outcome of this work has profound implications for artificial intelligence (AI) accelerated engineering applications in advancing the effectiveness of skin disease treatment through diagnosis systems. © The Author(s) 2025
Exploring the impact of autism and gender in online fan communities
No full textPresented to the 21st Annual Symposium on Graduate Research and Scholarly Projects (GRASP) held at the Rhatigan Student Center, Wichita State University, April 11, 2025.Research completed in the Department of Communication Sciences and Disorders, College of Health Professions.INTRODUCTION: This study explores the relationship between autism, gender, and engagement with online fan communities and fan-created transformative works. Deficits associated with autism in the areas of social communication, imagination, and narrative skills (American Psychiatric Association, 2013) have the potential to limit participation in fan communities. However, some have argued that elements of these communities may play to autistic individuals' strengths while removing barriers to social interaction and creative expression (Browning & Veit, 2022; Peeples et al., 2018). This interaction may be influenced by gender differences in autism (Boorse et al., 2019; Mazurek et al., 2012; McFayden et al., 2019). Very little research has investigated autistic individuals' participation in online fan communities and none has investigated the role of gender.
PURPOSE: This study takes the first step towards filling that gap by exploring how autistic and non-autistic people of multiple genders engage with online fan communities.
METHODS: This study utilized a survey that was distributed across multiple fandom platforms. Responses (N = 149, age 18-71, M = 28.55, SD = 8.08) included demographic characteristics, diagnosis, and participation in fan activities.
RESULTS: A chi-square test of independence revealed no significant difference in gender or sex distribution by autism diagnostic status in online fan communities, with females most prevalent across all diagnostic groups. Two-way MANOVA revealed no significant effect of diagnostic status or gender on participation in creative, analytical, narrative, or interactive activities. Two-way repeated measures MANOVA revealed significantly greater preferences for system-oriented genres in original fiction and significantly greater preferences for character-oriented and transformative genres in fanfiction across all diagnostic groups and genders.
CONCLUSION: In online fan communities, autistic people do not show reduced social engagement, creativity, or literacy, or avoidance of social cognition, fiction, or imagination. Findings may offer clinicians and educators new strength- and interest-based avenues for facilitating literacy, communication, and social skill development.Graduate School, Academic Affairs, University Librarie
Forward Together: A monthly newsletter, August 2025
No full text‘Student Centered, Innovation Driven’: Our story in print -- ‘Student Centered, Innovation Driven’ is available now -- University leadership hosts 2025 Fall Address -- Wichita State shares class of spring 2025 graduates and honors students -- AfterShocks win TBT Championship in front of record number of fans -- Wichita Biomedical Campus marks massive milestone with beam signing event -- Shocker Athletics returns for another season with renewed vigor and new facilities -- Shane and Kianne Prill invest in Shocker Athletics and the community -- Rick and Barbara Hemphill amplify community spirit with season ticket donation -- From classroom to career: WSU engineering student powers up her career through paid internship -- Mechanical engineering student finds confidence, community with internship at Shuttlewagon -- WSU senior takes next career step through paid internship with the ATF -- Wichita State student gains career clarity through Homeland Security internship -- WSU senior prepares for aviation workforce with NIAR internship -- Applied learning opportunity took WSU student Liam Crawford behind the scenes of NCAA basketball tournamen
Advanced Education Program in General Dentistry graduates 2016-2017
No full textSchool composite: students included in composite: Daniel Cepeda, Daniel Sucher, Emily Ruthven, Kelli Steinlage, Parth Patel, Richie Gallant.
Group photo: from left to right: Dr. Elledge, Richie Gallant, Kelli Steinlage, Emily Ruthven, Daniel Cepeda, Parth Patel, (in back Daniel Sucher).Digitized by University Libraries' Technical Services Institutional Repository & Digitization group.Personal and non-profit use only. Contact [email protected] if you have any questions
Being ghosted: determining the progenitor genomes and biogeographic origin of the invasive fern giant salvinia
No full textClick on the DOI link to access this article at the publishers website (may not be free).Giant salvinia (Salvinia molesta, Salviniaceae) is an invasive fern species infesting aquatic habitats in over 30 countries. It is a polyploid and thought to be of hybrid origin. While a variety of evidence suggests that this hybrid origin occurred in South America, its initial discovery in Zimbabwe sets up the alternative scenario of an origin in Africa or even Eurasia. DNA sequences from over 400 nuclear genes were obtained from a set of native and invasive S. molesta samples, as well as from specimens representing all extant Salvinia species. Phylogenetic analyses of these data suggest giant salvinia originated in South America from two ‘ghost’ progenitor taxa that are extinct or extant but highly cryptic and unknown to science. These analyses also indicate that considerable genetic diversity exists within S. molesta, a surprising finding given this species’ reported lack of sexual reproduction. Alternative hypotheses for the source of this diversity and the management implications for each scenario are discussed. © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2025.Wichita State University Department of Biological Sciences; University of Louisiana at Lafayette Department of Biology; National Science Foundation, NSF, (OIA 1920858); National Science Foundation, NSFThis work was supported by National Science Foundation award OIA 1920858 to EMS and JBB, by the Wichita State University Department of Biological Sciences, and by the University of Louisiana at Lafayette Department of Biology
Neutron detection with GAGG crystals using double pulse methods
No full textThesis (M.S.)-- Wichita State University, College of Liberal Arts and Sciences, Dept. of Mathematics, Statistics, and PhysicsGadolinium Aluminum Gallium Garnet, also known as GAGG for short, is a scintillating crystal used in the neutrino detector for the collaboration project. While the ability of this material to be able to detect and be used as a gamma and charged particle detector is apparent, its use as a neutron detector has been relatively unexplored specifically using a multi-pulse method with events from the neutron capture within the gadolinium of the crystal itself. In this paper, I will explore GAGG as a possible neutron detector with both simulations and lab studies, using this multi-pulse technique and present my findings and analysis on its viability as a possible multi-use detector
High-current-density electrosynthesis of formate from captured CO2 solution by MOF-derived bismuth nanosheets
No full textClick on the DOI link to access this article at the publishers website (may not be free).Greenhouse gas emissions present a significant challenge to humanity, and utilizing renewable electricity to convert emitted CO2 into value-added products offers a promising solution; however, traditional CO2 capture and regeneration processes remain energy-intensive, restricting the overall system efficiency and decarbonization efficacy. In this study, an advanced direct reduction of captured CO2 with large current densities for formate electrosynthesis was demonstrated without the need for CO2 regeneration or compression. The bismuth nanosheet (DRM-BiNS) was synthesized by direct reduction of a Bi-based MOF, representing a new class of catalytic materials with a large surface area and interconnected pores, suitable for the direct reduction of captured CO2. By seamlessly combining experimentation and simulation, insights into the structure-parameter-performance relation were acquired in a flow cell setting, including critical membrane-electrode distance, cell orientation, and pumping flow rate. Important flow-cell components, such as catholyte volume, electrode substrate, membrane choice, and ionomer type, were also carefully examined to enhance the cell performance. In sharp contrast to prior studies limited to current densities below 20 mA/cm2 in bicarbonate-based captured CO2 solutions, this work demonstrates a remarkable current density of 300 mA/cm2 with an FE to formate comparable to the case with gas-fed CO2 reduction. Moreover, the process sustained an FE above 50% at a high current density of 500 mA/cm2. The DRM-BiNS catalyst exhibited outstanding selectivity, activity, and stability, significantly outperforming oxide-derived bismuth nanosheets (OD-BiNS) in captured CO2 reduction. These findings offer critical insights into the development of sustainable and scalable CO2 utilization technologies. © 2025 Elsevier B.V.Established Program to Stimulate Competitive Research, EPSCoR, (2316481, 2316482); Established Program to Stimulate Competitive Research, EPSCoR; National Science Foundation, NSF, (2219162, 2219172); National Science Foundation, NSF; U.S. Department of Energy, FEA, (DE-SC0025376); U.S. Department of Energy, FEAThis research of this work was supported by the U.S. NSF agency through both ECO-CBET program (2219162 & 2219172) and EPSCoR RII-Track-2 FEC program (2316481 & 2316482), and the U.S. DOE agency (DE-SC0025376). W.L. is grateful to his Herbert L. Stiles Professor fellowship. S.G. appreciates the Sam Bloomfield Chair Professor endowment. T.L. acknowledges the fruitful discussion with Mohammad Albloushi, Xiaopeng Liu, and Rod Alexei De Guzman. The authors also acknowledge Tracey P. Stewart for TEM analyses
Approaches to training speech-language pathologists to work with people with aphasia: A systematic review
No full textPurpose: The aim of this systematic review was to analyze the published research on training approaches for preparing current and future speech-language pathologists (SLPs) to work effectively with people with aphasia (PWA). The review addresses key questions regarding the described training approaches, their key features, research quality, and efficacy.
Method: The review was conducted in accordance with Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines and registered with International Prospective Register of Systematic Reviews (CRD42024453742). The literature search spanned nine databases. Eligibility criteria included peer-reviewed publications in English that reported on training programs involving current or future SLPs working with PWA. Studies were assessed for research quality using the American Speech-Language-Hearing Association's (ASHA's) levels of evidence framework. Narrative synthesis was used to identify key features in the training programs.
Results: A total of 920 citations were identified, with 50 studies meeting the inclusion criteria for analysis. Included studies represented a broad range of training approaches, research designs, and research quality. The 50 studies fell into ASHA levels of evidence Ib, IIa, IIb, III, and IV. Statistical meta-analysis was not possible because of variability in research design and outcome measures, but the studies revealed statistically significant findings relevant to the question of what makes training effective for preparing clinicians to work with people with aphasia.
Conclusions: The current literature related to training of clinicians to work with PWA is heterogeneous in approach, outcome metrics, and methodological quality. There is evidence supporting several recommendations for training clinicians including integrating direct interactions with PWA during training, combining didactic and experiential learning, and incorporating reflective practices. Overall, the review highlights the need for well-described evidence-based training standards for speech-language pathology students working with PWA. Future research should aim to develop and validate comprehensive training guidelines to improve care quality for individuals with aphasia
Validation of PCL-5 symptom validity indices in a Cross-Cultural forensic sample
No full textClick on the DOI link to access this article at the publishers website (may not be free).Introduction: Three symptom validity indices have recently been developed for the PTSD Checklist for DSM-5 (PCL-5). To date, these validity indices have been examined in North American research and clinical samples, generally with promising results. The current study aimed to cross-validate the symptom validity indices in a cross-cultural forensic sample. Method: Examinees (N = 79) were Balkan (Macedonian, Kosovar, and Serbian) contractors previously employed at United States military bases in Afghanistan and Iraq. Examinees claimed posttraumatic stress disorder (PTSD) secondary to alleged adverse experiences, and they were pursuing Federal Workers’ Compensation claims for PTSD under the auspices of the Defense Base Act. In this study, validity status was determined via outcome on the Inventory of Problems-29. Results: There were no significant differences between most demographic and background variables when groups were divided by validity status. Conversely, scores on all validity tests were significantly different between the group of examinees who were likely credibly presenting and the group that was likely noncredibly responding; medium to large effect sizes were observed. Area under the curve statistics ranged from.73 to.77. Sensitivity rates ranged from.33 to.47 when specificity was held at.90 or higher. Conclusions: The findings converge well with prior research results, extending the use of PCL-5 symptom validity indices to a cross-cultural forensic sample. © 2025 Informa UK Limited, trading as Taylor & Francis Group
