The HuMBL Lab specializes in high sampling rate inertial measurement unit sensors for collecting human motion during dynamic activities and impacts. Measurements during these activities are especially useful for studying injuries in real-world scenarios, and for generating life-like animations. To achieve accurate measurements during dynamic activities, we use a combination of mechanical design principles, multi-sensor fusion approaches, and machine learning. We also thoroughly evaluate our technologies in the laboratory to ensure reliability.
We are always looking for students to work in wearable sensing, from developing new technologies and algorithms to clinical application.
Motion capture technologies are necessary for a number of applications ranging from fundamental biomechanics to graphics and animation. Traditional video-based motion capture systems are restrictive due to their high hardware requirements. More recently, inertial measurement unit motion capture systems have come on the market and allow greater freedom in performing motion capture in the wild.
However, such systems lack high sampling rates and thus are incapable of capturing highly dynamic motion. Leveraging our expertise in high rate IMU sensors, we augmented traditional motion capture to more accurately estimate dynamic motions.
Kuo C, Liang Z, Fan Y, Blouin JS, Pai DK. Creating impactful characters: correcting human impact accelerations using high rate IMUs in dynamic activities. ACM Transactions on Graphics (TOG). 2019 Jul 12;38(4):1-2. Link.
Measuring human motion kinematics during dynamic activities, and in particular impacts, is a major challenge for IMU-based motion capture systems due to soft tissue artifacts. Most systems will apply heavy low-pass filters to eliminate soft tissue artifacts and capture gross human motion, but this also removes the high frequency content of highly dynamic human motion.
Another solution, is to use a network of sensors placed on a single body segment, and optimally combine sensor measurements using Bayesian estimators. While each sensor might have significant errors from soft tissue artifacts, the combined estimate can be much more accurate.
The instrumented mouthguard, originally developed at Stanford University, is an inertial measurement unit embedded device designed specifically to measure head kinematics during head impacts. The mouthguards are custom formed to couple tightly to the upper dentition, which itself is coupled directly to the skull via stiff ligaments. This eliminates artifacts introduced by soft tissue or headgear motion with respect to the skull during impacts.
As part of development, we have tested the instrumented mouthguard in live human subjects, in dummy headforms, and in cadaveric heads. Each model system has its advantages and disadvantages, but in testing the mouthguards in all systems we are confident in its capabilities. We have also assessed the mouthguards ability to accurately simulate brain tissue strains, which are the current best metrics for identifying mTBI.
Upon thorough validation, we also deployed these mouthguards to contact sports athletes (primarily American football) to study head impact exposure and mTBI mechanisms. This has helped us develop better methods of tracking head impact exposure using a combination of sensor and video verification, and automated detection algorithms. The HuMBL lab currently collaborates with SiMPL in the UBC department of Mechanical Engineering to continue development and deployment of the instrumented mouthguard.
1. Kuo C, Wu L, Zhao W, Fanton M, Ji S, Camarillo DB. Propagation of errors from skull kinematic measurements to finite element tissue responses. Biomechanics and modeling in mechanobiology. 2018 Feb 1;17(1):235-47. Link.
2. Kuo C, Wu L, Loza J, Senif D, Anderson SC, Camarillo DB. Comparison of video-based and sensor-based head impact exposure. PloS one. 2018;13(6). Link.
3. Kuo C, Wu LC, Hammoor BT, Luck JF, Cutcliffe HC, Lynall RC, Kait JR, Campbell KR, Mihalik JP, Bass CR, Camarillo DB. Effect of the mandible on mouthguard measurements of head kinematics. Journal of biomechanics. 2016 Jun 14;49(9):1845-53. Link.
4. Wu LC, Nangia V, Bui K, Hammoor B, Kurt M, Hernandez F, Kuo C, Camarillo DB. In vivo evaluation of wearable head impact sensors. Annals of biomedical engineering. 2016 Apr 1;44(4):1234-45. Link.