Health Care Sciences Research Articles

A general framework for developing computable clinical phenotype algorithms.

To present a general framework providing high-level guidance to developers of computable algorithms for identifying patients with specific clinical conditions (phenotypes) through a variety of approaches, including but not limited to machine learning and natural language processing methods to incorporate rich electronic health record data. Drawing on extensive prior phenotyping experiences and insights derived from 3 algorithm development projects conducted specifically for this purpose, our team with expertise in clinical medicine, statistics, informatics, pharmacoepidemiology, and healthcare data science methods conceptualized stages of development and corresponding sets of principles, strategies, and practical guidelines for improving the algorithm development process. We propose 5 stages of algorithm development and corresponding principles, strategies, and guidelines: (1) assessing fitness-for-purpose, (2) creating gold standard data, (3) feature engineering, (4) model development, and (5) model evaluation. This framework is intended to provide practical guidance and serve as a basis for future elaboration and extension.

Understanding and Addressing AI Hallucinations in Healthcare and Life Sciences

Purpose: This paper investigates the phenomenon of "AI hallucinations" in healthcare and life sciences, where large language models (LLMs) produce outputs that, while coherent, are factually incorrect, irrelevant, or misleading. Understanding and mitigating such errors is critical given the high stakes of accurate and reliable information in healthcare and life sciences. We classify hallucinations into three types input-conflicting, context-conflicting, and fact-conflicting and examine their implications through real-world cases. Methodology: Our methodology combines the Fact Score, Med-HALT, and adversarial testing to evaluate the fidelity of AI outputs. We propose several mitigation strategies, including Retrieval-Augmented Generation (RAG), Chain-of-Verification (CoVe), and Human-in-the-Loop (HITL) systems, to enhance model reliability. Findings: As artificial intelligence continues to permeate various sectors of society, the issue of hallucinations in AI-generated text poses significant challenges, especially in contexts where precision and reliability are paramount. This paper has delineated the types of hallucinations commonly observed in AI systems input-conflicting, context-conflicting, and fact-conflicting and highlighted their potential to undermine trust and efficacy in critical domains such as healthcare and legal proceedings. Unique contribution to theory, policy and practice: This study's unique contribution lies in its comprehensive analysis of AI hallucinations' types and impacts and the development of robust controls that advance theoretical understanding, practical application, and policy formulation in AI deployment. These efforts aim to foster safer, more effective AI integration across healthcare and life sciences sectors

Translation, Cross-Cultural Adaptation, and Validation of Measurement Instruments: A Practical Guideline for Novice Researchers.

Cross-cultural validation of self-reported measurement instruments for research is a long and complex process, which involves specific risks of bias that could affect the research process and results. Furthermore, it requires researchers to have a wide range of technical knowledge about the translation, adaptation and pre-test aspects, their purposes and options, about the different psychometric properties, and the required evidence for their assessment and knowledge about the quantitative data processing and analysis using statistical software. This article aimed: 1) identify all guidelines and recommendations for translation, cross-cultural adaptation, and validation within the healthcare sciences; 2) describe the methodological approaches established in these guidelines for conducting translation, adaptation, and cross-cultural validation; and 3) provide a practical guideline featuring various methodological options for novice researchers involved in translating, adapting, and validating measurement instruments. Forty-two guidelines on translation, adaptation, or cross-cultural validation of measurement instruments were obtained from "CINAHL with Full Text" (via EBSCO) and "MEDLINE with Full Text". A content analysis was conducted to identify the similarities and differences in the methodological approaches recommended. Bases on these similarities and differences, we proposed an eight-step guideline that includes: a) forward translation; 2) synthesis of translations; 3) back translation; 4) harmonization; 5) pre-testing; 6) field testing; 7) psychometric validation, and 8) analysis of psychometric properties. It is a practical guideline because it provides extensive and comprehensive information on the methodological approaches available to researchers. This is the first methodological literature review carried out in the healthcare sciences regarding the methodological approaches recommended by existing guidelines.

Role of artificial intelligence (Google bard) in morphological, histopathological, and radiological image identifications: Objective Structured Practical Examination (OSPE) type-based performance.

To evaluate the role of artificial intelligence (Google Bard) in figures, scans, and image identifications and interpretations in medical education and healthcare sciences through an Objective Structured Practical Examination (OSPE) type of performance. The OSPE type of question bank was created with a pool of medical sciences figures, scans, and images. For assessment, 60 figures, scans and images were selected and entered into the given area of the Google Bard to evaluate the knowledge level. The marks obtained by Google Bard in brain structures, morphological and radiological images 7/10 (70%); bone structures, radiological images 9/10 (90%); liver structure and morphological, pathological images 4/10 (40%); kidneys structure and morphological images 2/7 (28.57%); neuro-radiological images 4/7 (57.14%); and endocrine glands including the thyroid, pancreas, breast morphological and radiological images 8/16 (50%). The overall total marks obtained by Google Bard in various OSPE figures, scans, and image identification questions were 34/60 (56.7%). Google Bard scored satisfactorily in morphological, histopathological, and radiological image identifications and their interpretations. Google Bard may assist medical students, faculty in medical education and physicians in healthcare settings.

Thermal Welding of Liquids.

Welding of thermoplastics is a common practice in many industrial sectors, but it has yet to be realized with fluids. Here, the thermal welding of liquids by using the assembly and jamming of nanoparticle surfactants (NPSs) at liquid-liquid interfaces is reported. By fine-tuning the dynamic interaction strength within NPSs, the interfacial activity of NPSs, as well as the binding energy of NPSs to the interface can be precisely controlled, leading to a dynamic exchange of NPSs, maximizing the reduction in the interfacial energy. With NPSs jammed at the interface, the structures of liquids can be manipulated to complex geometries by applying an external force and, due to the temperature responsiveness of NPSs, when bringing liquids into contact and heating the system, welding of liquids can be achieved. This work provides a straightforward strategy for the construction of modular all-liquid fluidics, opening up numerous opportunities in fields like biotechnology, healthcare, and materials science.

Data science in healthcare: techniques, challenges and opportunities

Data science in healthcare: techniques, challenges and opportunities

Data science techniques to gain novel insights into quality of care: a scoping review of long-term care for older adults

Background: The increase in powerful computers and technological devices as well as new forms of data analysis such as machine learning have resulted in the widespread availability of data science in healthcare. However, its role in organizations providing long-term care (LTC) for older people LTC for older adults has yet to be systematically synthesized. This analysis provides a state-of-the-art overview of 1) data science techniques that are used with data accumulated in LTC and for what specific purposes and, 2) the results of these techniques in researching the study objectives at hand. Methods: A scoping review based on guidelines of the Joanna Briggs Institute. PubMed and Cumulative Index to Nursing and Allied Health Literature (CINAHL) were searched using keywords related to data science techniques and LTC. The screening and selection process was carried out by two authors and was not limited by any research design or publication date. A narrative synthesis was conducted based on the two aims. Results: The search strategy yielded 1,488 studies: 27 studies were included of which the majority were conducted in the US and in a nursing home setting. Text-mining/natural language processing (NLP) and support vector machines (SVMs) were the most deployed methods; accuracy was the most used metric. These techniques were primarily utilized for researching specific adverse outcomes including the identification of risk factors for falls and the prediction of frailty. All studies concluded that these techniques are valuable for their specific purposes. Discussion: This review reveals the limited use of data science techniques on data accumulated in or by LTC facilities. The low number of included articles in this review indicate the need for strategies aimed at the effective utilization of data with data science techniques and evidence of their practical benefits. There is a need for a wider adoption of these techniques in order to exploit data to their full potential and, consequently, improve the quality of care in LTC by making data-informed decisions.

Associations between change in BMI and the risk of hypertension and dyslipidaemia in people receiving integrase strand-transfer inhibitors, tenofovir alafenamide, or both compared with other contemporary antiretroviral regimens: a multicentre, prospective observational study from the RESPOND consortium cohorts

Integrase strand-transfer inhibitors (INSTIs) and tenofovir alafenamide have been associated with weight gain in several clinical trials and observational cohorts. However, whether weight gain associated with INSTIs and tenofovir alafenamide confers a higher risk of weight-related clinical events is unclear. We aimed to assess whether changes in BMI differentially increase hypertension or dyslipidaemia risk in people with HIV receiving INSTIs, tenofovir alafenamide, or both versus other contemporary regimens. This multicentre, prospective observational study analysed prospective data from RESPOND, an international consortium of HIV cohorts for which recruitment began in 2017 and is still ongoing from HIV clinics and hospitals in 37 European countries and Australia. Participants were eligible if they were aged 18 years or older, receiving INSTI-containing antiretroviral therapy (ART) regimens or a contemporary non-INSTI, did not have hypertension or dyslipidaemia at baseline, and had baseline and at least two follow-up BMI, lipid, and blood pressure measurements. We excluded participants without baseline CD4 or HIV RNA results and those receiving non-ART medications associated with weight changes, including antipsychotics and mood stabilisers, corticosteroids, insulin, and insulin secretagogues. They were followed up from baseline until the earliest hypertension or dyslipidaemia event, their last visit, or Dec 31, 2021, whichever was earlier. The primary outcomes were incidence of hypertension and dyslipidaemia, for which we used multivariable Poisson regression adjusted for time-updated BMI to determine unadjusted and adjusted incidence rate ratios (IRRs) of hypertension and dyslipidaemia in people receiving INSTIs, tenofovir alafenamide, or both, and tested for interaction between time-updated ART regimen and BMI. Of the 35 941 RESPOND participants, 9704 (7327 [75·5 %] male and 2377 [24·5%] female) were included in the hypertension analysis and 5231 (3796 [72·6%] male and 1435 [27·4%] female) were included in the dyslipidaemia analysis. In the univariable model, hypertension was more common in individuals receiving an INSTI with tenofovir alafenamide (IRR 1·70, 95% CI 1·54-1·88) or an INSTI without tenofovir alafenamide (1·41, 1·30-1·53) compared with those receiving neither INSTIs nor tenofovir alafenamide. Adjustment for time-updated BMI and confounders attenuated risk in participants receiving an INSTI with (IRR 1·48, 1·31-1·68) or without (1·25, 1·13-1·39) tenofovir alafenamide. Similarly, dyslipidaemia was more common in participants using tenofovir alafenamide with an INSTI (IRR 1·24, 1·10-1·40) and tenofovir alafenamide alone (1·22, 1·03-1·44) than in participants using neither INSTI nor tenofovir alafenamide. Adjustment for BMI and confounders attenuated the risk in participants receiving tenofovir alafenamide with an INSTI (adjusted IRR 1·21, 1·07-1·37), whereas the risk in those receiving tenofovir alafenamide alone became non-significant (1·15, 0·96-1·38). The associations between increasing BMI and risk of hypertension and dyslipidaemia did not differ between participants receiving different ART regimens (pinteraction=0·46 for hypertension; pinteraction=0·31 for dyslipidaemia). Although residual confounding cannot be entirely excluded, the use of INSTIs was associated with incident hypertension, and the use of tenofovir alafenamide was associated with dyslipidaemia, with the latter association partly mediated by weight gain. These results reiterate the need for hypertension and dyslipidaemia screening in people with HIV. The CHU St Pierre Brussels HIV Cohort, The Austrian HIV Cohort Study, The Australian HIV Observational Database, The AIDS Therapy Evaluation in the Netherlands national observational HIV cohort, The Brighton HIV Cohort, The National Croatian HIV Cohort, The EuroSIDA cohort, The Frankfurt HIV Cohort Study, The Georgian National AIDS Health Information System, The Nice HIV Cohort, The ICONA Foundation, The Modena HIV Cohort, The PISCIS Cohort Study, The Swiss HIV Cohort Study, The Swedish InfCare HIV Cohort, The Royal Free HIV Cohort Study, The San Raffaele Scientific Institute, The University Hospital Bonn HIV Cohort, The University of Cologne HIV Cohort, Merck Life Sciences, ViiV Healthcare, and Gilead Sciences.

Artificial Intelligence in Health, Health Care, and Biomedical Science: An AI Code of Conduct Principles and Commitments Discussion Draft

Artificial Intelligence in Health, Health Care, and Biomedical Science: An AI Code of Conduct Principles and Commitments Discussion Draft

How can electronic health records serve as a tool for clinical trials?

AbstractIncreasing volumes of information are being collected via electronic health records and there is growing multi-site expertise in utlising these for research. This emerging field of healthcare data science is not only concerned with the technical challenges associated with complex data, but also with the need for effective security and governance in the use of sensitive information with robust structures for stakeholder input and guidance. To date, most of the focus has been on supporting observational cohort studies nested within clinical records data - particularly investigating research questions around treatment response and course/prognosis. It is likely that electronic health records will become increasingly integrated with clinical trials, providing opportunities for pre-study feasibility scoping, targeted recruitment, and enhanced and extended follow-up. In addition, there is interest in emulated trials using routine data. For mental health data science, key challenges lie in the quality and quantity of data made accessible, with a particular need for natural language processing to derive structured data from extensive clinical text. Many of the challenges have been addressed for observational research, creating exciting prospects for a transformed trials landscape.Disclosure of InterestR. Stewart Grant / Research support from: Janssen, GSK, Takeda

SELECTED ASPECTS OF THE USE OF ARTIFICIAL INTELLIGENCE IN PUBLIC ADMINISTRATION

Artificial intelligence grows ever more impactful in various social and economic domains. Public administration is among increasingly prominent fields undergoing technology-driven transformations. Today’s administration bodies, local and central both, are facing the challenge of accommodating new outlooks brought by artificial intelligence. This dynamic metamorphosis entails both benefits and difficulties worth investigating. The primary research problem here is the use of artificial intelligence in public administration and potential future changes in this regard. The main research goal of the article is to analyse the benefits of employing artificial intelligence in public administration followed by an analysis of any relevant ethical and pragmatic questions. The article further presents selected aspects of the problem, focusing on the application of data science in public health care.

A comparative study of the deep learning based model and the conventional machine learning based models in human activity recognition

Human activity recognition (HAR) has been widely studied as a research field in human behavior analysis due to its huge potential in various application domains such as health care and behavioral science. Recently, deep learning (DL) based methods have also been successfully applied to predict various human activities. This research aims at building different Python-based models to perform HAR using smartphones and calculating and comparing the accuracy of the models to select the optimal one. Four models were built to classify and predict human activities: Deep Convolutional Neural Network (DCNN), Support Vector Machine (SVM), Decision Tree (DT), and Random Forest (RF). The results of the experiments in this paper show that the Deep Convolutional Neural Network achieves an average recognition accuracy rate of 95.49%, exceeding the other three models. The underlying reason may be that Deep Convolutional Neural Network is based on a more advanced algorithm deep learning technique.

Artificial Intelligence in Health, Health Care, and Biomedical Science: An AI Code of Conduct Principles and Commitments Discussion Draft

Artificial Intelligence in Health, Health Care, and Biomedical Science: An AI Code of Conduct Principles and Commitments Discussion Draft

Exploring the Safety, Efficacy, and Bioactivity of Herbal Medicines: Bridging Traditional Wisdom and Modern Science in Healthcare

Exploring the Safety, Efficacy, and Bioactivity of Herbal Medicines: Bridging Traditional Wisdom and Modern Science in Healthcare

GraFITi: Graphs for Forecasting Irregularly Sampled Time Series

Forecasting irregularly sampled time series with missing values is a crucial task for numerous real-world applications such as healthcare, astronomy, and climate sciences. State-of-the-art approaches to this problem rely on Ordinary Differential Equations (ODEs) which are known to be slow and often require additional features to handle missing values. To address this issue, we propose a novel model using Graphs for Forecasting Irregularly Sampled Time Series with missing values which we call GraFITi. GraFITi first converts the time series to a Sparsity Structure Graph which is a sparse bipartite graph, and then reformulates the forecasting problem as the edge weight prediction task in the graph. It uses the power of Graph Neural Networks to learn the graph and predict the target edge weights. GraFITi has been tested on 3 real-world and 1 synthetic irregularly sampled time series dataset with missing values and compared with various state-of-the-art models. The experimental results demonstrate that GraFITi improves the forecasting accuracy by up to 17% and reduces the run time up to 5 times compared to the state-of-the-art forecasting models.

Abstract 1615: Balanced molecular feature selection unveils universally tumor-lineage informative methylation sites in colorectal cancer

Abstract In the era of precision medicine, performing comparative analysis over diverse patient populations is a fundamental step toward tailoring healthcare interventions. However, the critical aspect of equitably selecting molecular features across multiple patients is often overlooked. To address this challenge, we introduce FALAFL (FAir muLti-sAmple Feature seLection), a novel algorithmic approach based on integer linear programming. FALAFL is designed to bridge the gap between molecular feature selection and algorithmic fairness, ensuring a balanced selection of molecular features across different patient samples. We have applied FALAFL to the problem of selecting lineage-informative CpG sites within a cohort of 9 colorectal cancer patients subjected to low-coverage single-cell methylation sequencing. Our results demonstrate that FALAFL excels at rapidly determining the optimal set of CpG sites, which are well covered by cells across the vast majority of the patients, while ensuring that each patient contributes a similar number of sites to the final selection. An analysis of the FALAFL-selected CpG sites reveals that their lineage informativeness exhibits a strong association across diverse patient profiles, suggesting the non-stochasticity and universality of methylation changes in tumor evolution. Furthermore, these universally lineage-informative sites exhibit subclonal level alterations in the colorectal cancer primary tumors and are strongly associated with lineage-specific gene expression patterns. By integrating equity considerations into the molecular feature selection process, FALAFL is poised to propel equitable healthcare data science practices and advance our understanding of cancer and other complex diseases. Citation Format: Xuan Cindy Li, Yuelin Liu, Alejandro A. Schäffer, Stephen M. Mount, Süleyman C. Sahinalp. Balanced molecular feature selection unveils universally tumor-lineage informative methylation sites in colorectal cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 1615.

COVID-19 vaccination and potential predictors for vaccination among students in Bulgarian medical universities

University students, especially those in healthcare sciences, are of particular importance in COVID-19 vaccine views since their future careers will allow them to educate citizens regarding vaccination. The study examined COVID-19 immunisation coverage, behaviours, and attitudes towards COVID-19 vaccines among students in Bulgarian medical universities. A prospective cross-sectional study was undertaken in September 2021- January 2022. Demographics, university programme, year of study, general vaccine attitudes and behaviour, and personal experience with COVID-19 and vaccines were collected. The chi-square test was used to test for associations and binominal logistic regression to identify possible predictors for vaccination. A total of 3050 students with a median age of 22 years, predominantly female participated. Three-quarters of the students (73.5%) have been vaccinated against COVID-19. An association was found between COVID-19 vaccination and influenza vaccination in the previous season (χ2 = 67.15, p < 0.001) as well as the intention to receive the flu vaccine in the current season (χ2 = 142.83, p < 0.001). Vaccine hesitation mainly stemmed from fear of side effects and safety concerns, but non-vaccinated students voiced them more often. Respondents who were less likely to be vaccinated considered themselves to have: (a) limited access to sufficient information to allow them to make an informed decision to vaccinate against COVID-19; (b) lack of public awareness and education campaigns about COVID-19 vaccines; (c) insufficient information about COVID-19 vaccines during university education that would allow them to make an informed decision to vaccinate; and (d) had a positive COVID-19 diagnosis. Informational initiatives that emphasise vaccine safety will boost COVID-19 immunisation coverage.

Staff perception on including students with physical disabilities at a South African university.

International and local policy frameworks on disability promote inclusive higher education practices for students with disabilities (SWD). However, the actual application of these frameworks concerning students with physical disabilities (SWPD) in any School of Health Care Sciences (SHCS) is uncertain in South African universities. This study aimed to explore the perceptions of academic and admission staff on the inclusion of SWPD in SHCS at a South African university. The study was carried out at a University of Health Sciences in South Africa. A qualitative study in which respondents (n = 12) were interviewed in depth about their perceptions on the inclusion of SWPD in the SHCS. Thematic analysis was used in the data assessment. The results revealed three main themes: policy discourse, environmental effects on inclusion and SWPD enrolment. Respondents reported the lack of a disability inclusion policy and disability unit to support SWD in general. The respondents also noted that there were environmental challenges that could potentially affect the inclusion of SWPD in SHCS study programmes. Respondents also indicated that there was no SWPD enrolment as the university's current inclusion and/or quota system does not include SWD. The findings of the study showed a lack of disability inclusion policy, environmental challenges and lack of SWPD enrolment. Based on the study findings, it can be concluded that inclusion of SWPD at this university may be negatively influenced. The study findings contribute to the field of disability and the inclusion of SWPD in higher education institutions (HEIs).

Bridging, Mapping, and Addressing Research Gaps in Health Sciences: The Naqvi-Gabr Research Gap Framework.

Innovations pertaining to the ever-evolving needs of the medical and healthcare sciences remain constant. This creates a gap between the rationalized needs of the study and the proposed research question. However, classifying, identifying, and addressing these research gaps require a systematic and precise structured map. Using the Medical Subject Heading (MeSH) terms "Research Gaps" AND "Healthcare" AND "Framework" in MEDLINE, Scopus, and CINAHL databases with the filters yielded no relevant literature. Therefore, this review aims to fill this practical and clinical knowledge gap by developing the Naqvi-Gabr Research Gap Framework through critical synthesis based on extensive research on medical and healthcare research gaps. Fourteen research gaps are distributed for allocation as per the healthcare delivery system approach: developing new treatments or prevention strategies, improving diagnostic tools and techniques, addressing health disparities, and improving access to healthcare services. This structured framework determines the strategic mapping of research gaps corresponding to the nature of the research. The identification and classification of the appropriate research gap led to precise and concise conclusions corresponding to the research process proposed in this study. Hence, the Naqvi-Gabr Research Gap Framework is a valuable tool for determining the potential application of gaps by researchers, policymakers, and other stakeholders with a productive address.

Application of Complementary Therapy as an Effort to Mitigate Health Crisis from Malnutrition Risk Factors in Vulnerable Groups

Introduction Malnutrition, a critical global health issue, continues to affect vulnerable populations despite advances in healthcare and nutrition science, highlighting the need for integrated intervention strategies that combine traditional nutritional support with complementary therapies. Objective: This research aimed to assess the effectiveness of a holistic intervention combining nutritional strategies and complementary therapies in mitigating malnutrition risks within vulnerable families, specifically those with toddlers or elderly members. Method: Conducted at a community health center in Purwomartani, Kalasan, Sleman, Yogyakarta, the study involved 43 families, with 24 having toddlers and the remainder with elderly members. Employing purposive sampling, participants were selected based on their regular attendance at the health center, literacy, and willingness to participate. The intervention, implemented through two sessions in October-November 2023, included measures to enhance access to nutritious food, provide nutritional supplements, offer educational outreach, monitor nutritional status, and administer personalized nutritional therapy. Result: Data, recorded on a scale of 0-5 for each indicator and sub-indicator, demonstrated significant improvements across all parameters. The results indicated positive shifts in food security, healthcare access, socioeconomic status, dietary diversity, and micronutrient intake. Conclusion: The integrated approach effectively addressed malnutrition risks, showcasing tangible improvements in key indicators. This study contributes valuable insights into holistic strategies for mitigating malnutrition in vulnerable populations.