32584 research outputs found
Sort by
Spartan Daily, December 03, 2025
No full textVolume 165, Issue 41https://scholarworks.sjsu.edu/spartan_daily_2025/1085/thumbnail.jp
Prevalence of Exposures and Moral Injury in First Responders
No full textObjectives: Workers in high-stakes occupations, such as first responders, are at risk of exposure to potentially morally injurious events (PMIEs) and moral injury, yet research with first responders has been scarce. This study aimed to assess the frequency of exposure to PMIEs, the prevalence of moral injury, and the correlation between moral injury and other mental health symptoms. Methods: In this cross-sectional study, firefighters, emergency medical technicians (EMTs), and paramedics working for a large urban Fire Department in California were invited to complete an online, confidential survey using validated scales to assess exposure to PMIEs, moral injury symptoms, and symptoms of posttraumatic stress disorder, anxiety, depression, alcohol use, and burnout between February 21, 2024, and April 16, 2024. We calculated item frequencies and correlation and reliability analysis on each scale separately and between the total scores from all measures using Pearson, polyserial, and polychoric methods. Reliability was assessed using Guttman\u27s lambda and Cronbach\u27s alpha. Results: Participants (N = 292) endorsed a range of PMIEs, including exposure by commission (48.6%), omission (48.6%), and witnessing (80.8%). Among those that reported PMIE exposure and moral injury symptoms (N = 147), 18.4% met the threshold for clinically meaningful moral injury. Moral injury symptoms were strongly correlated with each of the mental health measures (r = 0.49-0.59), with the exception of alcohol use (r = 0.08). Conclusion: Exposure to PMIEs was common in these professional first responders, and a substantial proportion of participants reported clinically meaningful moral injury symptoms. These data fill an important gap and provide information that can help with moral injury assessment and treatment for first responders
Hemispheric Asymmetry of Phase Partition in Mixed-Phase Clouds Based on Near Global-Scale Airborne Observations
No full textMixed-phase clouds contribute to substantial uncertainties in global climate models due to their complex microphysical properties. Former model evaluations almost exclusively rely on satellite observations to assess cloud phase distributions globally. This study investigated mixed-phase cloud properties using near global-scale in situ observation data sets from 14 flight campaigns in combination with collocated output from a global climate model. The Southern Hemisphere (SH) shows significantly higher occurrence frequencies and higher mass fractions of supercooled liquid water than Northern Hemisphere (NH) based on observations at 0.2 and 100 km horizontal scales. Such hemispheric asymmetry is not captured by the model. The model also consistently overestimates liquid water content (LWC) in all cloud phases but shows ice water content (IWC) biases that vary with phase. Key processes contributing to model biases in phase partition can be identified through the combination of evaluation of phase frequency, liquid mass fraction, LWC and IWC
Deep Learning for the Early Detection of Invasive Ductal Carcinoma in Histopathological Images: Convolutional Neural Network Approach With Transfer Learning
No full textBackground: Invasive ductal carcinoma (IDC) is considered the most common form of breast cancer, accounting for a significant percentage of mortality worldwide. Therefore, its early detection is vital to further improve patients’ outcomes and survival rates. However, conventional diagnostic methods in the form of manual histopathological examinations are time-consuming, subjective, and prone to errors. Therefore, there is an urgent need to develop automated solutions for accurate IDC detection in histopathology images to assist pathologists in clinical decision-making. Objective: We aim to develop and validate a convolutional neural network (CNN) model for early detection of IDC by analyzing histopathological images. The specific objectives are designing a deep learning–based technique for automated detection of IDC, assessing its performance compared to traditional diagnostic methods, and evaluating its utility in a clinical setup for early breast cancer diagnosis. These methods will be available to practitioners in underdeveloped countries via an open-source application. Methods: The dataset for the research included 277,524 publicly available histopathological images from Kaggle, comprising both IDC-positive and IDC-negative images. About 71.6% of images were IDC-positive (class 0), while 28.4% were IDC-negative (class 1). Since our data are unbalanced, we created a weighted loss function to overcome the class imbalance problem. Further development was based on a CNN using the approach of transfer learning with a pretrained architecture called Visual Geometry Group to uplift feature extraction so that performance may improve; hence, images were preprocessed and normalized to perform augmentation with robustness. The model was developed using a split of 80% for training and 20% for testing. Model performance was measured for accuracy, sensitivity, specificity, precision, recall, and F1-score in the confusion matrix and classification report. Results: From our CNN base model, we obtained an accuracy of 89% on the test set. Later, the base model was used with a weighted loss function to balance the class weights, giving a lower accuracy of 86% on the test set. Data augmentation was performed but did not improve the results. To deal with the class imbalance effectively, we performed transfer learning with a pretrained model, which gave an accuracy of 90% on the test set. Conclusions: The CNN-based model thus showed accuracy and reliability for early detection of IDC from histopathological images. This technique will potentially act as an efficient and accurate assistant tool for pathologists, contributing to the early diagnosis of breast cancer and improving clinical outcomes. This paper provides an important contribution toward refining the performance of this model and widening its applications in a clinical setting by integrating it with other diagnostic techniques for better outcomes
Homelessness Following Jail Exit Among Previously Housed Individuals
No full textIncarceration is a recognized risk factor for homelessness. However, most research focuses on the relationship between homelessness and prison incarceration. Jail incarceration is more common compared to prison incarceration, but little data exists on its impact on housing. The objective of this study is to examine the occurrence of housing loss after jail incarceration among individuals without prior evidence of homelessness and the associated risk of reincarceration. In this retrospective cross-sectional study, we identified adults without evidence of homelessness who became unhoused within 6 months of jail incarceration. We compare pre-incarceration emergent and urgent health and social services utilization among housed and unhoused individuals, as well as the risk of reincarceration. Data are from the San Francisco (SF) Department of Public Health Coordinated Care Management System linked with SF City and County criminal justice data during fiscal years 2015–2018. We find that a quarter (25.1%) of individuals lost housing after jail incarceration, with a median incarceration length of 4 days in both the housed and unhoused groups. Compared to those without evidence of housing loss, more unhoused individuals had pre-incarceration substance use and mental health diagnoses and related service utilization. Unhoused individuals had 1.9 greater odds of reincarceration. In conclusion, we find that a significant number of individuals had evidence of housing loss after even a short jail incarceration; behavioral health diagnoses were more common among this group. Housing loss was associated with subsequent reincarceration. Given our findings, jail re-entry programs would benefit from incorporating housing assistance and housing loss mitigation strategies
Dynamics and lipid membrane coupling of the RAS-RAF complex revealed via multiscale simulations
No full textTo gain molecular and mechanistic insights into initiation of the RAS-RAF signaling cascade, we developed and used a combination of multiscale simulation and experimental approaches. The influence and impact of the membrane on RAS and RAF proteins is a factor we are just beginning to understand and appreciate in more detail. Molecular simulation is an ideal methodology to further study this complicated relationship between the membrane and associated proteins. Our previous work using Multiscale Machine-learned Modeling Infrastructure investigated different lipid compositions solely around the KRAS4b protein and the interplay between protein behavior and these membrane environments. Multiscale Machine-learned Modeling Infrastructure uses machine learning to couple adjacent simulation scales and has been efficiently scaled across some of the world\u27s largest high-performance computers. Recently, we have expanded this multiresolution framework to include the all-atom simulation scale and to incorporate the RAF RBDCRD domains. Here, we present the overall analysis results from this new simulation campaign comprising a mixture of RAS and RAF RBDCRD proteins. Approximately 35,000 coarse-grained and 10,000 all-atom molecular dynamics simulations were completed, sampled from a variety of protein/lipid composition configurations that were generated from a micron-scale continuum simulation containing hundreds of copies of the proteins. Our studies suggest that orientations of the RAS-RBDCRD complex on the membrane occupy distinct configurational states, and the spatial patterns of lipid arrangements around these different protein states are unique to each state. The extent and size of lipid “fingerprints” imposed on the membrane by the RAS-RBDCRD protein complex are significantly larger than observed for just the RAS protein on its own. These protein complexes strongly associate, but we do not observe statistically significant preferred protein-protein orientations. These observations indicate that spatial colocalization of RAS-RBDCRD proteins in the same vicinity may be assisted by specific membrane environments, acting to increase the probability of signaling complex formation
The Role of Advocacy and Occupational Justice in Increasing the Accessibility of Parks for Parents with Disabilities
No full textThe purpose of this doctoral capstone project was to address ongoing accessibility barriers experienced by adults with disabilities, particularly those in caregiving roles, in public parks and playgrounds in the greater San Diego region. Although many parks meet the minimum requirements of the Americans with Disabilities Act (1990), research shows that caregivers with disabilities continue to encounter obstacles that limit safe, meaningful participation in community spaces (Dalpra, 2022; Fernelius, 2017; Firkin et al., 2024; Movahed et al., 2023; Ogletree et al., 2020; Prellwitz & Skar, 2007). These persistent gaps indicate a need for stronger advocacy and clearer public awareness of real-world accessibility conditions. This project included assessing seven parks and playgrounds using the mPARCs tool, collaborating with mentors at Through the Looking Glass and the Adaptive Parenting Project, and creating educational materials that translate accessibility issues for a general audience. Key outcomes included two PSA style videos and Google Maps photo documentation, which has received more than fourteen thousand views and provides a sustainable resource for residents, service providers, and community planners. This project reinforces the role of occupational therapy in advocacy by illuminating common access barriers and promoting more inclusive park and playground design for adults with disabilities and their families
Detecting Malicious Encrypted Network Traffic Using Deep Learning and CNN-Based Feature Representations
No full textEncrypted HTTPS traffic now dominates the Internet, and malware increasingly uses TLS to conceal command-and-control activity. Since payloads cannot be inspected, detection must rely on metadata such as TLS handshake fields and certificate attributes, which prior work has shown can still reveal malicious behavior. This research evaluates whether malicious HTTPS connections can be detected using only metadata from Zeek logs. Using the CTU-SME-11 dataset, we build a reproducible preprocessing pipeline and a 33-feature connection-level representation capturing flow statistics, TLS behavior, and certificate validity characteristics. We evaluate XGBoost, multilayer perceptrons, and several CNN variants - including 1D and 2D grid-based embeddings - using a stratified capture-level split and 5-fold capture-aware cross-validation to prevent leakage. Results show strong discriminative performance, with XGBoost achieving the highest ROC-AUC and PR-AUC, and CNN-based models, particularly an 8×8 architecture, achieving the strongest malicious-class F1-scores. These findings show that metadata-based models can accurately detect encrypted malicious traffic and motivate future work on generalization, calibration and explainability
Detection and Mitigation for Poisoned Textual Datasets
No full textData poisoning occurs in various datasets; however, it is more challenging to detect poisoning in textual datasets compared to image datasets. The focus of this paper is to determine how to detect poisoning in textual datasets. We focused on four poisoning attacks, mislabeling, injection, targeted, and non targeted attacks. Recurrent Neural Networks (RNN), Support Vector Machine (SVM), and Resource Scheme Multilinear Regression (RSMLR) are used for detecting poisoning. A custom RNN class containing an encoder and decoder was created for the RNN. 10% of the data set was used for the SVM to determine whether the rest of the dataset was poisoned. The target column was processed separately from the entire dataset for the RSMLR. To improve each model, a threshold equation was used to determine the poisoned that needed to be flagged. Using the best parameter values, the models are used for a Federated Learning (FL) for multiple passes and shuffling. Based on the experimental results, the use of the RNN and SVM together in shuffling yields the best results for poisoning attacks. The RSMLR had the poorest performance but performed well when detecting poisoning in shuffled datasets. Based on the model shuffling experiment, the models yield average accuracies of 41% for mislabeling datasets, 92% for injection datasets and 68% for targeted datasets. For the Non Targeted attacks, both RNN and SVM yield accuracies of 100%