Facial artificial intelligence in ophthalmology and medicine: fundamental and transformative applications.
The integration of artificial intelligence (AI) in healthcare, particularly in the domain of facial processing tasks, has witnessed substantial growth in the 21st century. However, this requires sufficient appraisal for clinicians and researchers to adequately understand nomenclature and key concepts commonly used in this field. This article aims to elucidate the diverse applications of facial processing tasks, such as facial landmark extraction, face detection, face tracking, facial expression recognition and action unit detection, and their relevance to ophthalmology and other medical specialties. The keywords 'ophthalmology', 'facial artificial intelligence', 'facial recognition' and 'periorbital measurements' were used on PubMed and Ovid, between September 2012 and September 2022, to identify and screen for eligible articles. Studies reporting on human patients in ophthalmology, plastic, maxillofacial and cosmetic surgery with ocular lesions whose facial biometrics were processed by AI and written in the English language were included. A total of 291 and 513 articles were identified on PubMed and Ovid respectively. Twenty articles were included for analysis in this literature review after duplicates, inaccessible articles and articles without full manuscripts were excluded. Although fully automated algorithms can share the workload in healthcare systems and relieve strains on manpower, rigorous testing is crucial, followed by the challenges of convincing management bodies that it would work in reality, coupled with the costs of implementing specialised functional hardware and software. While patients have a valid concern that it would reduce physical contact with clinicians, it is important for clinicians not to replace clinical decision-making with AI alone.
Chan JJH
，Leung PW
，Kilgour H
，Dervenis P
... -
《-》
Australia in 2030: what is our path to health for all?
CHAPTER 1: HOW AUSTRALIA IMPROVED HEALTH EQUITY THROUGH ACTION ON THE SOCIAL DETERMINANTS OF HEALTH: Do not think that the social determinants of health equity are old hat. In reality, Australia is very far away from addressing the societal level drivers of health inequity. There is little progressive policy that touches on the conditions of daily life that matter for health, and action to redress inequities in power, money and resources is almost non-existent. In this chapter we ask you to pause this reality and come on a fantastic journey where we envisage how COVID-19 was a great disruptor and accelerator of positive progressive action. We offer glimmers of what life could be like if there was committed and real policy action on the social determinants of health equity. It is vital that the health sector assists in convening the multisectoral stakeholders necessary to turn this fantasy into reality. CHAPTER 2: ABORIGINAL AND TORRES STRAIT ISLANDER CONNECTION TO CULTURE: BUILDING STRONGER INDIVIDUAL AND COLLECTIVE WELLBEING: Aboriginal and Torres Strait Islander peoples have long maintained that culture (ie, practising, maintaining and reclaiming it) is vital to good health and wellbeing. However, this knowledge and understanding has been dismissed or described as anecdotal or intangible by Western research methods and science. As a result, Aboriginal and Torres Strait Islander culture is a poorly acknowledged determinant of health and wellbeing, despite its significant role in shaping individuals, communities and societies. By extension, the cultural determinants of health have been poorly defined until recently. However, an increasing amount of scientific evidence supports what Aboriginal and Torres Strait Islander people have always said - that strong culture plays a significant and positive role in improved health and wellbeing. Owing to known gaps in knowledge, we aim to define the cultural determinants of health and describe their relationship with the social determinants of health, to provide a full understanding of Aboriginal and Torres Strait Islander wellbeing. We provide examples of evidence on cultural determinants of health and links to improved Aboriginal and Torres Strait Islander health and wellbeing. We also discuss future research directions that will enable a deeper understanding of the cultural determinants of health for Aboriginal and Torres Strait Islander people. CHAPTER 3: PHYSICAL DETERMINANTS OF HEALTH: HEALTHY, LIVEABLE AND SUSTAINABLE COMMUNITIES: Good city planning is essential for protecting and improving human and planetary health. Until recently, however, collaboration between city planners and the public health sector has languished. We review the evidence on the health benefits of good city planning and propose an agenda for public health advocacy relating to health-promoting city planning for all by 2030. Over the next 10 years, there is an urgent need for public health leaders to collaborate with city planners - to advocate for evidence-informed policy, and to evaluate the health effects of city planning efforts. Importantly, we need integrated planning across and between all levels of government and sectors, to create healthy, liveable and sustainable cities for all. CHAPTER 4: HEALTH PROMOTION IN THE ANTHROPOCENE: THE ECOLOGICAL DETERMINANTS OF HEALTH: Human health is inextricably linked to the health of the natural environment. In this chapter, we focus on ecological determinants of health, including the urgent and critical threats to the natural environment, and opportunities for health promotion arising from the human health co-benefits of actions to protect the health of the planet. We characterise ecological determinants in the Anthropocene and provide a sobering snapshot of planetary health science, particularly the momentous climate change health impacts in Australia. We highlight Australia's position as a major fossil fuel producer and exporter, and a country lacking cohesive and timely emissions reduction policy. We offer a roadmap for action, with four priority directions, and point to a scaffold of guiding approaches - planetary health, Indigenous people's knowledge systems, ecological economics, health co-benefits and climate-resilient development. Our situation requires a paradigm shift, and this demands a recalibration of health promotion education, research and practice in Australia over the coming decade. CHAPTER 5: DISRUPTING THE COMMERCIAL DETERMINANTS OF HEALTH: Our vision for 2030 is an Australian economy that promotes optimal human and planetary health for current and future generations. To achieve this, current patterns of corporate practice and consumption of harmful commodities and services need to change. In this chapter, we suggest ways forward for Australia, focusing on pragmatic actions that can be taken now to redress the power imbalances between corporations and Australian governments and citizens. We begin by exploring how the terms of health policy making must change to protect it from conflicted commercial interests. We also examine how marketing unhealthy products and services can be more effectively regulated, and how healthier business practices can be incentivised. Finally, we make recommendations on how various public health stakeholders can hold corporations to account, to ensure that people come before profits in a healthy and prosperous future Australia. CHAPTER 6: DIGITAL DETERMINANTS OF HEALTH: THE DIGITAL TRANSFORMATION: We live in an age of rapid and exponential technological change. Extraordinary digital advancements and the fusion of technologies, such as artificial intelligence, robotics, the Internet of Things and quantum computing constitute what is often referred to as the digital revolution or the Fourth Industrial Revolution (Industry 4.0). Reflections on the future of public health and health promotion require thorough consideration of the role of digital technologies and the systems they influence. Just how the digital revolution will unfold is unknown, but it is clear that advancements and integrations of technologies will fundamentally influence our health and wellbeing in the future. The public health response must be proactive, involving many stakeholders, and thoughtfully considered to ensure equitable and ethical applications and use. CHAPTER 7: GOVERNANCE FOR HEALTH AND EQUITY: A VISION FOR OUR FUTURE: Coronavirus disease 2019 has caused many people and communities to take stock on Australia's direction in relation to health, community, jobs, environmental sustainability, income and wealth. A desire for change is in the air. This chapter imagines how changes in the way we govern our lives and what we value as a society could solve many of the issues Australia is facing - most pressingly, the climate crisis and growing economic and health inequities. We present an imagined future for 2030 where governance structures are designed to ensure transparent and fair behaviour from those in power and to increase the involvement of citizens in these decisions, including a constitutional voice for Indigenous peoples. We imagine that these changes were made by measuring social progress in new ways, ensuring taxation for public good, enshrining human rights (including to health) in legislation, and protecting and encouraging an independent media. Measures to overcome the climate crisis were adopted and democratic processes introduced in the provision of housing, education and community development.
Backholer K
，Baum F
，Finlay SM
，Friel S
，Giles-Corti B
，Jones A
，Patrick R
，Shill J
，Townsend B
，Armstrong F
，Baker P
，Bowen K
，Browne J
，Büsst C
，Butt A
，Canuto K
，Canuto K
，Capon A
，Corben K
，Daube M
，Goldfeld S
，Grenfell R
，Gunn L
，Harris P
，Horton K
，Keane L
，Lacy-Nichols J
，Lo SN
，Lovett RW
，Lowe M
，Martin JE
，Neal N
，Peeters A
，Pettman T
，Thoms A
，Thow AMT
，Timperio A
，Williams C
，Wright A
，Zapata-Diomedi B
，Demaio S
... -
《-》
Ensuring Appropriate Representation in Artificial Intelligence-Generated Medical Imagery: Protocol for a Methodological Approach to Address Skin Tone Bias.
In medical education, particularly in anatomy and dermatology, generative artificial intelligence (AI) can be used to create customized illustrations. However, the underrepresentation of darker skin tones in medical textbooks and elsewhere, which serve as training data for AI, poses a significant challenge in ensuring diverse and inclusive educational materials.
This study aims to evaluate the extent of skin tone diversity in AI-generated medical images and to test whether the representation of skin tones can be improved by modifying AI prompts to better reflect the demographic makeup of the US population.
In total, 2 standard AI models (Dall-E [OpenAI] and Midjourney [Midjourney Inc]) each generated 100 images of people with psoriasis. In addition, a custom model was developed that incorporated a prompt injection aimed at "forcing" the AI (Dall-E 3) to reflect the skin tone distribution of the US population according to the 2012 American National Election Survey. This custom model generated another set of 100 images. The skin tones in these images were assessed by 3 researchers using the New Immigrant Survey skin tone scale, with the median value representing each image. A chi-square goodness of fit analysis compared the skin tone distributions from each set of images to that of the US population.
The standard AI models (Dalle-3 and Midjourney) demonstrated a significant difference between the expected skin tones of the US population and the observed tones in the generated images (P<.001). Both standard AI models overrepresented lighter skin. Conversely, the custom model with the modified prompt yielded a distribution of skin tones that closely matched the expected demographic representation, showing no significant difference (P=.04).
This study reveals a notable bias in AI-generated medical images, predominantly underrepresenting darker skin tones. This bias can be effectively addressed by modifying AI prompts to incorporate real-life demographic distributions. The findings emphasize the need for conscious efforts in AI development to ensure diverse and representative outputs, particularly in educational and medical contexts. Users of generative AI tools should be aware that these biases exist, and that similar tendencies may also exist in other types of generative AI (eg, large language models) and in other characteristics (eg, sex, gender, culture, and ethnicity). Injecting demographic data into AI prompts may effectively counteract these biases, ensuring a more accurate representation of the general population.
O'Malley A
，Veenhuizen M
，Ahmed A
《-》
Theory of trust and acceptance of artificial intelligence technology (TrAAIT): An instrument to assess clinician trust and acceptance of artificial intelligence.
Artificial intelligence and machine learning (AI/ML) technologies like generative and ambient AI solutions are proliferating in real-world healthcare settings. Clinician trust affects adoption and impact of these systems. Organizations need a validated method to assess factors underlying trust and acceptance of AI for clinical workflows in order to improve adoption and the impact of AI.
Our study set out to develop and assess a novel clinician-centered model to measure and explain trust and adoption of AI technology. We hypothesized that clinicians' system-specific Trust in AI is the primary predictor of both Acceptance (i.e., willingness to adopt), and post-adoption Trusting Stance (i.e., general stance towards any AI system). We validated the new model at an urban comprehensive cancer center. We produced an easily implemented survey tool for measuring clinician trust and adoption of AI.
This survey-based, cross-sectional, psychometric study included a model development phase and validation phase. Measurement was done with five-point ascending unidirectional Likert scales. The development sample included N = 93 clinicians (physicians, advanced practice providers, nurses) that used an AI-based communication application. The validation sample included N = 73 clinicians that used a commercially available AI-powered speech-to-text application for note-writing in an electronic health record (EHR). Analytical procedures included exploratory factor analysis (EFA), confirmatory factor analysis (CFA), and partial least squares structural equation modeling (PLS-SEM). The Johnson-Neyman (JN) methodology was used to determine moderator effects.
In the fully moderated causal model, clinician trust explained a large amount of variance in their acceptance of a specific AI application (56%) and their post-adoption general trusting stance towards AI in general (36%). Moderators included organizational assurances, length of time using the application, and clinician age. The final validated instrument has 20 items and takes 5 min to complete on average.
We found that clinician acceptance of AI is determined by their degree of trust formed via information credibility, perceived application value, and reliability. The novel model, TrAAIT, explains factors underlying AI trustworthiness and acceptance for clinicians. With its easy-to-use instrument and Summative Score Dashboard, TrAAIT can help organizations implementing AI to identify and intercept barriers to clinician adoption in real-world settings.
Stevens AF
，Stetson P
《-》
ResearchGate
(全网免费下载)
钛学术
(全网免费下载)
