Bayesian modeling of human-AI complementarity

Artificial intelligence-enhanced patient evaluation: bridging art and science

The advent of digital health and artificial intelligence (AI) has promised to revolutionize clinical care, but real-world patient evaluation has yet to witness transformative changes. As history taking and physical examination continue to rely on long-established practices, a growing pipeline of AI-enhanced digital tools may soon augment the traditional clinical encounter into a data-driven process. This article presents an evidence-backed vision of how promising AI applications may enhance traditional practices, streamlining tedious tasks while elevating diverse data sources, including AI-enabled stethoscopes, cameras, and wearable sensors, to platforms for personalized medicine and efficient care delivery. Through the lens of traditional patient evaluation, we illustrate how digital technologies may soon be interwoven into routine clinical workflows, introducing a novel paradigm of longitudinal monitoring. Finally, we provide a skeptic's view on the practical, ethical, and regulatory challenges that limit the uptake of such technologies.

Three Challenges for AI-Assisted Decision-Making

Artificial intelligence (AI) has the potential to improve human decision-making by providing decision recommendations and problem-relevant information to assist human decision-makers. However, the full realization of the potential of human-AI collaboration continues to face several challenges. First, the conditions that support complementarity (i.e., situations in which the performance of a human with AI assistance exceeds the performance of an unassisted human or the AI in isolation) must be understood. This task requires humans to be able to recognize situations in which the AI should be leveraged and to develop new AI systems that can learn to complement the human decision-maker. Second, human mental models of the AI, which contain both expectations of the AI and reliance strategies, must be accurately assessed. Third, the effects of different design choices for human-AI interaction must be understood, including both the timing of AI assistance and the amount of model information that should be presented to the human decision-maker to avoid cognitive overload and ineffective reliance strategies. In response to each of these three challenges, we present an interdisciplinary perspective based on recent empirical and theoretical findings and discuss new research directions.

Equation for Attractiveness: Integrating Multidimensional Factors Through Computational Neuroaesthetics

BACKGROUND

Understanding the multifaceted nature of attractiveness (A), which encompasses physical beauty (PB), genuineness (GEN), self-confidence (SC), and prior experience (RE), is crucial for various domains, including psychology and clinical aesthetics. Previous studies have often isolated specific elements, failing to capture their intricate interplay. This study aims to develop a comprehensive equation for attractiveness using computational neuroaesthetics.

METHOD

The study began with a pilot study involving 250 participants (50 experts and 200 laypersons) who prerated 500 facial images on a Likert scale for traits such as physical beauty, genuineness, self-confidence, and perceived prior experience. Following the pilot, the main study recruited 11,780 participants through diverse media channels to rate a new set of 1,000 facial images. Advanced computational techniques, including multiple linear regression and Bayesian hierarchical modelling, were employed to analyse the data and formulate an attractiveness equation.

RESULTS

The analysis identified genuineness as the most significant factor, followed by physical beauty, self-confidence, and prior experience. The proposed equation for attractiveness, refined through Bayesian modelling, is:A = β 0 + ( β 1 · PB + β 2 · GEN + β 3 · SC + β 4 · PE ) + ϵ A = 1.82 + ( 0.34 · PB + 0.44 · GEN + 0.26 · SC + 0.16 · PE ) + ϵ (β0 is the intercept; β1, β2, β3, β4 are the coefficients for each factor; and ϵ is the error term) CONCLUSION: The findings underscore the paramount importance of psychological traits in attractiveness assessments, suggesting a shift from purely physical enhancements to holistic interventions in clinical settings. This model provides a robust framework for understanding attractiveness and has potential applications in psychology, marketing, and AI.

LEVEL OF EVIDENCE IV

This journal requires that authors assign a level of evidence to each article. For a full description of these Evidence-Based Medicine ratings, please refer to the Table of Contents or the online Instructions to Authors www.springer.com/00266 .

Integrating Psychological Insights into Aesthetic Medicine: A Cross-Generational Analysis of Patient Archetypes (IMPACT Study)

BACKGROUND

Aesthetic medicine has evolved significantly, accommodating diverse demographics and motivations influenced by societal shifts and technological advancements. The IMPACT (integrative multigenerational psychological analysis for cosmetic treatment) study refines patient archetypes, integrating psychological theories to tailor treatments, especially for younger demographics and LGBTQIA + communities.

METHODS

This cross-sectional study utilized a comprehensive, validated survey with a Cronbach's alpha of 0.89 and a Content Validity Index (CVI) of 0.92, distributed across a globally diverse, generationally stratified sample. Techniques like regression analysis, ANOVA, Bayesian modelling, and factor analysis were employed to analyse the data, focusing on developing nuanced patient archetypes.

RESULTS

Among 5645 participants, 5340 complete responses highlighted significant generational differences in aesthetic preferences. Millennials showed a strong preference for non-invasive procedures (β = 0.65, p < 0.001). ANOVA results confirmed significant variances across generations [F (3, 5118) = 157.6, p < 0.001], with post-hoc analyses delineating specific inter-group differences. Bayesian modelling provided insights into the probability of non-invasive preferences among younger cohorts at over 92% certainty. Factor analysis revealed key dimensions such as 'Generational Influence' and 'Technological Adoption,' which helped in defining archetypes including Dynamic Self-Identity, Digital Native, Stability Seeker, Classic Conservatism, and Holistic Health, collectively explaining up to 78% of the variance in responses.

CONCLUSION

The IMPACT study underscores the influence of generational identity and digital exposure on aesthetic preferences, advocating for personalized, archetype-based treatment approaches. This aligns with enhancing patient satisfaction and treatment outcomes, promoting an adaptive aesthetic medicine practice that meets the evolving needs of modern patients.

LEVEL OF EVIDENCE IV

This journal requires that authors assign a level of evidence to each article. For a full description of these Evidence-Based Medicine ratings, please refer to the Table of Contents or the online Instructions to Authors www.springer.com/00266 .

Perceptions and detection of AI use in manuscript preparation for academic journals

The rapid advances in Generative AI tools have produced both excitement and worry about how AI will impact academic writing. However, little is known about what norms are emerging around AI use in manuscript preparation or how these norms might be enforced. We address both gaps in the literature by conducting a survey of 271 academics about whether it is necessary to report ChatGPT use in manuscript preparation and by running GPT-modified abstracts from 2,716 published papers through a leading AI detection software to see if these detectors can detect different AI uses in manuscript preparation. We find that most academics do not think that using ChatGPT to fix grammar needs to be reported, but detection software did not always draw this distinction, as abstracts for which GPT was used to fix grammar were often flagged as having a high chance of being written by AI. We also find disagreements among academics on whether more substantial use of ChatGPT to rewrite text needs to be reported, and these differences were related to perceptions of ethics, academic role, and English language background. Finally, we found little difference in their perceptions about reporting ChatGPT and research assistant help, but significant differences in reporting perceptions between these sources of assistance and paid proofreading and other AI assistant tools (Grammarly and Word). Our results suggest that there might be challenges in getting authors to report AI use in manuscript preparation because (i) there is not uniform agreement about what uses of AI should be reported and (ii) journals might have trouble enforcing nuanced reporting requirements using AI detection tools.

"Filler-Associated Acute Stroke Syndrome": Classification, Predictive Modelling of Hyaluronidase Efficacy, and Updated Case Review on Neurological and Visual Complications

INTRODUCTION

The rising use of soft tissue fillers for aesthetic procedures has seen an increase in complications, including vascular occlusions and neurological symptoms that resemble stroke. This study synthesizes information on central nervous system (CNS) complications post-filler injections and evaluates the effectiveness of hyaluronidase (HYAL) treatment.

METHODS

A thorough search of multiple databases, including PubMed, EMBASE, Scopus, Web of Science, Google Scholar, and Cochrane, focused on publications from January 2014 to January 2024. Criteria for inclusion covered reviews and case reports that documented CNS complications related to soft tissue fillers. Advanced statistical and computational techniques, including logistic regression, machine learning, and Bayesian analysis, were utilized to dissect the factors influencing therapeutic outcomes.

RESULTS

The analysis integrated findings from 20 reviews and systematic analyses, with 379 cases reported since 2018. Hyaluronic acid (HA) was the most commonly used filler, particularly in nasal region injections. The average age of patients was 38, with a notable increase in case reports in 2020. Initial presentation data revealed that 60.9% of patients experienced no light perception, while ptosis and ophthalmoplegia were present in 54.3 and 42.7% of cases, respectively. The statistical and machine learning analyses did not establish a significant linkage between the HYAL dosage and patient recovery; however, the injection site emerged as a critical determinant.

CONCLUSION

The study concludes that HYAL treatment, while vital for managing complications, varies in effectiveness based on the injection site and the timing of administration. The non-Newtonian characteristics of HA fillers may also affect the incidence of complications. The findings advocate for tailored treatment strategies incorporating individual patient variables, emphasizing prompt and precise intervention to mitigate the adverse effects of soft tissue fillers.

LEVEL OF EVIDENCE III

This journal requires that authors assign a level of evidence to each article. For a full description of these Evidence-Based Medicine ratings, please refer to the Table of Contents or the online Instructions to Authors www.springer.com/00266 .

Boosting wisdom of the crowd for medical image annotation using training performance and task features

A crucial bottleneck in medical artificial intelligence (AI) is high-quality labeled medical datasets. In this paper, we test a large variety of wisdom of the crowd algorithms to label medical images that were initially classified by individuals recruited through an app-based platform. Individuals classified skin lesions from the International Skin Lesion Challenge 2018 into 7 different categories. There was a large dispersion in the geographical location, experience, training, and performance of the recruited individuals. We tested several wisdom of the crowd algorithms of varying complexity from a simple unweighted average to more complex Bayesian models that account for individual patterns of errors. Using a switchboard analysis, we observe that the best-performing algorithms rely on selecting top performers, weighting decisions by training accuracy, and take into account the task environment. These algorithms far exceed expert performance. We conclude by discussing the implications of these approaches for the development of medical AI.

AI and machine learning in medical imaging: key points from development to translation

Innovation in medical imaging artificial intelligence (AI)/machine learning (ML) demands extensive data collection, algorithmic advancements, and rigorous performance assessments encompassing aspects such as generalizability, uncertainty, bias, fairness, trustworthiness, and interpretability. Achieving widespread integration of AI/ML algorithms into diverse clinical tasks will demand a steadfast commitment to overcoming issues in model design, development, and performance assessment. The complexities of AI/ML clinical translation present substantial challenges, requiring engagement with relevant stakeholders, assessment of cost-effectiveness for user and patient benefit, timely dissemination of information relevant to robust functioning throughout the AI/ML lifecycle, consideration of regulatory compliance, and feedback loops for real-world performance evidence. This commentary addresses several hurdles for the development and adoption of AI/ML technologies in medical imaging. Comprehensive attention to these underlying and often subtle factors is critical not only for tackling the challenges but also for exploring novel opportunities for the advancement of AI in radiology.

Machine learning augmentation reduces prediction error in collective forecasting: development and validation across prediction markets with application to COVID events

BACKGROUND

The recent COVID-19 pandemic highlighted the challenges for traditional forecasting. Prediction markets are a promising way to generate collective forecasts and could potentially be enhanced if high-quality crowdsourced inputs were identified and preferentially weighted for likely accuracy in real-time with machine learning.

METHODS

We aim to leverage human prediction markets with real-time machine weighting of likely higher accuracy trades to improve performance. The crowd sourced Almanis prediction market longitudinal platform (n = 1822) and Next Generation Social Science (NGS2) platform (n = 103) were utilised.

FINDINGS

A 43-feature model predicted accurate forecasters, those with top quintile relative Brier accuracy, with subsequent replication in two out-of-sample datasets (pboth <1 × 10-9). Trades graded by this model as having higher accuracy scores than others produced a greater AUC temporal gain in the overall market after vs before trade. Accuracy score-weighted forecasts had higher accuracy than market forecasts alone, particularly when the two systems disagreed by 5% or more for binary event prediction: the hybrid system demonstrating substantial % AUC gains of 13.2%, p = 1.35 × 10-14 and 13.8%, p = 0.003 in two out-of-sample datasets. When discordant, the hybrid model was correct for COVID-19 event occurrence 72.7% of the time vs 27.3% for market models, p = 0.007. This net classification benefit was replicated in the separate Almanis B dataset, p = 2.4 × 10-7.

INTERPRETATION

Real-time machine classification followed by weighting human trades according to likely accuracy improves collective forecasting performance. This could provide improved anticipation of and thus response to emerging risks.

FUNDING

This work was supported by an AusIndustry R and D tax incentive program from the Department of Industry, Science, Energy and Resources, Australia, to SlowVoice Pty Ltd. (IR 2101990) and Fellowship (GNT 1110200) and Investigator grant (GNT 1197234) to A-L Ponsonby by the National Health and Medical Research Council of Australia.

Experimental drugs in clinical trials for COPD: Artificial Intelligence via Machine Learning approach to predict the successful advance from early-stage development to approval

INTRODUCTION

Therapeutic advances in drug therapy of chronic obstructive pulmonary disease (COPD) really effective in suppressing the pathological processes underlying the disease deterioration are still needed. Artificial Intelligence (AI) via Machine Learning (ML) may represent an effective tool to predict clinical development of investigational agents.

AREAL COVERED

Experimental drugs in Phase I and II development for COPD from early 2014 to late 2022 were identified in the ClinicalTrials.gov database. Different ML models, trained from prior knowledge on clinical trial success, were used to predict the probability that experimental drugs will successfully advance toward approval in COPD, according to Bayesian inference as follows: ≤25% low probability, >25% and ≤ 50% moderate probability, >50% and ≤ 75% high probability, and > 75% very high probability.

EXPERT OPINION

The Artificial Neural Network and Random Forest ML models indicated that, among the current experimental drugs in clinical trials for COPD, only the bifunctional muscarinic antagonist - β₂-adrenoceptor agonists (MABA) navafenterol and batefenterol, the inhaled corticosteroid (ICS)/MABA fluticasone furoate/batefenterol, and the bifunctional phosphodiesterase (PDE) 3/4 inhibitor ensifentrine resulted to have a moderate to very high probability of being approved in the next future, however not before 2025.

Cognitive Science of Augmented Intelligence

Cognitive science has been traditionally organized around the individual as the basic unit of cognition. Despite developments in areas such as communication, human-machine interaction, group behavior, and community organization, the individual-centric approach heavily dominates both cognitive research and its application. A promising direction for cognitive science is the study of augmented intelligence, or the way social and technological systems interact with and extend individual cognition. The cognitive science of augmented intelligence holds promise in helping society tackle major real-world challenges that can only be discovered and solved by teams made of individuals and machines with complementary skills who can productively collaborate with each other.