Accuracy of Qinematic Posture Scan tracking point algorithms.

‘‘It can be stated that Qinematic’s algorithm is a highly valued and applicable device for a future health market. It can be used as a high-fidelity examination device that standardizes posture examination. Due to a short screening-time it will enable therapists and doctors to focus on treatment instead of time-consuming body screenings with serious human errors. However, more improvements towards special patient groups have to be done. This can be realized with a strong collaboration between research engineers and the company. Further improvements are also awaited through Qinematic’s machine learning algorithm that improves automatically with an increasing number of scans.’’

Download

Qinematic is a life science and technology company

Qinematic invests in a lot of clinical and technical research and development, internally and with partners. We are very fond of collaboration with universities, life & health insurers, and evidence-based service providers. Qinematic sees its role in the research community as a provider of ‘practice-based evidence’ rather than ‘evidence-based practice’. We bring big data to human movement science so that people of all ages can learn to move well and to move more using data insights, rather than following traditional dogma and trends in the health and fitness industries. We see some major challenges with traditional research and some publications, and these are also well recognized by other thought leaders:

• practitioners struggle to keep up with the pace of technical innovation, and many clinical studies are designed around old-fashioned ways of working. Furthermore, by the time a typical study is published, the technology currently in use has frequently evolved well beyond the technology used in the study. Qinematic software enables real-time and retrospective analysis of data.

• research studies are often based on too small populations and limited study design that does not reflect reality. This can indeed make publication easier (driven by a ‘publish or perish!’ attitude), but can also oversimplify the situation and thereby reach the wrong conclusions in practice. Qinematic software is fast and simple to allow easy access to movement capture outside of the lab.

• accurate movement data is often collected in artificial labs, rather than real-life situations involving practitioners and their clients. In most cases, function is merely subjectively self-reported using paper surveys rather than objective recording and measurement of actual performance. Qinematic software records and measures actual performance following standardised instructions, every time. It also uses health surveys to collect background information.

• variation in human performance is sometimes confused with variation in measurement accuracy. People can vary the way they move (like the weather), and this is exactly why we need to measure it (like a barometer). To identify and address problems, we need large amounts of data to define normal and abnormal variation in various populations of people, and under various circumstances. That is why Qinematic exists.

• for comparison, the gold standard in motion capture is expensive multicamera marker-based systems, however ‘there has been some evidence to suggest that during trials using marker-based and markerless motion capture systems concurrently, the reflective markers used for tracking marker-based assessments may distort the results gleaned from cameras used for markerless motion capture (Naeemabadi et al., 2018), therefore systems compared to marker-based approaches may be better than we currently realize’ (Armitago-Laog, C et al., 2022). Unfortunately, this has not been recognised as a limitation or source of error in many studies.

What does the research say about the Kinect sensor for measuring human motion.

Qinematic develops software to make recording, measuring and visualising human movement fast, easy and meaningful.

There have been three generations of the Microsoft Kinect sensor. Qinematic software can be used with the second and third versions - XBox Kinect and Kinect Azure. Kinect V2 (XBox) was a big improvement on V1, and in our opinion the Kinect Azure sensor is better again (not to mention better looking).

Please note that Qinematic has improved the accuracy of the body tracking using the Kinect sensor. The Kinect sensor is an impressive piece of high-tech kit, but it was originally designed by Microsoft for immersive gaming and not measuring movement. The following studies are using the native Kinect body tracking algorithms, unless stated otherwise. Also note that Qinematic software only records tasks and measurements if they are considered accurate - by the research and by our own extensive internal testing. be aware that a single Kinect sensor is not suitable for measuring all movements, like a full squat due to occlusion, or limb internal and external rotation due to low accuracy. In such cases, a multicamera system or inertial sensor system is more appropriate.

This is what some of the good quality studies say:

“The depth-sensing camera’s high rate of successful data acquisition and correlations of its hip and knee ROMs with other mobility measures suggest that this device can provide a cost-efficient means of quantifying joint motion in large numbers of community-dwelling older adults who span the health spectrum…. Overall, results of this study show that the depth-sensing camera is promising but has limitations for gait analysis. It cannot, for example, match the accuracy, temporal resolution, or functionality of a dedicated, laboratory-based gait analysis system. Such systems have very high temporal and spatial resolution and incorporate force plates to capture ground reaction forces, enabling determination of inverse kinematics. Bringing a force plate into the community setting or estimating ground reaction forces with wearable accelerometers are technically challenging endeavors….
In addition to its utility in research, depth-sensing camera technology could eventually provide patients and physicians with wider access to quantitative mobility testing, either in a primary care clinic or via telemedicine in the comfort of the patient’s home. This conceptual paradigm for easier, lower cost, and more frequent gait testing using portable devices has the potential to impact clinical practice by facilitating the detection of early signs of mobility impairment, adding easily collected functional data on joint motion to treatment decision trees (e.g. regarding joint replacements), and monitoring the progression of gait alterations and their response to clinical interventions.” Dawe RJ, Yu L, Leurgans SE, Truty T, Curran T, Hausdorff JM, et al. (2019) Expanding instrumented gait testing in the community setting: A portable, depth-sensing camera captures joint motion in older adults. PLoS ONE 14(5): e0215995.

“Excellent agreement between systems for the flexion/extension range of motion of the shin during all tests and for the thigh abduction/adduction during SLS were seen. For peak angles, results showed excellent agreement for knee flexion. Poor correlation was seen for the rotation movements.” Kotsifaki A, Whiteley R, Hansen C (2018), Dual Kinect v2 system can capture lower limb kinematics reasonably well in a clinical setting: concurrent validity of a dual camera markerless motion capture system in professional football players. BMJ Open Sport & Exercise Medicine 2018;4:e000441.

“Overall, the validity of the active system’s (dual Kinect based software) mean magnitude velocity estimates was high (compared to gold standard motion capture lab).” Finn MT, (2018) Open-Ended Measurement of Whole-Body Movement: A Feasibility Study, Quantitative Methods of Psychology

“The Kinect V2 has the potential to be used as a reliable and valid tool for the measurement of some aspects of balance performance. Concurrent validity of trunk angle data during the dynamic tasks and anterior-posterior range and path length in the static balance tasks was excellent (Pearson’s r>.75).” Clark R et al, Reliability and concurrent validity of the Microsoft Xbox One (V2) for assessment of standing balance and postural control, Gait & Posture 42 (2015) 210-213.

“The Kinect system is expected to be useful for balance training systems that require characterization of the changes in the COM, and joint angles during flexion–extension movements.” Dohyung Lim et al (2015). Use of the Microsoft Kinect (V2) system to characterize balance ability during balance training, Clinical Interventions in Aging (2015):10 1077–1083.

“The Kinect V2 has the potential to be used as a reliable and valid clinical measurement tool. Signal to noise ratio provided information about Kinect V2 landmark stability and indicated larger noise behaviour in feet and ankles. Most of the derived clinical parameters showed good to excellent absolute agreement (30 parameters showed ICC(3,1) > 0.7) and consistency (38 parameters showed r > 0.7) between both systems. Conclusion Given that this system is low-cost, portable and does not require any sensors to be attached to the body, it could provide numerous advantages when compared to established marker or wearable sensor-based system.” Otte et al (2016), Accuracy and Reliability of the Kinect Version 2 for Clinical Measurement of Motor Function, Nov 2016

“Total body center of mass (TBCM) is a useful kinematic measurement of body sway. However, expensive equipment and high technical requirement limit the use of motion capture systems in large-scale clinical settings. Center of pressure (CP) measurement obtained from force plates cannot accurately represent TBCM during large body sway movement. Kinect is a cost-effective alternative to a motion capture and force plate system for clinical assessment of TBCM sway. The Kinect device was more accurate in the medial-lateral than in the anterior-posterior direction, and performed better than the force plate in more challenging balance tasks, such as Eyes Closed whilst Standing on Foam (ECf) with larger TBCM sway. Our results revealed that the Kinect system produced a highly correlated measurement of TBCM sway (mean RMSE=4.38 mm; mean CORR=0.94 in Kinect-Vicon comparison), as well as comparable intra-session reliability to Vicon. However, the Kinect device consistently overestimated the 95% CL of sway by about 3mm.” Yueng et al (2014), Evaluation of the Microsoft Kinect as a clinical assessment tool of body sway. Gait Posture. 2014 Sep;40(4):532-8

“These results illustrate the feasibility of a single camera markerless motion capture system to accurately measure lower extremity kinematics and provide a first step in using this technology to discern clinically relevant differences in the joint kinematics of patient populations. Joint angles were calculated with marker data from both motion capture systems using successive body-fixed rotations. The sagittal and frontal plane joint angles calculated from the marker-based and markerless system agreed with inclinometer measurements by <0.5°. The systems agreed with each other by <0.5° for sagittal and frontal plane joint angles and <2° for transverse plane rotation. Both systems showed a coefficient of reliability <0.5° for all angles.” Schmidt A (2013), Accuracy and repeatability of joint angles measured using a single camera markerless motion capture system. J Biomech. 2014 Jan 22;47(2):587-91.

Scientific overview for Qinematic software and the use of Kinect in clinical practice

Download

“Measurements of the shoulder ROM using Kinect show excellent agreement with those taken using a goniometer. These results indicate that the Kinect can be used to measure shoulder ROM and to diagnose AC as an alternative to goniometer.” Seung Hak Lee (2014), Accuracy and repeatability of joint angles measured using a single camera markerless motion capture system.

“It is important to note that with this agreement between systems, there has been some evidence to suggest that during trials using marker-based and markerless motion capture systems concurrently, the reflective markers used for tracking marker-based assessments may distort the results gleaned from cameras used for markerless motion capture (Naeemabadi et al., 2018), therefore systems compared to marker-based approaches may be better than we currently realize.” Armitago-Laog, C et al (2022) A SWOT Analysis of Portable and Low-Cost Markerless Motion Capture Systems to Assess Lower-Limb Musculoskeletal Kinematics in Sport, Front. Sports Act. Living, 25 January 2022

“Body point’s time series and gait parameters obtained with a multi-Kinect v2 set-up match well with those derived with a gold standard in 3D measurement accuracy. Future studies are recommended to test the clinical utility of the multi-Kinect v2 set-up to automate 10MWT assessments, thereby complementing the time to walk 10 meters with reliable spatiotemporal gait parameters obtained objectively in a quick, unobtrusive and patient-friendly manner.” Geerse DJ, Coolen BH, Roerdink M (2015) Kinematic Validation of a Multi-Kinect v2 Instrumented 10-Meter Walkway for Quantitative Gait Assessments. PLoS ONE 10(10): e0139913

“The Kinect V2 demonstrated excellent ICC of the KASR at IC and PF during the DVJ when compared with the Vicon system. A customized Kinect V2 software program demonstrated good reliability in identifying the KASR at IC and PF during the DVJ”, Aaron D. Gray, MD, et al (2017) Development and Validation of a Portable and Inexpensive Tool to Measure the Drop Vertical Jump Using the Microsoft Kinect V2, Sport Health, Dec 2017

“Given that this system is low-cost, portable and does not require any sensors to be attached to the body, it could provide numerous advantages when compared to established marker or wearable sensor based system. The Kinect V2 has the potential to be used as a reliable and valid clinical measurement tool.” Otte K, Kayser B, Mansow-Model S, Verrel J, Paul F, Brandt AU, et al. (2016) Accuracy and Reliability of the Kinect Version 2 for Clinical Measurement of Motor Function. PLoS ONE 11 (11): e0166532