The ability to measure human gait is essential in many clinical contexts, including diagnosing gait deviations, optimising treatment pathways and assessing the impact of interventions.
A wide range of motion analysis technologies are available on the market, meeting different needs for validity and reliability, measurement outputs, portability, space and cost. Choosing the ‘right’ gait analysis solution for your clinic not only depends on your clinical requirements, but also on other important factors including your budget, commercial vision and space availability.
This article attempts to provide clinicians with a head-start in this decision-making process. The first section presents an overview of the gait analysis technology options available, including the main benefits and limitations of each. The second section provides a more general overview of the key clinical decision-making questions you should be asking yourself before investing in new gait analysis technology.
Part 1: Which technology is right for you?
Broadly speaking, gait analysis technologies are classified as being either two-dimensional (2D) or three-dimensional (3D).
Two-dimensional gait analysis
Two-dimensional gait analysis measures movement in two planes: side-to-side and up-and-down. It usually takes the form of pictures or videos measured in the sagittal and/or frontal planes.
Video-based systems represent the most commonly used technology for two-dimensional gait analysis and they tend to be straight forward to implement, easily accessible (everyone has a camera on their phone!) and affordable.
At its most basic, simply recording videos of a patient walking enhances a visual (observational) gait analysis by enabling the clinician to re-play the video repeatedly and in slow-motion. Whilst this method is cheap and easy to perform, it presents serious limitations as it is entirely subjective, and the quality of the analysis depends on the experience, skill and interpretation of the clinician performing it.
Video 1: Example of a 2D video analysis. In some cases another camera placed either in front or behind the camera is used to record frontal plane motion. Note that in order to record the other side of the patient, either the camera needs to be moved or the treadmill needs to be turned around, which can be cumbersome.
More advanced two-dimensional video analysis set-ups include a calibration frame, anatomical markers and video analysis software, enabling the clinician to gather objective data and calculate 2D joint angles from the anatomical landmarks. This method of assessment enables you to calculate useful and objective information about how a person is moving and delivers a far better gait analysis solution compared to simply recording a patient using your phone. However, it is not without its limitations, the most significant being that these technologies are limited to two-dimensions and do not measure movement in the transverse plane. Humans move in three anatomical planes and so important information relating to a patient’s movement could be completely missed using two-dimensional analysis.
Furthermore, measurements taken from two-dimensional projections of a three-dimensional world are impacted by ‘real-life’ rotations. What this means is that real-world movements in the transverse plane give the appearance of movement in other planes when viewed in 2D, which leads to measurement error. This error can be significant, particularly in the frontal plane where the overall range of motion is small, and risks leading to inappropriate clinical decisions.
A complete understanding of human gait requires technologies that are able to measure movement in three-dimensions.
Two-dimensional video analysis is also susceptible to a phenomenon referred to parallax error, whereby the angle measured from the video is entirely dependent on the angle and position of the camera relative to the patient. In order to calculate true sagittal and/or frontal plane motion, the cameras must be positioned exactly perpendicular to the patient, and if they are not then the reliability of the measurement is compromised. This is clearly illustrated in Figure 1, which demonstrates how different joint measurements result from cameras positioned with different angles of view.
Example 2D video-based technologies: Dartfish, Contemplas.
Three-dimensional gait analysis
Three-dimensional gait analysis measures data in three planes: side-to-side, up-and-down, and front-to-back, thereby quantifying movement in the sagittal, frontal and transverse planes, and providing a complete understanding of the way a person is moving throughout the gait cycle.
Three-dimensional motion capture technologies can be categorised into three groups: optoelectronic, inertial and marker-less.
Optoelectronic three-dimensional motion capture is the industry gold-standard and has been used for decades by biomechanics research laboratories and specialist orthopaedic hospitals throughout the world (Figure 3). These systems are frequently used alongside force-plates and EMG to gain a complete understanding of a person’s kinematics, kinetics and muscle activity during gait.
Optoelectronic 3D motion capture systems use multiple cameras to measure the 3D positions of anatomical markers as the patient walks, runs or performs an activity of choice within the capture volume. The measured marker positions are input into a musculoskeletal model, and the musculoskeletal model is used to calculate the 3D joint kinematics.
Optoelectronic 3D motion capture systems are fully validated and offer the highest level of accuracy compared to any other gait analysis technology. However, despite these systems being the most trusted gait analysis systems on the market, they are expensive, usually require a dedicated purpose-built space and can be complicated to use.
Example systems on the market: Vicon, Qualisys, MotionAnalysisCorp.
Video 2: Example of a Vicon 3D motion capture system. The infrared cameras track the 3D positions of markers attached to the patient. The marker positions are input into a kinematic model in order to calculate sagittal, frontal and transverse plane joint angles.
Inertial systems use wearable accelerometer-based sensors to measure accelerations and subsequently calculate information on velocity and displacement. This technology is becoming increasingly accepted as a useful tool in motion capture, the main advantage being that the data can be captured anywhere and the gait assessment does not have to take place within the clinic.
Whilst there have been substantial advances in inertial-based motion capture systems in recent years, there are known limitations. Inertial-based systems are susceptible to a phenomenon referred to as environmental drift, which manifests itself as a bias or an offset in the data, thereby compromising the accuracy of your results. Furthermore, these technologies are less well established in the clinical world compared to video and optoelectronic systems, which can make data processing, interpretation of the results and their clinical application more difficult.
Example inertial systems: XSens, IMeasureU.
Three-dimensional marker-less motion capture is an advancing motion capture technology, which does not require the use of anatomical markers or wearable sensors. Instead, synchronised, calibrated video-data is processed with advanced image processing and machine learning techniques in order to calculate joint kinematics.
Example 3D marker-less systems: Theia3D
Part 2: It's not all about the tech!
Before investing in any new technology, it is essential to consider some key clinical and commercial matters in order to ensure it is integrated successfully in your clinic. In this section of this article, we therefore provide a more general overview of some of the questions you should be asking yourself before investing in any of the gait analysis technologies discussed above.
What information do you need to measure?
Windt et al. [1] presented a critical decision-making framework of four simple questions to help practitioners decide whether to purchase and implement a new piece of technology.
Assuming that you have already identified a need for gait analysis at your clinic, the first step is to define the information you want to measure and to identify the technology options that satisfy your demands.
For example, if want to measure cadence and sagittal plane motion, then a 2D video-based system will suffice. If you require a more detailed understanding of the human gait cycle and need to measure rotational movements, then the 3D technology options should be reviewed.
Can you trust the data?
As discussed in Section 1 of this article, optoelectronic three-dimensional motion capture is the industry gold-standard and has been used for decades by biomechanics research laboratories and specialist orthopaedic hospitals throughout the world. These systems are fully validated, provide comprehensive data in the three planes of human movement and offer the highest level of accuracy compared to any other gait analysis technology. Whilst inertial measurement units and 3D marker-less systems are advancing technologies, they are less well validated and less widely-used in both clinical and academic settings compared to 3D optoelectronic systems.
Can take the information from a given technology and be confident in making a decision based on the data it provides?
Two-dimensional video-based systems do not measure transverse plane motion and cannot therefore measure joint rotations. However, with a good-set-up that minimises parallax error, video-analysis can provide objective outputs and 2D sagittal plane measurements have been shown to be reliable. This information certainly has a very useful role in clinical practice as long as it is interpreted with due consideration of its limitations.
Can you use the data effectively?
Once your data has been collected, the next question to answer is whether you are able to process and analyse this data effectively, and then generate results that are interpretable and clinically useful. Thankfully, scientific advancements have enabled many technologies to automate their data-processing, thereby lessening the need for a technical expert to process the data and enabling the results to be processed immediately after they have been captured, or even in real-time.
It is important to note that the time and expertise required to process motion capture data and generate clinically meaningful results varies widely with different gait analysis systems. Whilst we will not comment on specific technologies in this article, we highly recommend requesting a live demonstration from the manufacturers or better still, speaking to other practitioners who utilise the systems in which you are interested, to gain a realistic insight into their time requirements and ease-of-use.
Once the data has been processed and a report has been generated, interpreting the results can present its own challenges.
Understanding a gait report and successfully using the information to make clinical decisions presents a steep learning curve. It is therefore important to understand how you will educate everyone to use it effectively, both from the technical perspective to capture the data and from the clinical perspective in order to use the results. Again, knowledge of the training and on-going support supplied by the technology providers and understanding how much you will need to take-on yourself is highly recommended.
Can you implement the technology in your practice?
No matter how excellent any piece of technology, it will only be of clinical and financial value if you have a clear vision for how it is going to be implemented in your clinic.
Successful implementation needs careful consideration of time and spatial requirements, staff-training, as well as new service procedures, patient pathway integration and well-defined decision-making channels for understanding who will benefit from the technology and how the results will be delivered and used to inform practice.
The best technologies are useless if they are not implemented in a way that informs decision making or changes practice.
Education is key, and upskilling your team so that they fully understand the new technology will empower them, giving them the confidence they need to recommend, trust and use the technology themselves. A clinician who does not understand the rationale and benefits associated with a technology, or is nervous to use it or interpret its results, will quite simply not use it at all.
It is impossible to comment on the ‘right’ implementation strategy for any individual clinic, however reaching out to or shadowing similar clinics who have successfully implemented the technology you are reviewing will certainly be of benefit.
If the staff are not educated about the potential benefits, lack the desire to collect the data appropriately, and do not believe the technology will provide useful information, the investment is flawed before it begins [1].
Is the technology worth the investment?
Assuming that a technology satisfies all of the above, the final step is a business plan, setting-out your pricing-model and patient pathways, as well as a cost-benefit analysis to assess the financial gain or burden of implementing the new technology in your clinic. Again, there is no one-size-fits-all answer, but some basic calculations similar to those shown in the examples below can help to streamline your thought process and can serve as a useful starting point.
Example Clinic 1
Clinic 1 is an independent, private practice, which carries-out 10,000 patient consultations a year. 10% of patients are advised to have gait analysis, of which 2% do, resulting in 200 gait analysis assessments per year. The clinic charges £300 for an appointment, thereby generating £60,000 per annum. The gait analysis system the clinic uses charges £1000/month, resulting in £48,000 profit for the clinic per year, making the investment financially as well as clinically worthwhile.
Example Clinic 2
Clinic 2 is part-time practice, which carries-out 1000 patient consultations a year. 10% of patients are advised to have gait analysis, of which 1% do, resulting in 10 gait analysis assessments per year. The clinic charges £100 for an appointment, thereby generating £1000 per annum. In this case, neither the patient-numbers nor the appointment-price justify investment in gait analysis technology.
Where does Run3D fit?
The Run3D system was developed with the aim of taking the ‘gold-standard’ of gait analysis out of the lab and putting it into everyday clinical practice.
We have successfully overcome the financial, spatial and temporal limitations that have previously restricted the use of 3D motion analysis in day-to-day practice and created a real-time 3D gait analysis system based on Vicon infrared technology that is affordable and easy-to-use.
Our expanding network of progressive, forward-thinking clinicians truly sets us apart, enabling us to learn from our most successful partners, and share this insight and knowledge with new users. We recognise that it takes more than just excellent technology to make a technology clinically and commercially successful, and our training and on-going support reflects this.
If you want to find out more about Run3D, please don’t hesitate to contact us directly. We can also happily introduce you to one of our partner clinics so that you can speak to them directly.
Supplementary Reading: What does the literature say?
A quick literature search for papers comparing 2D video analysis to the gold-standard 3D optoelectronic systems generates 1000s of results, a selection of which are cited below and serve so as to substantiate some of the comments presented in this article.
In 1990, Arablad et al. (1990) presented a novel, three-dimensional model for measuring rear-foot motion during running. The authors compared the results to those measured using a two dimensional model. They concluded that the 2D angles measured from the posterior view are influenced by movements about other axes and advised using a three-dimensional model when studying rear-foot motion during running.
In 1998, McClay and Manal [2] also compared two-dimensional and three-dimensional measures of rear-foot motion during running. The authors concluded that the two-dimensional measures should be treated with caution.
Whilst McLean et al. [3] refer to three-dimensional motion analysis as the ‘gold standard’ of biomechanical assessment; their paper highlighted the considerable financial, spatial and temporal costs, which restricts the use of this technology in day-to-day clinical practice. The authors therefore set-out to investigate whether a two-dimensional video measure of frontal plane knee angle was consistent with that measured using 3D techniques. The peak knee valgus angles for three different functional activities (a side-step, a side jump and a shuttle run) were measured using both two-dimensional video and three-dimensional analysis. The 2D technology over-estimated the knee valgus angle by 26 degrees during running and by 18 degrees and 13 degrees during sidestepping and side jumping respectively. The results suggest that a 2D video-based technology should be avoided when precise and accurate measures of frontal plane knee joint motion are required.
In a similar study, Cormack et al. (2011) investigated whether 2D measures of hip and knee frontal plane kinematics are comparable to 3D measures during running. Peak knee abduction and peak hip adduction angles were consistently greater when measured using 2D video compared to a 3D infrared system. The authors concluded that 2D systems are not valid for assessing running kinematics and that caution should be taken when measuring frontal plane joint angles using 2D video systems.
References
[1] Windt, J., MacDonald, K., Taylor, D., Zumbo, B. D., Sporer, B. C., & Martin, D. T. (2020). "To Tech or Not to Tech?" A Critical Decision-Making Framework for Implementing Technology in Sport. Journal of athletic training, 55(9), 902–910. https://doi.org/10.4085/1062-6050-0540.19
[2] McClay I., and Manal K., (1998) The influence of foot abduction on differences between two-dimensional and three-dimensional rearfoot motion. Foot Ankle Int. Jan;19(1):26-31. doi: 10.1177/107110079801900105. PMID: 9462909.
[3] McLean, S. G., Walker, K., Ford, K. R., Myer, G. D., Hewett, T. E., & van den Bogert, A. J. (2005). Evaluation of a two dimensional analysis method as a screening and evaluation tool for anterior cruciate ligament injury. British journal of sports medicine, 39(6), 355–362. https://doi.org/10.1136/bjsm.2005.018598
[4] . Cormack, S., Kendall, K.D., Ferber, R. (2011). Validation of 2D Measures of Hip and Knee Frontal Plane Biomechanics During Running. Journal of Athletic Training.46(3), S163.