Health inequalities amplified by the COVID-19 pandemic have disproportionately affected racialized and equity-deserving communities across Canada. In the Municipality of Peel, existing data, while limited, illustrates that individuals from racialized and equity-deserving communities continue to suffer, receive delayed care, and die prematurely. In response to these troubling statistics, grassroots community advocacy has called on health systems leaders in Peel to work with community and non-profit organizations to address the critical data and infrastructure gaps that hinder addressing the social determinants of health in the region. To support these advocacy efforts, we used a community-based participatory research approach to understand how we might build a data collection ecosystem across sectors, alongside community residents and service providers, to accurately capture the data about the social determinants of health. This approach involved developing a community engagement council, defining the problem with the community, mapping what data is actively collected and what is excluded, and understanding experiences of sociodemographic data collection from community members and service providers. Guided by community voices, our study focused on sociodemographic data collection in the primary care context and identified which service providers use and collect these data, how data are used in their work, the facilitators and barriers to data use and collection. Additionally, we gained insight into how sociodemographic data collection could be respectful, safe, and properly governed from the perspectives of community members. From this study, we identify a set of eight recommendations for sociodemographic data collection and highlight limitations. This foundational community-based work will inform future research in establishing data governance in partnership with diverse and equity-deserving communities.

Understanding the health effects of disasters is critical for effective preparedness, response, recovery, and mitigation. However, research is negatively impacted by both the limited availability of disaster data and the difficulty of identifying and utilizing disaster-specific and health data sources relevant to disaster research and management. In response to numerous requests from disaster researchers, emergency managers, and operational response organizations, 73 distinct data sources at the intersection of disasters and health were compiled and categorized. These data sources generally cover the entire United States, address both disasters and health, and are available to researchers at little or no cost. Data sources are described and characterized to support improved research and guide evidence-based decision making. Current gaps and potential solutions are presented to improve disaster data collection, utilization, and dissemination.

BACKGROUND: Chronic stress is highly prevalent in the German population. It has known adverse effects on mental health, such as burnout and depression. Known long-term effects of chronic stress are cardiovascular disease, diabetes, and cancer.
OBJECTIVE: This study aims to derive an interpretable multiclass machine learning model for predicting chronic stress levels and factors protecting against chronic stress based on representative nationwide data from the German Health Interview and Examination Survey for Adults, which is part of the national health monitoring program.
METHODS: A data set from the German Health Interview and Examination Survey for Adults study including demographic, clinical, and laboratory data from 5801 participants was analyzed. A multiclass eXtreme Gradient Boosting (XGBoost) model was constructed to classify participants into 3 categories including low, middle, and high chronic stress levels. The model\'s performance was evaluated using the area under the receiver operating characteristic curve, precision, recall, specificity, and the F1-score. Additionally, SHapley Additive exPlanations was used to interpret the prediction XGBoost model and to identify factors protecting against chronic stress.
RESULTS: The multiclass XGBoost model exhibited the macroaverage scores, with an area under the receiver operating characteristic curve of 81%, precision of 63%, recall of 52%, specificity of 78%, and F1-score of 54%. The most important features for low-level chronic stress were male gender, very good general health, high satisfaction with living space, and strong social support.
CONCLUSIONS: This study presents a multiclass interpretable prediction model for chronic stress in adults in Germany. The explainable artificial intelligence technique SHapley Additive exPlanations identified relevant protective factors for chronic stress, which need to be considered when developing interventions to reduce chronic stress.

BACKGROUND: In health care, diagnosis codes in claims data and electronic health records (EHRs) play an important role in data-driven decision making. Any analysis that uses a patient\'s diagnosis codes to predict future outcomes or describe morbidity requires a numerical representation of this diagnosis profile made up of string-based diagnosis codes. These numerical representations are especially important for machine learning models. Most commonly, binary-encoded representations have been used, usually for a subset of diagnoses. In real-world health care applications, several issues arise: patient profiles show high variability even when the underlying diseases are the same, they may have gaps and not contain all available information, and a large number of appropriate diagnoses must be considered.
OBJECTIVE: We herein present Pat2Vec, a self-supervised machine learning framework inspired by neural network-based natural language processing that embeds complete diagnosis profiles into a small real-valued numerical vector.
METHODS: Based on German outpatient claims data with diagnosis codes according to the International Statistical Classification of Diseases and Related Health Problems, 10th Revision (ICD-10), we discovered an optimal vectorization embedding model for patient diagnosis profiles with Bayesian optimization for the hyperparameters. The calibration process ensured a robust embedding model for health care-relevant tasks by aggregating the metrics of different regression and classification tasks using different machine learning algorithms (linear and logistic regression as well as gradient-boosted trees). The models were tested against a baseline model that binary encodes the most common diagnoses. The study used diagnosis profiles and supplementary data from more than 10 million patients from 2016 to 2019 and was based on the largest German ambulatory claims data set. To describe subpopulations in health care, we identified clusters (via density-based clustering) and visualized patient vectors in 2D (via dimensionality reduction with uniform manifold approximation). Furthermore, we applied our vectorization model to predict prospective drug prescription costs based on patients\' diagnoses.
RESULTS: Our final models outperform the baseline model (binary encoding) with equal dimensions. They are more robust to missing data and show large performance gains, particularly in lower dimensions, demonstrating the embedding model\'s compression of nonlinear information. In the future, other sources of health care data can be integrated into the current diagnosis-based framework. Other researchers can apply our publicly shared embedding model to their own diagnosis data.
CONCLUSIONS: We envision a wide range of applications for Pat2Vec that will improve health care quality, including personalized prevention and signal detection in patient surveillance as well as health care resource planning based on subcohorts identified by our data-driven machine learning framework.

Information, as the most elusive subject, is central to all forms of thought, governance, economic structure, science, and society. Regulation of information, especially within the healthcare field, is proving to be a difficult task globally, given the lack of a qualitative framework and understanding of the concept and properties of information (or data) itself. The presentation of the overall qualitative framework, comprising a qualitative analysis of information, data, and knowledge, will be valuable and of great assistance in delineating regulatory, ethical, and strategic trajectories. In addition, this framework provides insights (and answers) regarding (1) data privacy and protection; (2) delineations between information, data, and knowledge based on the important notion of trust; (3) a structured approach to establishing the necessary conditions for an open society and system, and the maintenance of said openness, based on the work of Karl Popper and Georg Wilhelm Friedrich Hegel; (4) an active agent approach that promotes autonomy and freedom and protects the open society; and (5) a data governance mechanism based on the work of Friedrich Hayek, which structures the current legal-ethical-financial and social society. This is insightful for questions relating to the extent of rights and duties, the extent of biological bodies and freedom, and the structure of relations in distributed networked systems. There is great value offered in this framework; furthermore, it provides critical insights and thoughts about (and uncovers the interplay between) academic culture, politics, science, society, and societal decay. Note that, in line with the ideas expressed in this manuscript, such as incorporation of personal experience (thereby mending the Kantian and Cartesian gap), a first-person perspective will be used, where relevant.

We aimed to assess salivary and seroprevalence of Toxoplasma immunoglobulins in risky populations and evaluate drug docking targeting TgERP. A cross-sectional study was conducted in Alexandria University hospitals\' outpatient clinics. 192 participants were enrolled from September 2022 to November 2023. Anti-Toxoplasma IgG and IgM were determined in serum and saliva by ELISA. An in-Silico study examined TgERP\'s protein-protein interactions (PPIs) with pro-inflammatory cytokine receptors, anti-inflammatory cytokine, cell cycle progression regulatory proteins, a proliferation marker, and nuclear envelope integrity-related protein Lamin B1. Our findings revealed that anti-T. gondii IgG were detected in serum (66.1%) and saliva (54.7%), with 2.1% of both samples were positive for IgM. Salivary IgG had 75.59% sensitivity, 86.15% specificity, 91.40% PPV, 64.40% NPP, 79.17% accuracy and fair agreement with serum IgG. On the other hand, the sensitivity, specificity, PPV, NPV, and accuracy in detecting salivary IgM were 75.0%, 99.47%, 75.0%, 99.47%, and 98.96%. AUC 0.859 indicates good discriminatory power. Examined synthetic drugs and natural products can target specific amino acids residues of TgERP that lie at the same binding interface with LB1 and Ki67, subsequently, hindering their interaction. Hence, salivary samples can be a promising diagnostic approach. The studied drugs can counteract the pro-inflammatory action of TgERP.

There is an increasing desire to study neurodevelopmental disorders (NDDs) together to understand commonalities to develop generic health promotion strategies and improve clinical treatment. Common data elements (CDEs) collected across studies involving children with NDDs afford an opportunity to answer clinically meaningful questions. We undertook a retrospective, secondary analysis of data pertaining to sleep in children with different NDDs collected through various research studies. The objective of this paper is to share lessons learned for data management, collation, and harmonization from a sleep study in children within and across NDDs from large, collaborative research networks in the Ontario Brain Institute (OBI). Three collaborative research networks contributed demographic data and data pertaining to sleep, internalizing symptoms, health-related quality of life, and severity of disorder for children with six different NDDs: autism spectrum disorder; attention deficit/hyperactivity disorder; obsessive compulsive disorder; intellectual disability; cerebral palsy; and epilepsy. Procedures for data harmonization, derivations, and merging were shared and examples pertaining to severity of disorder and sleep disturbances were described in detail. Important lessons emerged from data harmonizing procedures: prioritizing the collection of CDEs to ensure data completeness; ensuring unprocessed data are uploaded for harmonization in order to facilitate timely analytic procedures; the value of maintaining variable naming that is consistent with data dictionaries at time of project validation; and the value of regular meetings with the research networks to discuss and overcome challenges with data harmonization. Buy-in from all research networks involved at study inception and oversight from a centralized infrastructure (OBI) identified the importance of collaboration to collect CDEs and facilitate data harmonization to improve outcomes for children with NDDs.

Reproducibility has become a requirement in the hard sciences, and its adoption is gradually extending to the digital humanities. The FAIR criteria and the publication of data papers are both indicative of this trend. However, the question that arises is whether the strict prerequisites of digital reproducibility serve only to exclude digital humanities from broader humanities scholarship. Instead of adopting a binary approach, an alternative method acknowledges the unique features of the objects, inquiries, and techniques of the humanities, including digital humanities, as well as the social and historical contexts in which the concept of reproducibility has developed in the human sciences. In the first part of this paper, I propose to examine the historical and disciplinary context in which the concept of reproducibility has developed within the human sciences, and the disciplinary struggles involved in this process, especially for art history and literature studies. In the second part, I will explore the question of reproducibility through two art history research projects that utilize various computational methods. I argue that issues of corpus, method, and interpretation cannot be separated, rendering a procedural definition of reproducibility impractical. Consequently, I propose the adoption of \'post-computational reproducibility\', which is based on FAIREST criteria as far as digital corpora are concerned (FAIR + Ethics and Expertise, Source mention + Time-Stamp), but extended to include further sources that confirm computational results with other non-computational methodologies.

Using information-theoretic quantities in practical applications with continuous data is often hindered by the fact that probability density functions need to be estimated in higher dimensions, which can become unreliable or even computationally unfeasible. To make these useful quantities more accessible, alternative approaches such as binned frequencies using histograms and k-nearest neighbors (k-NN) have been proposed. However, a systematic comparison of the applicability of these methods has been lacking. We wish to fill this gap by comparing kernel-density-based estimation (KDE) with these two alternatives in carefully designed synthetic test cases. Specifically, we wish to estimate the information-theoretic quantities: entropy, Kullback-Leibler divergence, and mutual information, from sample data. As a reference, the results are compared to closed-form solutions or numerical integrals. We generate samples from distributions of various shapes in dimensions ranging from one to ten. We evaluate the estimators\' performance as a function of sample size, distribution characteristics, and chosen hyperparameters. We further compare the required computation time and specific implementation challenges. Notably, k-NN estimation tends to outperform other methods, considering algorithmic implementation, computational efficiency, and estimation accuracy, especially with sufficient data. This study provides valuable insights into the strengths and limitations of the different estimation methods for information-theoretic quantities. It also highlights the significance of considering the characteristics of the data, as well as the targeted information-theoretic quantity when selecting an appropriate estimation technique. These findings will assist scientists and practitioners in choosing the most suitable method, considering their specific application and available data. We have collected the compared estimation methods in a ready-to-use open-source Python 3 toolbox and, thereby, hope to promote the use of information-theoretic quantities by researchers and practitioners to evaluate the information in data and models in various disciplines.

Technology has a major impact on the way nurses work. Data-driven technologies, such as artificial intelligence (AI), have particularly strong potential to support nurses in their work. However, their use also introduces ambiguities. An example of such a technology is AI-driven lifestyle monitoring in long-term care for older adults, based on data collected from ambient sensors in an older adult\'s home. Designing and implementing this technology in such an intimate setting requires collaboration with nurses experienced in long-term and older adult care. This viewpoint paper emphasizes the need to incorporate nurses and the nursing perspective into every stage of designing, using, and implementing AI-driven lifestyle monitoring in long-term care settings. It is argued that the technology will not replace nurses, but rather act as a new digital colleague, complementing the humane qualities of nurses and seamlessly integrating into nursing workflows. Several advantages of such a collaboration between nurses and technology are highlighted, as are potential risks such as decreased patient empowerment, depersonalization, lack of transparency, and loss of human contact. Finally, practical suggestions are offered to move forward with integrating the digital colleague.