The challenge asks researchers to submit a walking-visualization-tool of gait sequences based on given inertial sensor-based gait recordings acquired using intelligent engineering technology, which was developed at FAU Erlangen-Nürnberg. Seeing is believing: Doctors want to see how the patient was walking - even if they cannot see the patient. Your innovative visualization tool should help medical experts to get a visual impression of gait impairments experienced by their patients - not only during ambulatory visits in the hospital but also during everyday life in the patients’ home environment. With your approach, you get the chance to significantly contribute to comprehensive medical care concepts.
Personalized Medical Engineering in Gait Disorders
Imagine … your intelligent solution supports medical experts with personalized and disease specific visualizations of gait impairments, even though the doctor is not present.
(Photo: University Hospital Erlangen)
Gait Impairment & Assessment Technology
Gait impairment is the major characteristic of many musculoskeletal or neurologic movement diseases, such as osteoarthrosis or Parkinson’s disease (PD). A reduced ability to walk substantially limits the patient´s quality of life and the related characteristics are frequently used to clinically rate staging and progression of movement disorders. Nevertheless, clinical gait assessment by the physician is rather descriptive and heavily depends on the rater’s personal experience. Hence, more objective measures of gait quality are desired.
The cyclic nature and unique biomechanical patterns of human gait make it an ideal candidate for automated movement assessment using motion sensors. Numerous studies address the technical development of motion sensors to objectively assess motor symptoms in movement disorders and the technology is about to reach clinical applicability (–).
Since mobile gait analysis solutions are not confined to laboratory environments, they allow gait-monitoring in the patient's home environment. This aspect is particularly interesting from a clinical perspective as it promises continuous monitoring as opposed to instantaneous examinations by a physician that are prone to day-to-day variation within movement impairments or time-of-day dependent symptoms. One challenging aspect, however, is the validation of these home-monitoring concepts since the usual laboratory-based validation techniques involving video or motion capture systems are prohibitive for practical and ethical reasons.
Our interdisciplinary consortium of medical, technological and engineering expertise at FAU has developed a mobile sensor-based gait analysis system providing objective gait parameters for diagnostic application and individual therapy monitoring in movement disorders ([5-7]). Thereby, this system allows for an objective rater-independent and individualized assessment of motor symptoms using inertial sensors. Thesystem consists of three building blocks:
- An inertial sensor (currently provided by Shimmer Sensing®) attached laterally below the ankle on the patient's shoe. The sensor collects inertial data and sends it via Bluetooth to a tablet for data storage and analysis.
- A software for recording of gait sequences running on the tablet offers a variety of standardized gait tests such as a 40m walk or a stop & go sequence to be selected and then supervised by the therapist. The core of the system is a specifically designed analysis framework that provides the clinician with validated gait parameters such as stride length, gait speed, foot clearance, heel strike angle, etc. for each standardized clinical gait test [7–10]. Moreover, it provides the basis for learning gait signature characteristic for specific motor symptoms or movement disorders using large amounts of clinically labeled gait recordings and machine learning approaches.
- The recorded data as well as the analysis results can be transmitted to a secure online platform so that the clinicians can get a full overview of all gait parameters for individual, longitudinal monitoring of their patients. Furthermore, the platform is designed for multicenter applications enabling both integration of standard care units and multicenter study support.
The Research Challenge
(Image: Julius Hannink (FAU), based on vector created by Freepik)
Dataset as basis for the challenge
You will receive 10 gait recordings (sensor-derived raw data) from healthy subjects performing a standardized gait test. It’s your challenge to develop a visualization showing the movement of the lower extremities that mimics the provided video footage based on the inertial sensor data recorded at the subjects' feet. The motivation for this is the lack of validation techniques in at-home monitoring scenarios. Together with you, we want to evaluate a sensor-based replay of gait recorded in unconstrained environments. As a final goal, this replay will be rated by experienced clinicians and compared against an automated assessment.
To receive the dataset, please contact us via email (email@example.com) or via the platform.
What if we win?
The best 3 proposals will be invited to Erlangen in order to present their concept on movement visualization based on the gait data in the healthy control subjects. You will meet the local researchers during a 1-week visit and the best concept will be selected.
The winner of the research challenge will get access to gait recordings from patients with neuromuscular disease that show specific forms of gait impairment. Now, it's your time to challenge the clinicians. Based on the patient data, you present them with visualizations of the gait patterns for different neuromuscular diseases. The experienced clinicians now have to diagnose the underlying neuromuscular disease based on your visualization.
Please submit your visualization (videos), a short description of your project, and the project code here on the platform. The official end of the challenge was July 28, 2017 BUT: we are going to accept late submission until Friday, August 4, 2017.
- Visualization: Please submit as a result of your work a video that shows a shoe /an avatar moving along a trajectory (given by the provided inertial measurement unit data).
- Description: Shortly explain how you solved the visualization task (please describe and reference your methods).
- Project code: Please provide your source code (e.g. Matlab, Python or similar format). Take care for sufficient documentation so that your code is fully comprehensible to us, and please make sure that your submitted code will compile with no errors (also on our machines).
For further information and data, please contact
 B. Mariani, C. Hoskovec, S. Rochat, C. Büla, J. Penders, and K. Aminian, “3D gait assessment in young and elderly subjects using foot-worn inertial sensors,” J. Biomech., vol. 43, no. 15, pp. 2999–3006, Nov. 2010.
 J. R. Rebula, L. V. Ojeda, P. G. Adamczyk, and A. D. Kuo, “Measurement of foot placement and its variability with inertial sensors,” Gait Posture, vol. 38, no. 4, pp. 974–980, Sep. 2013.
 A. M. Sabatini, “Quaternion-based strap-down integration method for applications of inertial sensing to gait analysis,” Med. Biol. Eng. Comput., vol. 43, no. 1, pp. 94–101, 2005.
 D. Trojaniello, A. Cereatti, E. Pelosin, L. Avanzino, A. Mirelman, J. M. Hausdorff, and U. Della Croce, “Estimation of step-by-step spatio-temporal parameters of normal and impaired gait using shank-mounted magneto-inertial sensors: application to elderly, hemiparetic, parkinsonian and choreic gait,” J. Neuroeng. Rehabil., vol. 11, no. 1, p. 152, 2014.
 J. Klucken, J. Barth, P. Kugler, J. Schlachetzki, T. Henze, F. Marxreiter, Z. Kohl, R. Steidl, J. Hornegger, B. M. Eskofier, and J. Winkler, “Unbiased and Mobile Gait Analysis Detects Motor Impairment in Parkinson’s Disease,” PLoS One, vol. 8, no. 2, 2013.
 J. Barth, J. Klucken, P. Kugler, T. Kammerer, R. Steidl, J. Winkler, J. Hornegger, and B. Eskofier, “Biometric and mobile gait analysis for early diagnosis and therapy monitoring in Parkinson’s disease,” 2011 Annu. Int. Conf. IEEE Eng. Med. Biol. Soc., pp. 868–871, 2011.
 S. Schülein, J. Barth, A. Rampp, B. M. Eskofier, J. Winkler, K. G. Gaßmann and J. Klucken, "Objective gait analysis: A novel approach to detect gait improvement by a wheeled walker in geriatric patients." J Neuroeng Rehab accepted 2017.
 J. Barth, C. Oberndorfer, P. Kugler, D. Schuldhaus, J. Winkler, J. Klucken, and B. Eskofier, “Subsequence dynamic time warping as a method for robust step segmentation using gyroscope signals of daily life activities,” in Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS, 2013, vol. 2013, pp. 6744–6747.
 A. Rampp, J. Barth, S. Schülein, K.-G. Gaßmann, J. Klucken, and B. M. Eskofier, “Inertial Sensor Based Stride Parameter Calculation from Gait Sequences in Geriatric Patients.,” IEEE Trans. Biomed. Eng., vol. 62, no. 4, pp. 1089–1097, 2014.
 C. M. Kanzler, J. Barth, A. Rampp, H. Schlarb, F. Rott, J. Klucken, and B. M. Eskofier, “Inertial Sensor based and Shoe Size Independent Gait Analysis including Heel and Toe Clearance Estimation,” Eng. Med. Biol. Soc. (EMBC), 2015 Annu. Int. Conf. IEEE, pp. 5424–5427, 2015.