PHOTOGRAMMETRY IN MEDICAL GAIT ANALYSIS
Baumann, J.U., Meier, G., Schaer A.R.
Motion Analysis Laboratory, Dept. of Orthopedic Surgery,
University of Basel, Felix Platter-Spital, CH-4055 Basel, Switzerland.
Sheffer D.B.
Biostereometrics Laboratory, Dept. of Biomedical Engineering
University of Akron, Akron, Ohio 44325, USA.
ABSTRACT
For understanding the complex dynamics of total body movements of human locomotion, information on shape, dimensions, mass
and proportions of the 15 major body segments is needed. In physically handicapped persons, these parameters vary considerably.
Evaluation of the individual is feasible using photogrammetry. Graphic presentations of results which are easy to understand are
decisive for general acceptance of gait analysis as a useful diagnostic tool. Linking limited measurements to images of the whole
person in motion permits local measurements to be incorporated into the general information of total body movements.
A pilot study is reported and propositions for cost efficient performance using todays possibilities of electronic imaging technology
are discussed.
KEY WORDS: Photogrammetry, 3D Human Motion Analysis, Joint Forces and Moments, Anatomy Based Coordinate Systems,
Human Body Deformity.
INTRODUCTION
As a cause for medical consultation, diseases and injuries to
the organs of motricity, the neuro-musculo-skeletal system,
rank among the most frequent. Long term disabilities
affecting independance in daily living as well as working
capacity are most often caused by functional insufficiency
and pain within the locomotor system. Human beings are
three dimensional bodies who to move in space within time.
It seems astonishing therefore, that todays methods for
recording and measuring the changes in the relative spatial
positions of our 15 major body segments are still
unsatisfactory. The task is formidable, however. From the
standpoint of the photogrammetrist, the systems in practical
use for quantitative assessment of such disorders seem
inadequate. Better methods at a reasonable price could not
only support further improvements in ^ medico-surgical
treatment but also help to prevent the development of
disabling conditions. This is particularly important in a
population reaching much higher ages than only 50 years
ago.
A number of commercial systems for gait analysis are
available in which output generally consists of the recording
of the spatial positions of body segments in space as
successive frames or instances of time. Kinematic
information describing joint angle changes relative to time
and position are represented in stick figures or curves
describing the joint movements. These data coupled with the
simultaneous recording of ground reaction forces have served
as descriptions of the kinematic activitities of the human
body during ambulation. Continuous development in
methodology depending on the advancement of electronic
data collection and processing devices has tended to attempt
to move the major acquisition protocol from cinematography
to that of video based systems such as CINTEL (Winters,
1972), VICON (Macleod et al., 1990), ELITE (Ferrigno
1990). Opto-electronic systems that use position sensitive
detectors (SELSPOT, Woltring 1980), or CCD line sensors
such as COSTEL (Bianchi et al. 1990) and OPTOTRAK
(Crouch et al., 1990). These latter technologies employ non-
-television based systems of active markers and
non-simultaneous sampling of the body segment markers.
We have tried over many years to improve qualitative and
quantitative assessment of locomotion in patients and normal
persons from their first year of life to old age. Particular
focus was centered on disorders of the neural control of
movements particularly in children with cerebral palsy. on
ligamentous knee injuries in sportive individuals as well as
on effects of aging on walking patterns.
Practical experience showed, that photogrammetry and
remote sensing using telemetry for some electrical signals are
needed to evaluate the movement patterns within the
segmented bodies of children and adults under natural
conditions of daily life. Body shape and movements in space
can thus be recorded and analysed without interference to the
phenomena to be measured.
Because orthopedic surgeons should be even more interested
in forces and moments occuring at the major joints than in
the movements alone, measuring and calculating reactive
force actions must be included in many instances. For this
purpose, 3D force plates for measuring floor reaction forces
to the loading by the feet have proved to be of great help.
During walking, only one leg has floor contact most of the
time, while the other three extremities pursue pendular
swinging movements. In order to calculate the forces and
moments occuring at the hip, knee and ankle joints during
walking and running, the segmental masses which make up
the pendulums as well as the segmental centers of mass must
be known. The total mass of the extremities makes up some
44% of total body mass in a healthy young male. Their
rhythmic swinging during walking must have considerable
effect, but has so far been largely neglected by measurements
in medical movement analysis. Methods and tables for
evaluating density distribution within the different body
segments are available. Information on translational and
angular accelerations of the segmental masses in the 3
dimensions of space is needed to calculate reactive forces and
moments at the joints. Because human body proportions
change considerably from birth to adulthood and are often
highly abnormal in physically impaired person, individual
evaluations of segmental mass distribution is frequently
required. Photogrammetry seems ideal as a basis for
calculating segmental body volumes and marking average
