State-of-the-art automatic sleep staging methods have already demonstrated comparable reliability and superior time efficiency to manual sleep staging. However, fully automatic black-box solutions are difficult to adapt into clinical workflow and the interaction between explainable automatic methods and the work of sleep technologists remains underexplored and inadequately conceptualized. Thus, we propose a human-in-the-loop concept for sleep analysis, presenting an automatic sleep staging model (aSAGA), that performs effectively with both clinical polysomnographic recordings and home sleep studies. To validate the model, extensive testing was conducted, employing a preclinical validation approach with three retrospective datasets; open-access, clinical, and research-driven. Furthermore, we validate the utilization of uncertainty mapping to identify ambiguous regions, conceptualized as gray areas, in automatic sleep analysis that warrants manual re-evaluation. The results demonstrate that the automatic sleep analysis achieved a comparable level of agreement with manual analysis across different sleep recording types. Moreover, validation of the gray area concept revealed its potential to enhance sleep staging accuracy and identify areas in the recordings where sleep technologists struggle to reach a consensus. In conclusion, this study introduces and validates a concept from explainable artificial intelligence into sleep medicine and provides the basis for integrating human-in-the-loop automatic sleep staging into clinical workflows, aiming to reduce black-box criticism and the burden associated with manual sleep staging.

Land-use decision-making processes have a long history of producing globally pervasive systemic equity and sustainability concerns. Quantitative, optimization-based planning approaches, e.g. Multi-Objective Land Allocation (MOLA), seemingly open the possibility to improve objectivity and transparency by explicitly evaluating planning priorities by the type, amount, and location of land uses. Here, we show that optimization-based planning approaches with generic planning criteria generate a series of unstable "flashpoints" whereby tiny changes in planning priorities produce large-scale changes in the amount of land use by type. We give quantitative arguments that the flashpoints we uncover in MOLA models are examples of a more general family of instabilities that occur whenever planning accounts for factors that coordinate use on- and between-sites, regardless of whether these planning factors are formulated explicitly or implicitly. We show that instabilities lead to regions of ambiguity in land-use type that we term "gray areas". By directly mapping gray areas between flashpoints, we show that quantitative methods retain utility by reducing combinatorially large spaces of possible land-use patterns to a small, characteristic set that can engage stakeholders to arrive at more efficient and just outcomes.

An engineer-turned-entrepreneur helping small businesses survive and thrive with AI. We've had some impeccable growth, development and innovation in artificial intelligence (AI) and machine learning (ML). The two niches of IT are being lauded as the technology that will solve the most significant problems of our planet, if not all. Although that may or may not be accurate, AI systems are becoming pretty popular and valuable in industries such as healthcare and automobiles, with systems that can diagnose diseases based on symptoms, enable self-driving cars and more. This is because they need more and better training datasets to become more accurate and precise.

Business and technology professionals have been continuing to face challenges in operationalizing ML for effective development, deployment, and governance. Many of us still view the operationalization process as more of an art than a systemic approach. Because ML initiatives are different from traditional IT product development initiatives. ML initiatives are very experimental and require skills from many more domains, for example-- statistical analysis, data analysis, platform engineering, and application development. Also, there is often a lack of process understanding, communication gap between teams involved, and development and ops teams' unwillingness to engage in each other domains for effective alignment of ML models' development and operationalization.

When the novel coronavirus startled the world earlier this year, San Francisco quickly took action by suspending large public gatherings in the city. The order meant many artists -- and art exhibitors -- had to quickly consider how they must pivot and prepare for a society that exists with the virus, one that might not allow the same artistic interactions we've come to take for granted. As such, the pandemic is accelerating technology's already rapid transformation of visual arts, from their creation process to their discovery and experience. Within visual arts, artificial intelligence (AI) has already redefined who can be an artist. In 2018, a portrait created by an AI was sold at auction for $432,000, reaching a new milestone for conceptual and generative art.

The new decade opened with some intriguing news: The journal Nature reported that artificial intelligence was better at identifying breast cancers on mammograms than radiologists. Researchers at Google Health teamed up with academic medical centers in the United States and Britain to train an AI system using tens of thousands of mammograms. To understand why, it helps to have a sense of how AI systems learn. In this case, the system was trained with images labeled as either "cancer" or "not cancer." From them, it learned to deduce features -- such as shape, density and edges -- that are associated with the cancer label.

The new decade opened with some intriguing news: The journal Nature reported that artificial intelligence was better at identifying breast cancers on mammograms than radiologists. Researchers at Google Health teamed up with academic medical centers in the United States and Britain to train an AI system using tens of thousands of mammograms. To understand why, it helps to have a sense of how AI systems learn. In this case, the system was trained with images labeled as either "cancer" or "not cancer." From them, it learned to deduce features -- such as shape, density and edges -- that are associated with the cancer label.

The new decade opened with some intriguing news: the journal Nature reported that artificial intelligence was better at identifying breast cancers on mammograms than radiologists. Researchers at Google Health teamed up with academic medical centers in the United States and Britain to train an AI system using tens of thousands of mammograms. To understand why, it helps to have a sense of how AI systems learn. In this case, the system was trained with images labeled as either "cancer" or "not cancer." From them, it learned to deduce features from the images -- such as shape, density and edges -- that are associated with the cancer label.

By the end of the 20th century, celebrity endorsements were certainly not unusual, except for the rather important detail that Astaire had been dead for 10 years at that point. Unsurprisingly then, reaction was mixed, bordering on negative. But a precedent had been set, so by the time Peter Cushing was digitally inserted into 2016's Rogue One, the resulting controversy was barely a blip. A year later, though, we learned about deepfakes. A deepfake is a video that's been created using a combination of a kind of machine learning, algorithms, and possibly magic.

The fuzzy ROC extends Receiver Operating Curve (ROC) visualization to the situation where some data points, falling in an indeterminacy region, are not classified. It addresses two challenges: definition of sensitivity and specificity bounds under indeterminacy; and visual summarization of the large number of possibilities arising from different choices of indeterminacy zones.