Designing Videogames to Crowdsource Accelerometer Data Annotation for Activity Recognition Research

"Crying in the Wilderness"-The Use of Web-Based Support in Telomere Biology Disorders: Thematic Analysis

BACKGROUND

Web-based information and social support are commonly used in rare disease communities where geographic dispersion and limited provider expertise complicate in-person support. We examined web-based resource use among caregivers of individuals with telomere biology disorders (TBDs), which are rare genetic conditions with long diagnostic odysseys and uncertain prognoses including multiorgan system cancer risk.

OBJECTIVE

This study explored internet-based information-seeking and social support practices and perspectives of patients with TBDs and their caregivers.

METHODS

Our qualitative descriptive study used semistructured interviews of patients with TBDs and caregivers. Data were transcribed verbatim and thematically analyzed by an interdisciplinary team.

RESULTS

A total of 32 adults completed interviews. Participant ages ranged from 27 to 74 years. The majority (n=28, 88%) were female, occupied multiple TBD roles (eg, patient and parent), and had undergone genetic testing. Most engaged in web-based information-seeking (n=29, 91%) and TBD-specific social media (n=26, 81%). Participants found web-based resources useful for information-seeking but reported privacy concerns and frustration with forming supportive relationships. Most participants described ambivalence toward web-based resource use, citing tensions between hunger for information versus distrust, empowerment versus overwhelm, disclosure versus privacy, and accessibility versus connection. Fluctuations in web-based support use arose from perceived harms, information saturation, or decreased relevance over the course of TBD illness experience.

CONCLUSIONS

Individuals with TBDs and their caregivers reported frequent use of web-based informational and emotional support. However, ambivalence about the benefits and liabilities of web-based resources and persistent medical uncertainty may impact the adoption of and adherence to web-based support among patients with TBD and caregivers. Our findings suggest web-based psychosocial support should target long-term and multifaceted informational and emotional needs, be user-initiated, be offered alongside in-person formats, provide expert-informed information, and be attentive to personal privacy and evolving support needs of the TBD community. This study suggests web-based resources will be most effective in the TBD context when they achieve the following features: (1) offer a variety of ways to engage (eg, active and passive), (2) provide privacy protections in moderated "safe spaces" designed for personal disclosure, (3) offer separate venues for informational versus emotional support, (4) combine web-based relationship formation with opportunities for in-person gathering, (5) provide information that is reliable, easy to access, and informed by medical professionals, (6) remain mindful of user distress, and (7) are responsive to variations in levels and types of engagement. Additionally, advocacy organizations may wish to avoid traditional social media platforms when designing safe spaces for web-based emotional support, instead pivoting to internet-based tools that minimize privacy threats and limit the perpetual public availability of shared information.

Hang-Time HAR: A Benchmark Dataset for Basketball Activity Recognition Using Wrist-Worn Inertial Sensors

We present a benchmark dataset for evaluating physical human activity recognition methods from wrist-worn sensors, for the specific setting of basketball training, drills, and games. Basketball activities lend themselves well for measurement by wrist-worn inertial sensors, and systems that are able to detect such sport-relevant activities could be used in applications of game analysis, guided training, and personal physical activity tracking. The dataset was recorded from two teams in separate countries (USA and Germany) with a total of 24 players who wore an inertial sensor on their wrist, during both a repetitive basketball training session and a game. Particular features of this dataset include an inherent variance through cultural differences in game rules and styles as the data was recorded in two countries, as well as different sport skill levels since the participants were heterogeneous in terms of prior basketball experience. We illustrate the dataset's features in several time-series analyses and report on a baseline classification performance study with two state-of-the-art deep learning architectures.

Blazing fire or breezy wind? A story-driven playful experience for annotating dance movement

The annotation of animated motion-captured segments is a challenging, interdisciplinary task, especially when it comes to characterizing movement qualitatively. The lack of intuitive, easy-to-learn-and-use frameworks is considered to be one of the biggest challenges in this process; another is the lack of approaches able to motivate a wide audience of users, from the broader public to dance experts, researchers and performers, to contribute with annotations. In this paper we present Motion Hollow, a story-driven playful experience that uses metaphors based on Laban Movement Analysis, an established framework for movement analysis and annotation, to familiarize novice users with the process of qualitative characterization of dance moves. This work proposes a first step into introducing movement annotation to non-expert users, and as such, its main goal is to explore the implications and potential of such an approach. The evaluation of the experience confirms its potential to transform the annotation of dance movement segments into an engaging and enjoyable experience as well as to foster a deeper understanding of movement annotation both as a concept and process.

Signaligner Pro: A Tool to Explore and Annotate Multi-day Raw Accelerometer Data

Human activity recognition using wearable accelerometers can enable in-situ detection of physical activities to support novel human-computer interfaces. Many of the machine-learning-based activity recognition algorithms require multi-person, multi-day, carefully annotated training data with precisely marked start and end times of the activities of interest. To date, there is a dearth of usable tools that enable researchers to conveniently visualize and annotate multiple days of high-sampling-rate raw accelerometer data. Thus, we developed Signaligner Pro, an interactive tool to enable researchers to conveniently explore and annotate multi-day high-sampling rate raw accelerometer data. The tool visualizes high-sampling-rate raw data and time-stamped annotations generated by existing activity recognition algorithms and human annotators; the annotations can then be directly modified by the researchers to create their own, improved, annotated datasets. In this paper, we describe the tool's features and implementation that facilitate convenient exploration and annotation of multi-day data and demonstrate its use in generating activity annotations.