Preliminary Technical Validation of LittleBeats™: A Multimodal Sensing Platform to Capture Cardiac Physiology, Motion, and Vocalizations

Abstract

Across five studies, we present the preliminary technical validation of an infant-wearable platform, LittleBeats™, that integrates electrocardiogram (ECG), inertial measurement unit (IMU), and audio sensors. Each sensor modality is validated against data from gold-standard equipment using established algorithms and laboratory tasks. Interbeat interval (IBI) data obtained from the LittleBeats™ ECG sensor indicate acceptable mean absolute percent error rates for both adults (Study 1, N = 16) and infants (Study 2, N = 5) across low- and high-challenge sessions and expected patterns of change in respiratory sinus arrythmia (RSA). For automated activity recognition (upright vs. walk vs. glide vs. squat) using accelerometer data from the LittleBeats™ IMU (Study 3, N = 12 adults), performance was good to excellent, with smartphone (industry standard) data outperforming LittleBeats™ by less than 4 percentage points. Speech emotion recognition (Study 4, N = 8 adults) applied to LittleBeats™ versus smartphone audio data indicated a comparable performance, with no significant difference in error rates. On an automatic speech recognition task (Study 5, N = 12 adults), the best performing algorithm yielded relatively low word error rates, although LittleBeats™ (4.16%) versus smartphone (2.73%) error rates were somewhat higher. Together, these validation studies indicate that LittleBeats™ sensors yield a data quality that is largely comparable to those obtained from gold-standard devices and established protocols used in prior research.

Evaluating Users’ Experiences of a Child Multimodal Wearable Device: Mixed Methods Approach

“…Furthermore, LittleBeats' multiple modalities provide especially rich data to assess an array of constructs central to child development researchers and clinicians, including parent-child vocal turn-taking, regulation of stress, sleep-wake cycles, physical activity, and developmental disorders. At the same time, although we have validated LittleBeats sensors and machine learning algorithms to accurately capture some of these key constructs [17][18][19]43], the degree to which LittleBeats and similar child wearables deliver benefits (eg, high ecological validity and low burden) will largely depend on acceptance by the end users (eg, parents and children), making user experience studies critical to this research space. In short, if the technology is not acceptable to the end user, it is less likely to be adopted and used as intended.…”

“…To minimize the risk of data being intercepted during transfer (ie, uploading data via wireless or Bluetooth networks), data were stored directly on a microSD card on the physical device, and files were configured in such a way that only study personnel could access the data in a human-readable format (eg, wav files for audio) using a data processing pipeline developed specifically for LittleBeats. Because LittleBeats is not a commercial device, simple modifications can be made to the device firmware (eg, "turning off" ≥1 of the sensors) to suit research goals (refer to the study by Islam et al [19] for details about technical specifications).…”

“…In prior papers, we reported on machine learning algorithms used to detect and classify child and parent vocalizations using audio data from the LittleBeats device [17] and child sleep states using all 3 sensor modalities [18]. We also conducted technical validation studies to assess the signal quality of each sensor modality in relation to established laboratory protocols and gold-standard equipment [19]. Complementing these prior reports, we focus here on the critical issue of "user experience" among families and their children aged 1 month to 9.5 years.…”