Collegiate student-athletes train and compete in a dense data ecology where information about their bodies and performances circulates among coaches, staff, and fans. To understand how student-athletes themselves engage with this data, we conducted interviews with 20 student-athletes, identifying four modes of engagement: 1) performance-directive, executing training and targeting improvement; 2) reflective-monitoring, assessing the body’s reaction to training and daily load; 3) coach-mediated, receiving insights through staff expertise; and 4) selective-disengagement, intentionally stepping back to protect confidence or avoid overload. These findings fill a gap left open by three related areas of research: SportsHCI, collegiate athletics, and personal data engagement. Each mode entails reasons, practices, and trade-offs. Student-athletes draw on different combinations of these modes as they respond to training demands, coaching oversight, and their own well-being. Our findings highlight how an evolving data ecology creates opportunities and pressures, requiring student-athletes to balance performance with protecting their state of mind.

Temporal information extraction is the task of identifying temporal entities in a text and relating them to each other. In medicine, electronic health records (EHRs) contain text that documents the sequence of events during an encounter with a patient, and sometimes the events prior to the encounter (e.g., psychosocial environment and history). Temporality is especially important for the specialty of psychiatry. In this work, we describe the updates to the guidelines that allowed us to create a corpus of temporally-annotated psychiatric discharge summaries and progress notes in English. These updated guidelines were used to create a corpus of over 18,000 events, 2,200 time expressions, and 13,000 temporal relations. Temporal information extraction performance with a baseline system trained on non-psychiatric data obtains an F1 score of 0.152 on relation extraction, indicating the importance of this new dataset for making progress on temporal information extraction in the psychiatric domain.

Recent progress in large language models (LLMs) has enabled the automated processing of lengthy documents even without supervised training on a task-specific dataset.Yet, their zero-shot performance in complex tasks as opposed to straightforward information extraction tasks remains suboptimal.One feasible approach for tasks with lengthy, complex input is to first summarize the document and then apply supervised fine-tuning to the summary.However, the summarization process inevitably results in some loss of information.In this study we present a method for processing the summaries of long documents aimed to capture different important aspects of the original document.We hypothesize that LLM summaries generated with different aspect-oriented prompts contain different information signals, and we propose methods to measure these differences.We introduce approaches to effectively integrate signals from these different summaries for supervised training of transformer models.We validate our hypotheses on a high-impact task -30-day readmission prediction from a psychiatric discharge -using real-world data from four hospitals, and show that our proposed method increases the prediction performance for the complex task of predicting patient outcome.

Mammographic screening is an effective method for detecting breast cancer, facilitating early diagnosis. However, the current need to manually inspect images places a heavy burden on healthcare systems, spurring a desire for automated diagnostic protocols. Techniques based on deep neural networks have been shown effective in some studies, but their tendency to overfit leaves considerable risk for poor generalisation and misdiagnosis, preventing their widespread adoption in clinical settings. Data augmentation schemes based on unpaired neural style transfer models have been proposed that improve generalisability by diversifying the representations of training image features in the absence of paired training data (images of the same tissue in either image style). But these models are similarly prone to various pathologies, and evaluating their performance is challenging without ground truths/large datasets (as is often the case in medical imaging). Here, we consider two frameworks/architectures: a GAN-based cycleGAN, and the more recently developed diffusion-based SynDiff. We evaluate their performance when trained on image patches parsed from three open access mammography datasets and one non-medical image dataset to demonstrate its generalisability. We consider the use of uncertainty quantification to assess model trustworthiness, and propose a scheme to evaluate calibration quality in unpaired training scenarios. This ultimately helps facilitate the trustworthy use of image-to-image translation models in domains where ground truths are not typically available.

We describe the design and implementation of a scientific database for the map assembly tasks performed by the geneticists at the University of Michigan Human Genome Center. Our system manages complex genomic data and supports the automation of the associated map assembly tasks. For the former, we present a genomic object model that integrates both experimental and derived data. For the latter, we describe operators to automate some of the analysis steps. To develop a framework for implementing our rule-based approach to physical mapping, we have designed and implemented an active object-oriented database (OODB) system, called Crystal, on GemStone. Crystal seamlessly integrates inference capabilities with complex object modeling and other typical database capabilities as required for physical mapping. We also discuss the implementation of a physical map assembly tool on top of Crystal. In conclusion, we provide a walk-through example that demonstrates how our approach can be used to effectively support physical contig assembly.>

Article An information model for genome map representation and assembly Share on Authors: A. J. Lee Department of EECS, University of Michigan, Ann Arbor, Ann Arbor, MI Department of EECS, University of Michigan, Ann Arbor, Ann Arbor, MIView Profile , E. A. Rendensteiner Department of EECS, University of Michigan, Ann Arbor, Ann Arbor, MI Department of EECS, University of Michigan, Ann Arbor, Ann Arbor, MIView Profile , S. Thomas Information Technology and Networking, University of Michigan Medical Center, Human Genome Center, 2570C MSRB II Information Technology and Networking, University of Michigan Medical Center, Human Genome Center, 2570C MSRB IIView Profile , S. Lafortune Department of EECS, University of Michigan, Ann Arbor, Ann Arbor, MI Department of EECS, University of Michigan, Ann Arbor, Ann Arbor, MIView Profile Authors Info & Claims CIKM '93: Proceedings of the second international conference on Information and knowledge managementDecember 1993 Pages 75–84https://doi.org/10.1145/170088.170107Published:01 December 1993 5citation1,221DownloadsMetricsTotal Citations5Total Downloads1,221Last 12 Months0Last 6 weeks0 Get Citation AlertsNew Citation Alert added!This alert has been successfully added and will be sent to:You will be notified whenever a record that you have chosen has been cited.To manage your alert preferences, click on the button below.Manage my AlertsNew Citation Alert!Please log in to your account Save to BinderSave to BinderCreate a New BinderNameCancelCreateExport CitationPublisher SiteGet Access