Although rapid advances in machine learning have made it increasingly applicable to expert decision-making, the delivery of accurate algorithmic predictions alone is insufficient for effective human-AI collaboration. In this work, we investigate the key types of information medical experts desire when they are first introduced to a diagnostic AI assistant. In a qualitative lab study, we interviewed 21 pathologists before, during, and after being presented deep neural network (DNN) predictions for prostate cancer diagnosis, to learn the types of information that they desired about the AI assistant. Our findings reveal that, far beyond understanding the local, case-specific reasoning behind any model decision, clinicians desired upfront information about basic, global properties of the model, such as its known strengths and limitations, its subjective point-of-view, and its overall design objective-what it's designed to be optimized for. Participants compared these information needs to the collaborative mental models they develop of their medical colleagues when seeking a second opinion: the medical perspectives and standards that those colleagues embody, and the compatibility of those perspectives with their own diagnostic patterns. These findings broaden and enrich discussions surrounding AI transparency for collaborative decision-making, providing a richer understanding of what experts find important in their introduction to AI assistants before integrating them into routine practice. CCS Concepts: • Human-centered computing → Human computer interaction (HCI).

IMPORTANCE Expert-level artificial intelligence (AI) algorithms for prostate biopsy grading have recently been developed. However, the potential impact of integrating such algorithms into pathologist workflows remains largely unexplored. OBJECTIVETo evaluate an expert-level AI-based assistive tool when used by pathologists for the grading of prostate biopsies. DESIGN, SETTING, AND PARTICIPANTSThis diagnostic study used a fully crossed multiple-reader, multiple-case design to evaluate an AI-based assistive tool for prostate biopsy grading. Retrospective grading of prostate core needle biopsies from 2 independent medical laboratories in the US was performed between October 2019 and January 2020. A total of 20 general pathologists reviewed 240 prostate core needle biopsies from 240 patients. Each pathologist was randomized to 1 of 2 study cohorts. The 2 cohorts reviewed every case in the opposite modality (with AI assistance vs without AI assistance) to each other, with the modality switching after every 10 cases. After a minimum 4-week washout period for each batch, the pathologists reviewed the cases for a second time using the opposite modality. The pathologist-provided grade group for each biopsy was compared with the majority opinion of urologic pathology subspecialists. EXPOSURE An AI-based assistive tool for Gleason grading of prostate biopsies. MAIN OUTCOMES AND MEASURES Agreement between pathologists and subspecialists with andwithout the use of an AI-based assistive tool for the grading of all prostate biopsies and Gleason grade group 1 biopsies. RESULTSBiopsies from 240 patients (median age, 67 years; range, 39-91 years) with a median prostate-specific antigen level of 6.5 ng/mL (range, 0.6-97.0 ng/mL) were included in the analyses.Artificial intelligence-assisted review by pathologists was associated with a 5.6% increase (95% CI, 3.2%-7.9%; P < .001) in agreement with subspecialists (from 69.7% for unassisted reviews to 75.3% for assisted reviews) across all biopsies and a 6.2% increase (95% CI, 2.7%-9.8%; P = .001) in agreement with subspecialists (from 72.3% for unassisted reviews to 78.5% for assisted reviews) for grade group 1 biopsies. A secondary analysis indicated that AI assistance was also associated with improvements in tumor detection, mean review time, mean self-reported confidence, and interpathologist agreement. CONCLUSIONS AND RELEVANCEIn this study, the use of an AI-based assistive tool for the review of prostate biopsies was associated with improvements in the quality, efficiency, and consistency of cancer detection and grading.

Elevated weight variability in young women may signal the degradation of body weight regulatory systems. In an obesogenic environment this may eventuate in accelerated weight gain, particularly in those with a genetic susceptibility toward overweight. Future research is needed to evaluate the reliability of weight variability as a predictor of future weight gain and the sources of its predictive effect. The trial on which this study is based is registered at clinicaltrials.gov as NCT00456131.

Background: Both an elevated brain-reward-region response to palatable food and elevated weight variability have been shown to predict future weight gain. Objective: We examined whether the brain-reward response to food is related to future weight variability. Design: A total of 162 healthy-weight adolescents, who were aged 14-18 y at baseline, were enrolled in the study and were assessed annually over a 3-y follow-up period with 127 participants completing the final 3-y follow-up assessment. With the use of functional magnetic resonance imaging, we tested whether the neural responses to a cue that signaled an impending milkshake receipt and the receipt of the milkshake predicted weight variability over the follow-up period. Weight variability was modeled with a root mean squared error method to reflect fluctuations in weight independent of the net weight change. Results: Elevated activation in the medial prefrontal cortex and supplementary motor area, cingulate gyrus, cuneus and occipital gyrus, and insula in response to milkshake receipt predicted greater weight variability. Greater activation in the precuneus and middle temporal gyrus predicted lower weight variability. Conclusions: From our study data, we suggest that the elevated activation of reward and emotional-regulation brain regions (medial prefrontal cortex, cingulate cortex, and insula) and lower activation in self-reference regions (precuneus) in response to milkshake receipt predict weight variability over 3 y of follow-up. The fact that the reward response in the current study emerged in response to high-calorie palatable food receipt suggests that weight variability may be a measure of propensity periods of a positive energy balance and should be examined in addition to measures of the net weight change. With our collective results, we suggest that weight variability and its brain correlates should be added to other variables that are predictive of weight gain to inform the design of obesity-preventive programs in adolescents. This trial was registered at clinicaltrials.gov as NCT01807572.Am J Clin Nutr 2017;105:781-9.

Recurrent binge eating, or overeating accompanied by a sense of loss of control, is a major public health concern. Identifying similarities and differences among individuals with binge eating and those with other psychiatric symptoms and characterizing the deficits that uniquely predispose individuals to eating problems are essential to improving treatment. Research suggests that altered reward and control-related processes may contribute to dysregulated eating and other impulsive behaviors in binge-eating populations, but the best methods for reliably assessing the contributions of these processes to binge eating are unclear. In this review, we summarize standard neurocognitive and neuroimaging tasks that assess reward and control-related processes, describe adaptations of these tasks used to study eating and food-specific responsivity and deficits, and consider the advantages and limitations of these tasks. Future studies integrating both general and food-specific tasks with neuroimaging will improve understanding of the neurocognitive processes and neural circuits that contribute to binge eating and could inform novel interventions that more directly target or prevent this transdiagnostic behavior.