Human-in-the-loop (HITL) AI may enable an ideal symbiosis of human experts and AI models, harnessing the advantages of both while at the same time overcoming their respective limitations. The purpose of this study was to investigate a novel collective intelligence technology designed to amplify the diagnostic accuracy of networked human groups by forming real-time systems modeled on biological swarms. Using small groups of radiologists, the swarm-based technology was applied to the diagnosis of pneumonia on chest radiographs and compared against human experts alone, as well as two state-of-the-art deep learning AI models. Our work demonstrates that both the swarm-based technology and deep-learning technology achieved superior diagnostic accuracy than the human experts alone. Our work further demonstrates that when used in combination, the swarm-based technology and deep-learning technology outperformed either method alone. The superior diagnostic accuracy of the combined HITL AI solution compared to radiologists and AI alone has broad implications for the surging clinical AI deployment and implementation strategies in future practice.

B ecause environmental and natural resource stakeholders hold valuable knowledge about social-ecological dynamics, researchers often attempt to incorporate local knowledge into environmental models and resource management (Gray et al. 2018). This knowledge is considered valuable because many resource users routinely engage with the natural environment through activities like fishing or hunting (Arlinghaus and Krause 2013). In addition, natural resource users may share information about environmental, policy, or social changes across their social networks (Barnes et al. 2016), allowing them to accumulate and refine knowledge and observations across years and locations (eg anglers moving among lakes; Carruthers et al. 2018). Such local ecological knowledge is useful and can be integrated with expert scientific knowledge to provide a more comprehensive understanding of pressing environmental issues (Gray et al. 2012). At the same time, recent technological developments have vastly improved the way this information is not only collected from stakeholders but also provided through new, cyberenabled forms of participatory environmental assessments (Voinov et al. 2018). For example, the emergence of citizen science (Cooper et al. 2007; Bonney et al. 2009; Shirk et al. 2012) and the internet has changed the nature of data-sourcing by extending the geographic reach of ecological surveys considerably, and in some cases has shifted monitoring away from the consultation of a small minority of scientific experts to now include larger scale collaborations involving dozens to thousands of local stakeholders and scientists (Shirk et al. 2012). More recently, traditional citizen science has been expanded to include open online collaborative communities, leading to scientific discoveries that blend the strengths of humans and machines (Mugar et al. 2014; Trouille et al. 2019). By integrating both local and expert knowledge, citizen science and other crowdsourcing approaches can reduce the uncer

No abstract

No abstract

Every year millions of people fill out brackets, trying to accurately predict the outcome of the NCAA Men’s Basketball March Madness tournament. This study examines how collective swarm intelligence might impact these choices in small groups. Rather than working by themselves, groups of people came together to combine their knowledge and opinions and pick brackets collectively. It is generally agreed that collective intelligence is effective in decision-making. However, how and why collective intelligence augments performance has not been totally agreed upon, and the theoretical explanations have been elusive. This study examines groups that are either highly dedicated or high in expertise to see whether they perform differently based on these dimensions.