The value of standards for health datasets in artificial intelligence-based applications.

Artificial intelligence as a medical device is increasingly being applied to healthcare for diagnosis, risk stratification and resource allocation. However, a growing body of evidence has highlighted the risk of algorithmic bias, which may perpetuate existing health inequity. This problem arises in part because of systemic inequalities in dataset curation, unequal opportunity to participate in research and inequalities of access. This study aims to explore existing standards, frameworks and best practices for ensuring adequate data diversity in health datasets. Exploring the body of existing literature and expert views is an important step towards the development of consensus-based guidelines. The study comprises two parts: a systematic review of existing standards, frameworks and best practices for healthcare datasets; and a survey and thematic analysis of stakeholder views of bias, health equity and best practices for artificial intelligence as a medical device. We found that the need for dataset diversity was well described in literature, and experts generally favored the development of a robust set of guidelines, but there were mixed views about how these could be implemented practically. The outputs of this study will be used to inform the development of standards for transparency of data diversity in health datasets (the STANDING Together initiative).

Impact of luminescent MoSe2 quantum dots on activity of trypsin under different pH environment.

It is vital that a straightforward detection approach for trypsin should be developed as it is important diagnostic tool for a number of diseases. Herein, the impact of luminescent MoSe2 quantum dots on trypsin activity under different pH environment has been studied. Addition of trypsin to MoSe2 quantum dots enhanced the fluorescence of quantum dots whereas quantum dots resulted in quenching of fluorescence of trypsin. The quenching behavior at various pH and temperature was examined and revealed that the MoSe2-trypsin complex stabilized through the electrostatic interactions. The obtained negative values of zeta potential of the complex -0.11 mV, -0.30 mV and -0.59 mV for pH 6.0,7.6 and 9.0 respectively confirmed the stability of the complex. The separation between the donor and acceptor atoms in energy transfer mechanism was found to decrease (1.48 nm to 1.44 nm to 1.30 nm) with increasing value of pH. It was also evident that trypsin retained its enzyme activity in the trypsin-MoSe2 complex and under different pH environment. The Vant Hoff plot from quenching revealed 1 binding site for quantum dots by trypsin for all pH of buffer solution. The complex formation of trypsin-MoSe2 quantum dots was verified for the first time using fluorescence spectroscopy and it revealed that tryspin form complex with MoSe2 quantum dots through electrostatic interactions. Our results revealed that the MoSe2 quantum dots stabilized and sheltered the active sites of trypsin, which was likely the cause of the increased bioavailability of MoSe2 quantum dots in enzymes.

Assessment of machine learning algorithms in national data to classify the risk of self-harm among young adults in hospital: A retrospective study.

BACKGROUND: Self-harm is one of the most common presentations at accident and emergency departments in the UK and is a strong predictor of suicide risk. The UK Government has prioritised identifying risk factors and developing preventative strategies for self-harm. Machine learning offers a potential method to identify complex patterns with predictive value for the risk of self-harm. METHODS: National data in the UK Mental Health Services Data Set were isolated for patients aged 18-30 years who started a mental health hospital admission between Aug 1, 2020 and Aug 1, 2021, and had been discharged by Jan 1, 2022. Data were obtained on age group, gender, ethnicity, employment status, marital status, accommodation status and source of admission to hospital and used to construct seven machine learning models that were used individually and as an ensemble to predict hospital stays that would be associated with a risk of self-harm. OUTCOMES: The training dataset included 23 808 items (including 1081 episodes of self-harm) and the testing dataset 5951 items (including 270 episodes of self-harm). The best performing algorithms were the random forest model (AUC-ROC 0.70, 95%CI:0.66-0.74) and the ensemble model (AUC-ROC 0.77 95%CI:0.75-0.79). INTERPRETATION: Machine learning algorithms could predict hospital stays with a high risk of self-harm based on readily available data that are routinely collected by health providers and recorded in the Mental Health Services Data Set. The findings should be validated externally with other real-world, prospective data. FUNDING: This study was supported by the Midlands and Lancashire Commissioning Support Unit.

Examining medical student volunteering during the COVID-19 pandemic as a prosocial behaviour during an emergency.

PURPOSE: Understanding the factors that influence prosocial behaviour during the COVID-19 pandemic is essential due to the disruption to healthcare provision. METHODS: We conducted an in-depth, mixed-methods cross-sectional survey, from 2 May 2020 to 15 June 2020, of medical students at medical schools in the United Kingdom. Data analysis was informed by Latané and Darley's theory of prosocial behaviour during an emergency. RESULTS: A total of 1145 medical students from 36 medical schools responded. Although 947 (82.7%) of students were willing to volunteer, only 391 (34.3%) had volunteered. Of the students, 92.7% understood they may be asked to volunteer; however, we found deciding one's responsibility to volunteer was mitigated by a complex interaction between the interests of others and self-interest. Further, concerns revolving around professional role boundaries influenced students' decisions over whether they had the required skills and knowledge. CONCLUSION: We propose two additional domains to Latané and Darley's theory that medical students consider before making their final decision to volunteer: 'logistics' and 'safety'. We highlight modifiable barriers to prosocial behaviour and provide suggestions regarding how the conceptual framework can be operationalized within educational strategies to address these barriers. Optimizing the process of volunteering can aid healthcare provision and may facilitate a safer volunteering process. Key messages  What is already known on this topic: There is a discrepancy between the number of students willing to volunteer during pandemics and disasters, and those who actually volunteer. Understanding the factors that influence prosocial behaviour during the current COVID-19 pandemic and future pandemics and disasters is essential. What this study adds: We expanded on Latané and Darley's theory of prosocial behaviour in an emergency and used this to conceptualize students' motivations to volunteer, highlighting a number of modifiable barriers to prosocial behaviour during the COVID-19 pandemic. How this study might affect research, practice, or policy: We provide suggestions regarding how the conceptual framework can be operationalized to support prosocial behaviours during emergencies for the ongoing COVID-19 pandemic and future crises.

Screening for Auditory Processing Difficulties in Older Adults with Hearing Impairment Using Screening Checklist for Auditory Processing in Adults.

BACKGROUND: Aging enhances changes in the central and peripheral auditory systems. It is expected that older adults population would experience auditory processing deficits. Therefore, early identification of these individuals will help in making appropriate referrals, which in turn might help in early diagnosis and management of the problem. METHODS: Fifty-five participants diagnosed with hearing impairment were screened for the existence of auditory processing difficulties using Screening Checklist for Auditory Processing for Adults-Modified 2-point rating scale. The data were collected using direct interview and telephonic interview with the participant. RESULTS: A total of 26 participants with bilateral symmetrical sensorineural hearing loss (47.3%) exhibited auditory processing difficulties. CONCLUSION: It can be understood that all older adults with hearing impairment need to undergo screening using Screening Checklist for Auditory Processing for Adults. This will further help in deciding and customizing the management options required for each older adult with bilateral symmetrical sensorineural hearing loss.

Clinical support during COVID-19: An opportunity for service and learning? A cross-sectional survey of UK medical students.

PURPOSE: Medical students providing support to clinical teams during Covid-19 may have been an opportunity for service and learning. We aimed to understand why the reported educational impact has been mixed to inform future placements. METHODS: We conducted a cross-sectional survey of medical students at UK medical schools during the first Covid-19 'lockdown' period in the UK (March-July 2020). Analysis was informed by the conceptual framework of service and learning. RESULTS: 1245 medical students from 37 UK medical schools responded. 57% of respondents provided clinical support across a variety of roles and reported benefits including increased preparedness for foundation year one compared to those who did not (p < 0.0001). However, not every individual's experience was equal. For some, roles complemented the curriculum and provided opportunities for clinical skill development, reflection, and meaningful contribution to the health service. For others, the relevance of their role to their education was limited; these roles typically focused on service provision, with few opportunities to develop. CONCLUSION: The conceptual framework of service and learning can help explain why student experiences have been heterogeneous. We highlight how this conceptual framework can be used to inform clinical placements in the future, in particular the risks, benefits, and structures.[Box: see text].

Machine learning models trained on synthetic datasets of multiple sample sizes for the use of predicting blood pressure from clinical data in a national dataset.

INTRODUCTION: The potential for synthetic data to act as a replacement for real data in research has attracted attention in recent months due to the prospect of increasing access to data and overcoming data privacy concerns when sharing data. The field of generative artificial intelligence and synthetic data is still early in its development, with a research gap evidencing that synthetic data can adequately be used to train algorithms that can be used on real data. This study compares the performance of a series machine learning models trained on real data and synthetic data, based on the National Diet and Nutrition Survey (NDNS). METHODS: Features identified to be potentially of relevance by directed acyclic graphs were isolated from the NDNS dataset and used to construct synthetic datasets and impute missing data. Recursive feature elimination identified only four variables needed to predict mean arterial blood pressure: age, sex, weight and height. Bayesian generalised linear regression, random forest and neural network models were constructed based on these four variables to predict blood pressure. Models were trained on the real data training set (n = 2408), a synthetic data training set (n = 2408) and larger synthetic data training set (n = 4816) and a combination of the real and synthetic data training set (n = 4816). The same test set (n = 424) was used for each model. RESULTS: Synthetic datasets demonstrated a high degree of fidelity with the real dataset. There was no significant difference between the performance of models trained on real, synthetic or combined datasets. Mean average error across all models and all training data ranged from 8.12 To 8.33. This indicates that synthetic data was capable of training equally accurate machine learning models as real data. DISCUSSION: Further research is needed on a variety of datasets to confirm the utility of synthetic data to replace the use of potentially identifiable patient data. There is also further urgent research needed into evidencing that synthetic data can truly protect patient privacy against adversarial attempts to re-identify real individuals from the synthetic dataset.

Ethics Principles for Artificial Intelligence-Based Telemedicine for Public Health.

The use of artificial intelligence (AI) in the field of telemedicine has grown exponentially over the past decade, along with the adoption of AI-based telemedicine to support public health systems. Although AI-based telemedicine can open up novel opportunities for the delivery of clinical health and care and become a strong aid to public health systems worldwide, it also comes with ethical risks that should be detected, prevented, or mitigated for the responsible use of AI-based telemedicine in and for public health. However, despite the current proliferation of AI ethics frameworks, thus far, none have been developed for the design of AI-based telemedicine, especially for the adoption of AI-based telemedicine in and for public health. We aimed to fill this gap by mapping the most relevant AI ethics principles for AI-based telemedicine for public health and by showing the need to revise them via major ethical themes emerging from bioethics, medical ethics, and public health ethics toward the definition of a unified set of 6 AI ethics principles for the implementation of AI-based telemedicine. (Am J Public Health. 2023;113(5):577-584. https://doi.org/10.2105/AJPH.2023.307225).

Generative adversarial networks and synthetic patient data: current challenges and future perspectives.

Artificial intelligence (AI) has been heralded as one of the key technological innovations of the 21st century. Within healthcare, much attention has been placed upon the ability of deductive AI systems to analyse large datasets to find patterns that would be unfeasible to program. Generative AI, including generative adversarial networks, are a newer type of machine learning that functions to create fake data after learning the properties of real data. Artificially generated patient data has the potential to revolutionise clinical research and protect patient privacy. Using novel techniques, it is increasingly possible to fully anonymise datasets to the point where no datapoint is traceable to any real individual. This can be used to expand and balance datasets as well as to replace the use of real patient data in certain contexts. This paper focuses upon three key uses of synthetic data: clinical research, data privacy and medical education. We also highlight ethical and practical concerns that require consideration.

Key considerations for the use of artificial intelligence in healthcare and clinical research.

Interest in artificial intelligence (AI) has grown exponentially in recent years, attracting sensational headlines and speculation. While there is considerable potential for AI to augment clinical practice, there remain numerous practical implications that must be considered when exploring AI solutions. These range from ethical concerns about algorithmic bias to legislative concerns in an uncertain regulatory environment. In the absence of established protocols and examples of best practice, there is a growing need for clear guidance both for innovators and early adopters. Broadly, there are three stages to the innovation process: invention, development and implementation. In this paper, we present key considerations for innovators at each stage and offer suggestions along the AI development pipeline, from bench to bedside.

Face mask fit hacks: Improving the fit of KN95 masks and surgical masks with fit alteration techniques.

INTRODUCTION: During the course of the COVID-19 pandemic, there have been suggestions that various techniques could be employed to improve the fit and, therefore, the effectiveness of face masks. It is well recognized that improving fit tends to improve mask effectiveness, but whether these fit modifiers are reliable remains unexplored. In this study, we assess a range of common "fit hacks" to determine their ability to improve mask performance. METHODS: Between July and September 2020, qualitative fit testing was performed in an indoor living space. We used quantitative fit testing to assess the fit of both surgical masks and KN95 masks, with and without 'fit hacks', on four participants. Seven fit hacks were evaluated to assess impact on fit. Additionally, one participant applied each fit hack multiple times to assess how reliable hacks were when reapplied. A convenience of four participants took part in the study, three females and one male with a head circumference range of 54 to 60 centimetres. RESULTS AND DISCUSSION: The use of pantyhose, tape, and rubber bands were effective for most participants. A pantyhose overlayer was observed to be the most effective hack. High degrees of variation were noted between participants. However, little variation was noted within participants, with hacks generally showing similar benefit each time they were applied on a single participant. An inspection of the fit hacks once applied showed that individual facial features may have a significant impact on fit, especially the nose bridge. CONCLUSIONS: Fit hacks can be used to effectively improve the fit of surgical and KN95 masks, enhancing the protection provided to the wearer. However, many of the most effective hacks are very uncomfortable and unlikely to be tolerated for extended periods of time. The development of effective fit-improvement solutions remains a critical issue in need of further development.

Experimental Measurement of the Size of Gaps Required to Compromise Fit of an N95 Respirator.

OBJECTIVE: The effectiveness of filtering facepiece respirators such as N95 respirators is heavily dependent on the fit. However, there have been limited efforts to discover the size of the gaps in the seal required to compromise filtering facepiece respirator performance, with prior studies estimating this size based on in vitro models. In this study, we measure the size of leak necessary to compromise the fit of N95 respirators. METHODS: Two methods were used to create a gap of specific dimensions. A set of 3D-printed resin spacers and hollow steel rods were used to generate gaps in N95 respirators while worn on 2 participants. Occupational Safety and Health Administration (OSHA) quantitative fit testing methods were used to quantify mask performance with gaps between 0.4 and 2.9-mm diameters. RESULTS: Gap size was regressed against fit factor, showing that overall, the minimum gap size to compromise N95 performance was between 1.5 mm2 and 3 mm2. CONCLUSIONS: These findings suggest the fit of a N95 respirator is compromised by gaps that may be difficult to visually detect. The study also adds to the body of evidence supporting the routine use of quantitative fit testing to ensure that masks are well-fitting.

Performing Qualitative Mask Fit Testing Without a Commercial Kit: Fit Testing Which Can Be Performed at Home and at Work.

OBJECTIVE: Qualitative fit testing is a popular method of ensuring the fit of sealing face masks such as N95 and FFP3 masks. Increased demand due to the coronavirus disease 2019 (COVID-19) pandemic has led to shortages in testing equipment and has forced many institutions to abandon fit testing. Three key materials are required for qualitative fit testing: the test solution, nebulizer, and testing hood. Accessible alternatives to the testing solution have been studied. This exploratory qualitative study evaluates alternatives to the nebulizer and hoods for performing qualitative fit testing. METHODS: Four devices were trialed to replace the test kit nebulizer. Two enclosures were tested for their ability to replace the test hood. Three researchers evaluated promising replacements under multiple mask fit conditions to assess functionality and accuracy. RESULTS: The aroma diffuser and smaller enclosures allowed participants to perform qualitative fit tests quickly and with high accuracy. CONCLUSIONS: Aroma diffusers show significant promise in their ability to allow individuals to quickly, easily, and inexpensively perform qualitative fit testing. Our findings indicate that aroma diffusers and homemade testing hoods may allow for qualitative fit testing when conventional apparatus is unavailable. Additional research is needed to evaluate the safety and reliability of these devices.

Lipidomics and Redox Lipidomics Indicate Early Stage Alcohol-Induced Liver Damage.

Alcoholic fatty liver disease (AFLD) is characterized by lipid accumulation and inflammation and can progress to cirrhosis and cancer in the liver. AFLD diagnosis currently relies on histological analysis of liver biopsies. Early detection permits interventions that would prevent progression to cirrhosis or later stages of the disease. Herein, we have conducted the first comprehensive time-course study of lipids using novel state-of-the art lipidomics methods in plasma and liver in the early stages of a mouse model of AFLD, i.e., Lieber-DeCarli diet model. In ethanol-treated mice, changes in liver tissue included up-regulation of triglycerides (TGs) and oxidized TGs and down-regulation of phosphatidylcholine, lysophosphatidylcholine, and 20-22-carbon-containing lipid-mediator precursors. An increase in oxidized TGs preceded histological signs of early AFLD, i.e., steatosis, with these changes observed in both the liver and plasma. The major lipid classes dysregulated by ethanol play important roles in hepatic inflammation, steatosis, and oxidative damage. Conclusion: Alcohol consumption alters the liver lipidome before overt histological markers of early AFLD. This introduces the exciting possibility that specific lipids may serve as earlier biomarkers of AFLD than those currently being used.