Background: Automated machine-learning systems are able to de-identify electronic medical records, including free-text clinical notes. Use of such systems would greatly boost the amount of data available to researchers, yet their deployment has been limited due to uncertainty about their performance when applied to new datasets. Objective: We present practical options for clinical note de-identification, assessing performance of machine learning systems ranging from off-the-shelf to fully customized. Methods: We implement a state-of-the-art machine learning de-identification system, training and testing on pairs of datasets that match the deployment scenarios. We use clinical notes from two i2b2 competition corpora, the Physionet Gold Standard corpus, and parts of the MIMIC-III dataset. Results: Fully customized systems remove 97-99% of personally identifying information. Performance of off-the-shelf systems varies by dataset, with performance mostly above 90%. Providing a small labeled dataset or large unlabeled dataset allows for fine-tuning that improves performance over off-the-shelf systems. Conclusion: Health organizations should be aware of the levels of customization available when selecting a deidentification deployment solution, in order to choose the one that best matches their resources and target performance level.

Digitally presenting physiological signals as biofeedback to users raises awareness of both body and mind. This paper describes the effectiveness of conveying a physiological signal often overlooked for communication: breathing. We present the design and development of digital breathing patterns and their evaluation along three output modalities: visual, audio, and haptic. We also present Breeze, a wearable pendant placed around the neck that measures breathing and sends biofeedback in real-time. We evaluated how the breathing patterns were interpreted in a fixed environment and gathered qualitative data on the wearable device's design. We found that participants intentionally modified their own breathing to match the biofeedback, as a technique for understanding the underlying emotion. Our results describe how the features of the breathing patterns and the feedback modalities influenced participants' perception. We include guidelines and suggested use cases, such as Breeze being used by loved ones to increase connectedness and empathy.Signals such as heart rate and electrodermal activity (EDA), which measures perspiration, can also serve as proxies for social interactions [37]. Breathing presents an advantage over such signals because it can easily be modulated. In the field of human-computer interaction (HCI), breathing and other physiological signals have been used mainly as input [35,39]. Instead, we employ these technologies to support humanhuman communication as advocated by [13].We envision a future where interfaces can "connect" people at a distance, enforcing a bond between close ones through shared wearable biofeedback. While trying to develop such an arXiv:1802.04995v1 [cs.HC]

The broad goal of physiological genomics research is to link genes to their functions using appropriate experimental and computational techniques. Modern genomics experiments enable the generation of vast quantities of data, and interpretation of this data requires the integration of information derived from many diverse sources. Computational biology and bioinformatics offer the ability to manage and channel this information torrent. The Rat Genome Database (RGD; http://rgd.mcw.edu) has developed computational tools and strategies specifically supporting the goal of linking genes to their functional roles in rat and, using comparative genomics, to human and mouse. We present an overview of the database with a focus on these unique computational tools and describe strategies for the use of these resources in the area of physiological genomics.

No abstract

We present an approach to designing input devices that focuses on the structure of materials. We explore and visualize how a material reacts under manipulation, and harness the material's properties to design new movement sensors. Two benefits spring out of this approach. One, simpler sensing emerges from making use of existing structure in the material. Two, by working with the natural structure of the material, we create input devices with readily recognizable affordances. We present six projects using this approach. We use the natural structure (coordination) of the human body to enable a mapping from five clothing-mounted accelerometers to high-quality motion capture data, creating a low-cost performance animation system. We design silicone input devices with embedded texture allowing single-camera tracking. We study squishable, conformable materials such as foam and silicone, and create a vocabulary of unit structures (shaped cuts in the material) for harnessing patterns of compression/tension to capture particular manipulations. We use this vocabulary to build soft sensing skeletons for stuffed animals, making foam cores with e-textile versions of our unit structures. We also use this vocabulary to design a tongue input device for a collaboration with Disney Imagineering. Finally, we rethink this vocabulary and apply it to capturing, using air pressure sensors, manipulations of hollow 3D-printed rubber shapes, and 3D-print several interactive robots incorporating the new vocabulary.