MOVEMENT ANALYSIS AT BAMBINO GESÙ
BY DR. PAOLO CAPPA
In October 1998 a new Paediatric
Neuro-rehabilitation Division was established at the Bambino Gesù Children's
Hospital,* Palidoro.
The laboratory multidisciplinary team is co-ordinated by the Department Director
of the Paediatric Neuro-rehabilitation Division, Prof. Marcello Mario Pierro,
and includes the following people:
* Dr. Giuseppe di Rosa, MD physician at 50% of his working time.
* Mr. Maurizio Petrarca, therapist at 100% of his working time.
* Dr. Paolo Cappa, Italian Degree (five years), with honours, in Mechanical Engineering, Associate Professor in "Mechanical and Thermal Measurement" at the Department of Mechanics and Aeronautics, University of Rome "La Sapienza".
* Dr. Fabrizio Patanè, Italian Degree (five years) in Mechanical Engineering PhD student in "Mechanical Measurement in Engineering" .
The Vicon Laboratory started in July 2000 after the installation of the first Vicon 512 motion analysis system in Italy. Currently, the system is the heart of numerous research projects. The system, Figure 2 is equipped with:
* 6 Vicon cameras with a maximum
frame rate of 240 Hz (red visible probe);
* 2 six-component AMTI 1000 force platforms;
* 8 channel-electromyograph Noraxon FM-transmitter;
* 1 digital video camera 3ccd. The signal, digitised by a PC, is stored with
other data.
The Movement Laboratory of Bambino Gesù Children's Hospital includes a Video Laboratory which is complementary to the Vicon system. This provides a digital movie, and is used to evaluate a considerable number of patients - in fact 20 each week. Up to last year the laboratory has completed about 600 valuations.
Two physiotherapists are included in the team: Mr. Marco Zazza working full time and Mr. Luciano Colibazzi working half time in the laboratory.
CLINICAL ACTIVITIES
The most important problems to solve in rehabilitation are: (a) the evaluation
and assessment of impairments, and (b) the estimation of the best rehabilitation
action. More importantly, the main task of the first mentioned problem is to
identify the disablement features of the patient, while in neurological rehabilitation
the main task is to understand the pattern used by patients during their natural
movement, so as to assess what kind of therapeutic processes could efficiently
affect them.
The approach in the laboratory is determined by trying to answer the following critical questions, "what can we measure?" and "what can we study?", in order to understand the pathological gait or movement pathways in the three sectors: (a) clinical, (b) informative-educational and finally, (c) movement resource management. The clinical assessments are relevant to: diagnosis, prognosis, degree of severity, treatment and outcome assessment, staff communication, research, internal audit and epidemiology. In the informative-educational sector we seek to improve the dialogue with patients, parents, and those involved in patient care and welfare. In terms of movement resource management, instead, we evaluate the efficiency, analyse the cost/benefit relationship, and estimate the use of resources.
The common task is to give intelligible, verifiable, up-to-date and, above all, quantitative answers. These are essential requirements to instruct patients and parents about what rehabilitation medicine can provide, and the range of advantages to patients and the whole community.
The typical procedure adopted in a clinical evaluation of pathological gait is a preliminary analysis conducted in the Video Laboratory where: (a) a qualitative analysis of locomotion, grasping, etc. is carried out and (b) the most significant frames are printed to carry out an image case history. Finally, from this archive the most interesting cases are selected for a more exhaustive analysis to be conducted in the Vicon Laboratory.
Up to now the Vicon Laboratory has carried out 140 studies, having observed both gait and upper limb activity. The routine analyses are determined with total body reconstruction, making use of the PluginGait marker set. We primarily study the following pathologies: emi-paresis; spastic and distonic tetra-paresis; para-paresis; peripheral neuropathy; brain injury and ataxia.
We also have begun a series of clinical research projects, including:
* A study of the effect of GH hormone on patients affected by Prader Willi Syndrome;
* A study of the effect of the inoculation
of botulinum toxin on the hypertonic muscles of the lower limb in
children with encephalopathies (see Figures 3-5);
* A study on the effect of trazodone in children affected by cerebral palsy;
* Evaluation of an optoelectronic-based gait analysis system when used to study obese children;
* Evaluation of the effectiveness of a rehabilitation model for the stimulation of missing motor impulse components (Grimaldi);
* Evaluation of gait modification in healthy and pathological children using AFO (Ankle Foot Orthosis)
BIOENGINEERING ACTIVITIES
When designing our gait analysis laboratory, a 6-camera configuration was selected
to ensure sufficient visibility of the marker trajectories. This was carried
out because the specific condition of constraints is particularly critical in
children affected by abnormal walking and at the same time unnecessary to handle
too much video data. The video cameras currently operate at 120 Hz to obtain
the best vertical resolution at the highest possible frame rate.
The sampling frequency of the multi-component force platforms depends on the EMG sampling because the a/d converter cannot gather the data source at various acquisition rates. Even though the frequency of 3kHz implies an oversampling of the force platform signals for the routine clinical analysis, this oversampling could be useful in many research fields to apply, for example, some data manipulation operations such as digital filtering or correlation analysis between EMG and other analog sources available. The relevant research will be conducted in the near future.
We would like to briefly describe a research project that has recently been concluded. It was devoted mainly to improving the effective clinical interpretation of collected data. An in-house post-processing procedure has been developed, verified and currently included in the usual gait analysis reports. Actually, the huge amount of data made available by the Vicon hardware and software causes some difficulties with clinical evaluation in obtaining synthetic reports on the patients examined. As is well known, the data available just for a lower limb investigation are: 28 3-dimensional data and 20 analog data (two force platforms and 8 EMG sources). Thus, it was decided to utilise the output provided by the Workstation package, i.e. the c3d data raw files, and these files are processed by means of a Matlab program. The main steps are here briefly described as follows:
* File import from c3d files and conversion in Matlab stream data variables, with auto-recognition of markers and analog labels.
* Auto-selective numerical filtering, depending on the type of signal being handled, i.e. EMG, force platform signal, angular displacement, angular velocities, moments, etc..
* Individualisation of step cycles and averaging in order to create the "mean step".
* Visualisation of the manipulated data.
As a general constraint at the developing stage of the package the friendliness of the User Interface is emphasised so that the package could be an effective diagnostic tool for physicians as well.
More precisely it is currently an effective tool in EMG representation, see Figure 4. In the figure two graphs are depicted: the first is the variation of the left ankle angle (in the sagittal plane) as a function of step percentage. The latter graph is a contour signal of muscle activity relative only to the ankle. The benefit of that representation is an easy interpretation, not only of the muscle activity but also the rate and "quality" of this activity. With the in-house developed software, it was decided to implement a tool to view the variations of all signals available, not only as a function of time or step percentage, but also as a function of each other, i.e. "the symmetry indicator diagrams". See Figure 5, in which the pathological step symmetry is easily visible and quantified by the difference in the patient curve from the reference normal curve i.e. the bisect.
The reason for developing this tool is primarily to facilitate a quick estimation of symmetry variations in gait which occur due to pharmacological (Botulinum Toxin A, Trazodone...), surgical or rehabilitation treatment. In fact, the previously mentioned estimation emerges from a comparative examination of the graphs obtained prior to and following the treatment.
The package described is supported by start-up software, a simple database in which the user can simply select the patient and view either the single session file or that calculated from the automatic-averaging of several nominally identical sessions.
* The Bambino Gesù Children's Hospital (about 800 bed medical facility), is a private and non-profit-making hospital located in the Vatican City, i.e. the independent Papal state within the city of Rome (Italy), and is officially recognised by the Italian Government as a "Research and Care Institute of a Scientific Nature".