Deep Automation Bias: How to Tackle a Wicked Problem of AI?

A machine learning approach using migration-related cultural stress to classify depression and post-traumatic stress disorder among hurricane survivors

BACKGROUND

Data science approaches have increasingly been used in behavioral health research and may be useful for addressing social factors contributing to disparities in health status. This study evaluated the importance of cultural stress-related factors in classifying depression and post-traumatic stress disorder (PTSD) among adult survivors (N = 319) of Hurricane Maria who migrated from Puerto Rico to the United States mainland.

METHODS

We evaluated the performance of random forests (RF) and logistic regression (LR) for classifying PTSD and depression. Models included demographic, hurricane exposure, and migration-related cultural stress variables. We inspected area under the receiver operating characteristic curve (AUC), accuracy, balanced accuracy, F1 score, precision, recall, and specificity.

RESULTS

Negative context of reception and language-related stressors were moderately important for accurately classifying depression and PTSD. For classifying depression, RF showed higher accuracy, balanced accuracy, specificity, precision, and F1. For classifying PTSD, RF showed higher accuracy, specificity, precision, and F1.

LIMITATIONS

A more thorough classification model would also include biomarkers (e.g., of allostatic load), family, community, or neighborhood-level attributes. Findings may not generalize to other groups who have experienced crisis-related migration.

CONCLUSIONS

Findings underscore the importance of culturally and linguistically appropriate and trauma-informed clinical services for recent migrants. Use of assessments to identify pre-migration and post-migration stressors could inform clinical practice with migrants presenting with behavioral health-related difficulties.

Ethical, legal, and social implications (ELSI) of virtual agents and virtual reality in healthcare

Virtual agents (VAs) and immersive virtual reality (VR) applications broaden the opportunities for accessing healthcare by transposing certain processes from the analogue world into a virtual realm. While these innovations offer a number of advantages including improved access for individuals in diverse geographic locations and novel therapeutic options, their implementation raises significant ethical, social, and legal implications. Key considerations pertain to the doctor-patient relationship, privacy and data protection, justice, fairness, and equal access as well as to issues of accountability, liability, and safety. This paper conducts a comprehensive review of the existing literature to analyse the ethical, social, and legal ramifications of employing VAs and VR applications in healthcare. It examines the recommended strategies to mitigate potential adverse effects and addresses current research gaps in this domain.

Artificial intelligence suppression as a strategy to mitigate artificial intelligence automation bias

BACKGROUND

Incorporating artificial intelligence (AI) into clinics brings the risk of automation bias, which potentially misleads the clinician's decision-making. The purpose of this study was to propose a potential strategy to mitigate automation bias.

METHODS

This was a laboratory study with a randomized cross-over design. The diagnosis of anterior cruciate ligament (ACL) rupture, a common injury, on magnetic resonance imaging (MRI) was used as an example. Forty clinicians were invited to diagnose 200 ACLs with and without AI assistance. The AI's correcting and misleading (automation bias) effects on the clinicians' decision-making processes were analyzed. An ordinal logistic regression model was employed to predict the correcting and misleading probabilities of the AI. We further proposed an AI suppression strategy that retracted AI diagnoses with a higher misleading probability and provided AI diagnoses with a higher correcting probability.

RESULTS

The AI significantly increased clinicians' accuracy from 87.2%±13.1% to 96.4%±1.9% (P < .001). However, the clinicians' errors in the AI-assisted round were associated with automation bias, accounting for 45.5% of the total mistakes. The automation bias was found to affect clinicians of all levels of expertise. Using a logistic regression model, we identified an AI output zone with higher probability to generate misleading diagnoses. The proposed AI suppression strategy was estimated to decrease clinicians' automation bias by 41.7%.

CONCLUSION

Although AI improved clinicians' diagnostic performance, automation bias was a serious problem that should be addressed in clinical practice. The proposed AI suppression strategy is a practical method for decreasing automation bias.

A Review of the Clinical Applications of Artificial Intelligence in Abdominal Imaging

Artificial intelligence (AI) has been a topic of substantial interest for radiologists in recent years. Although many of the first clinical applications were in the neuro, cardiothoracic, and breast imaging subspecialties, the number of investigated and real-world applications of body imaging has been increasing, with more than 30 FDA-approved algorithms now available for applications in the abdomen and pelvis. In this manuscript, we explore some of the fundamentals of artificial intelligence and machine learning, review major functions that AI algorithms may perform, introduce current and potential future applications of AI in abdominal imaging, provide a basic understanding of the pathways by which AI algorithms can receive FDA approval, and explore some of the challenges with the implementation of AI in clinical practice.

The news framing of artificial intelligence: a critical exploration of how media discourses make sense of automation

Analysing how news media portray A.I. reveals what interpretative frameworks around the technology circulate in public discourses. This allows for critical reflections on the making of meaning in prevalent narratives about A.I. and its impact. While research on the public perception of datafication and automation is growing, only a few studies investigate news framing practices. The present study connects to this nascent research area by charting A.I. news frames in four internationally renowned media outlets: The New York Times, The Guardian, Wired, and Gizmodo. The main goals are to identify dominant emphasis frames in AI news reporting over the past decade, to explore whether certain A.I. frames are associated with specific data risks (surveillance, data bias, cyber-war/cyber-crime, and information disorder), and what journalists and experts contribute to the media discourse. An automated content analysis serves for inductive frame detection (N = 3098), identification of risk references (dictionary-based), and network analysis of news writers. The results show how A.I.’s ubiquity emerged rapidly in the mid-2010s, and that the news discourse became more critical over time. It is further argued that A.I. news reporting is an important factor in building critical data literacy among lay audiences.

The Global Cyber Security Model: Counteracting Cyber Attacks through a Resilient Partnership Arrangement

In this paper, insights are provided into how senior managers can establish a global cyber security model that raises cyber security awareness among staff in a partnership arrangement and ensures that cyber attacks are anticipated and dealt with in real time. We deployed a qualitative research strategy that involved a group interview involving cyber security and intelligence experts. The coding approach was used to identify the themes in the data and, in addition, a number of categories and subcategories were identified. The mind map approach was utilized to identify the thought processes of senior managers in relation to ensuring that the cyber security management process is effective. The global cyber security model can be used by senior managers to establish a framework for dealing with a range of cyber security attacks, as well as to upgrade the cyber security skill and knowledge base of individuals. In order for a cyber security mentality to be established, senior managers need to ensure that staff are focused on organizational vulnerability and resilience, there is an open and transparent communication process in place, and staff are committed to sharing cyber security knowledge. By placing cyber security within the context of a partnership arrangement, senior managers can adopt a collectivist approach to cyber security and benefit from the knowledge of external experts.