Skip to main content
2024 BMES Annual Meeting Main banner
Icon Legend
This session is not in your schedule.
This session is in your schedule. Click again to remove it.

High School Student with a Traditionally Underrepresented Background

HS Poster

Poster 128 - Power Optimization of an Inertial Mass Unit Sensor for Monitoring Peristalsis, Heart Rate, and Respiratory Rate

Saturday, October 26, 2024
10:00 AM - 12:00 PM EST
Location: Exhibit Hall E, F & G

    Presenting Author(s)

  • Sophia You (she/her/hers)

    Student
    Massachusetts Institute of Technology
    Portland, Oregon, United States

    • Primary Investigator(s)

  • GT

    Giovanni Traverso

    Associate Professor
    Brigham and Women's Hospital, Harvard Medical School, Massachusetts Institute of Technology, United States

    • Last Author(s)

  • CB

    Ceara Byrne

    Massachusetts Institute of Technology, United States


Introduction: While traditional methods of monitoring gastrointestinal motility – such as manometry, scintigraphy, or endoscopy – are typically nonsurgical, they can still be inconvenient or uncomfortable for the patient. Novel research has shifted toward pioneering non-invasive, ingestible designs due to their small size and ease of use. Since the first capsule endoscopy in 2001, wireless ingestible sensors have found numerous applications like measuring biopotentials and automated drug delivery. Proof of a free-floating, pill-sized capsule device that successfully records heart and respiratory rate was recently published. However, during the testing period, only about six hours of data were collected per trial participant, reducing the device’s viability for long-term studies. Current research therefore hopes to maximize an ingestible device’s time in vivo by extending its battery life or reducing the power needs of its sensors. One such sensor, the inertial mass unit (IMU), has been proven to record gastrointestinal peristalsis, heart rate, and respiration by measuring acceleration, angular velocity, and magnetic heading. Though examinations comparing IMU sampling rate with battery life exist, less thoroughly analyzed are the techniques for which strategic duty-cycling of sensor data can both minimize power usage and record accurate physiological metrics. This study seeks to conflate effective IMU data collection with a low-power solution and investigates how tracking peristalsis, heart rate, and respiration with an ingestible sensor can be optimized for reduced current consumption.

Materials and

Methods: The ICM-20948 IMU was selected for experimentation since its supply current in 9-axis mode was lower than that of other commercial IMUs like the MPU-9250. The ICM-20948 was connected to an nRF5340 system-on-chip using I2C communication protocol, and firmware drivers were written for its compatibility with the Zephyr real-time operating system. Two factors were manipulated during testing: sampling frequency and use of low-power duty-cycling. While consistently reliable frequencies exceed 100 Hz, recent findings suggest those as low as 25 Hz can be successfully employed to record data for physiological measurements; therefore, values between 25-100 Hz were tested. A full-factorial experiment was designed (Table 1) to determine the most power-optimizable combination of variables (data collected from the IMU at 100 Hz with both duty-cycling and low-power modes disabled was considered the standard for accurate recordings).

During every trial, the device was to be placed on a nutating mixer operating at 24 rpm to emulate gastrointestinal behavior. Acceleration values were collected over twenty minutes. To analyze the accuracy of each trial, recorded sensor data would be divided into non-overlapping, one-minute windows, and the average, minimum, and maximum measurements from each window were calculated. Graphic visualizations would be plotted against the 100 Hz, duty-cycle- and sleep-disabled trial and compared for goodness of fit. The same windowing technique would be applied to each trial’s current measurements over time. Afterward, the IMU’s optimal settings would be determined by considering the sensor data’s relative accuracy with consideration to its power consumption.

Results, Conclusions, and Discussions: At the time of submission, not all data has been collected. However, preliminary results running each trial for ten seconds suggest that between sampling frequency and use of low-power duty-cycling, the latter offers a greater reduction in overall current consumption. When low-power duty-cycling was enabled for an IMU collecting data at 25 Hz (Figure 1) , the average current consumption decreased by nearly 50% (Table 2). In contrast, the 25 Hz low-power trial demonstrated comparable current consumption to the 100 Hz low-power trial (Figure 2), albeit lower variability. These results suggest that strategic duty-cycling may be more serviceable when seeking to optimize an IMU’s overall power consumption.
Nonetheless, the full procedure outlined previously has not been completed as the collection of accelerometer data is still in progress; only by comparing the sensor data’s accuracy alongside its power efficiency can an optimal method of data collection be established. Furthermore, replication of longer trials is necessary to confirm the preliminary results’ validity. These conclusions will provide valuable insights on the feasibility of future ingestible, vitals-monitoring pills as commercial products.
Future extensions of this project include measuring the IMU’s operating time frame when powered by a biodegradable battery (use of a safely ingestible power source is necessary for the sensor’s feasibility). Furthermore, a similar experiment conducted in vivo instead of through simulated gastrointestinal movement will provide data more reflective of realistic applications.

Acknowledgements (Optional): I would like to thank members of the Traverso Lab at the Massachusetts Institute of Technology for giving me the opportunity to learn and conduct research at the lab. In particular, I want to express my gratitude toward Dr. Ceara Byrne for mentoring me and advising this research project, Ian Ballinger for his guidance on developing sensor firmware, Mark Yang for lending his expertise in data transmission, and Dr. Giovanni Traverso for allowing my use of his resources and facilities.

References (Optional): Nakamura, T., & Terano, A. (2008). Capsule Endoscopy: Past, present, and future. Journal of Gastroenterology, 43(2), 93–99. https://doi.org/10.1007/s00535-007-2153-6

Traverso, G., Finomore, V., Mahoney, J., Kupec, J., Stansbury, R., Bacher, D., Pless, B., Schuetz, S., Hayward, A., Langer, R., & Rezai, A. (2023). First-in-human trial of an ingestible vitals-monitoring pill. Device, 1(5), 100125. https://doi.org/10.1016/j.device.2023.100125

You, S. S., Gierlach, A., Schmidt, P., Selsing, G., Moon, I., Ishida, K., Jenkins, J., Madani, W. A., Yang, S.-Y., Huang, H.-W., Owyang, S., Hayward, A., Chandrakasan, A. P., & Traverso, G. (2024). An ingestible device for gastric electrophysiology. Nature Electronics, 7(6), 497–508. https://doi.org/10.1038/s41928-024-01160-w