Videographic Study and Objective Classification of Gait in Patients with Neurological Diseases


Estudio Videografico y Clasificacion Objetiva de la Marcha en Pacientes con Enfermedades Neurologicas


Carlos Diaz Novo1, Arquimedes Montoya Pedron2,  Osiel Gamez Rodriguez2,

Larisa Zamora Matamoro1, Manuel Barbaro Cuadra Sanz3, Angel Valentin Molina Mojica3*


1Universidad de Oriente. Santiago de Cuba, Cuba

2Hospital Clinico Quirurgico Gineco-Obstetrico Juan Bruno Zayas, Santiago de Cuba, Cuba

3 Universidad ECCI, Bogota, Colombia


How to cite: Diaz, C., Montoya, A., Garcia, G., Gamez, O., Zamora, L., Cuadra, M., Molina, A., Videographic Study and Objective Classification of Gait in Patients with Neurological Diseases., TECCIENCIA, Vol. 14 No. 26, 29-36, 2019




Received: 15 Jun 2018                                                             Accepted: 23 Oct 2018                                        Available Online: 1 Mar 2019




Using videography, a cross-sectional descriptive-multivariate study was carried out, consisting of 61 patients with clinical, neurological and genetic diagnosis of Spinocerebellar Ataxia (SCA) and Cerebrovascular Accidents (CVA), as well as 51 healthy individuals chosen at random, attended in the Neurology consultation of the General Teaching Hospital Dr. Juan Bruno Zayas Alfonso in Santiago de Cuba. The results of the metrological evaluations performed on the instrumental proposal are included in the work, and the behavior of the kinematic variables (gait speed, excessive increase in step width and lateral oscillations of the hip) are analyzed, for 3 manifestations of the movement of the groups of healthy and sick subjects. Finally, the discriminant linear models obtained for different groups of healthy and sick subjects whose validation reveals their potential usefulness for the study of normal and pathological movement are included.


Keywords: Biomechanics, Human Movement Disorders, Gait Cycle, Cerebellar Ataxia, Videographic Techniques, Human Movement Kinematics.





Empleando la videografia se realizo un estudio descriptivo-multivariado transversal y de clasificacion, compuesto por 61 pacientes con diagnostico clinico, neurologico y genetico de Ataxia Espinocerebelosa (ASC) y Accidentes Cerebrovasculares (ACV), asi como 51 individuos sanos escogidos al azar, atendidos en la consulta de Neurologia del Hospital General Docente Dr. Juan Bruno Zayas Alfonso de Santiago de Cuba. Se incluyen en el trabajo los resultados de las evaluaciones metrologicas realizadas a la propuesta instrumental y se analiza el comportamiento de las variables cinematicas (velocidad de marcha, aumento desmesurado del ancho de paso y de las oscilaciones laterales de la cadera), para 3 manifestaciones del movimiento de los grupos de sujetos sanos y enfermos. Finalmente se incluyen los modelos lineales discriminantes obtenidos para diferentes grupos de sujetos sanos y enfermos cuya validacion revela su potencial utilidad para el estudio del movimiento normal y patologico


Palabras clave: Biomecanica, Trastornos del movimiento humano, Ciclo de Marcha, Ataxia Cerebelosa, Tecnicas Videograficas, Cinematica del Movimiento Humano.



1.     Introduction


Virtually all pathologies of the nervous and musculoskeletal systems are manifested through movement dysfunctions. Autosomal dominant hereditary ataxias form a heterogeneous group of neurodegenerative conditions characterized by progressive neuronal loss of the cerebellum, brainstem nuclei, and spinocerebellar tracts [1].


Within this family, type 2 cerebellar ataxias (SCA2) are the most frequent in Cuba, with the main clinical manifestations the ataxia of gait, dysmetria, dysarthria and adiadochokinesia, and anticipating in all cases the age of onset of the disease in 6 years compared to previous generations [2]. Another distinctive feature of the disease is the progressive deterioration of the gait and mobility of the person who suffers from it until death, making it a serious health problem.


Since there is no cure, an important part of the therapeutic treatment of SCAs focuses on rehabilitation [3] [4] [5] and there are few studies carried out to clarify the causes of their kinematic manifestations, which is due to factors related to the micro-localization of cases worldwide and factors associated with difficulties with homogeneity in patient recruitment and the diversity of methods used in the biomechanical studies of movement disorders, which has led in many cases to draw contradictory conclusions [6] [7] [8] [9].


In Cuba, cerebrovascular accidents (CVA) rank third as the cause of general death, the first cause of death of neurological origin, as well as the most frequent basic cause of death among the deceased. They generate a high cost and expense of material and financial resources.

This is due to severity, length of hospitalization and rehabilitation time [11] [12]. Associated with the anatomical location and magnitude of the CVA, patients frequently present sequelae related to biomechanical disorders associated with the execution of movements, which complicate the process of walking, contribute to the loss of balance and cause falls [13] [14].


From a diagnostic point of view, visual analysis of patients with complex movement disorders is limited. Typically, the evaluation is conducted with a certain degree of subjectivity, through qualitative assessments involving clinical scales and the criteria of medical experts [15] [17].


In addition to the above problem, the fact that the uniqueness of human movements depends on the socio-economic and life conditions of the populations of individuals, and currently have a quantitative method for the evaluation of kinematic disorders of the march, which hinders the planning of rehabilitation actions that should be applied.


2. Methodology


The methodology followed for the development of the study carried out consisted of 3 stages. The first focused on the assembly and metrological validation of a videography system adapted to the study of motor disorders in patients in a hospital setting [18].


The second stage included the analysis and descriptive-multivariate evaluation of 3 movement structures executed by groups of healthy and sick subjects under laboratory conditions and the third stage focused on obtaining and validating gait models and classifiers from the kinematic data of the groups of people studied (figure 1). 



Figure 1 Methodological proposal. (In graphic: * Laboratory proposal for the analysis of human movements. * KINEMATIC DATA: Healthy, Sick Ataxia, Sick ICTUS. * STATISTICS: Descriptive, Multivariate, Dimensionality reduction. *Classification of the subjects vs Kinematics of the march. *Characterization of the effect of soft tissue artefacts in measurements. *Characterization of measurement system errors. * Validation)



Figure 2 Measurement set, composed of anatomical markers, spatial distribution of cameras and computer for motion capture.



The passive markers used were positioned on the surface of the body according to different spatial configurations and their trajectories were obtained by means of an ATI model 9600 video capture card, with 64 MB of video memory and a spatial resolution of 240 x 480 pixels which was coupled to a computer. It was worked at a frequency of frames corresponding to (30 c/s) and a shutter ratio of 1/500 seconds.


      2.2 Metrological validation


For the metrological validation of the instrumentation used, an electromechanical device was designed and constructed from which the dimensional errors of the study volumes in each of the laboratories employed were obtained and analyzed.


This task required a multifactorial study in which speed, diameter and inter-marker distances were analyzed to determine the precision and accuracy of the measurements made with the videography system, as well as the simulation of the occlusion of the marker trajectories by means of the rotation of a cluster of markers placed at the end of the arm of the built device (figure 3).


Absolute errors (ABSE) were calculated as the difference between the inter-marker length of the pattern used, measured with a digital caliper (precision 0.001 m) and the one obtained after the digitization of the cluster markers with the videographic technique (equation 1). Similarly, the mean of the errors was used as a reference to evaluate the accuracy of the measurement system [19].



(In the equation: Distance patterns)


Where: X2, X1, Y2, Y1, Z2, Z1 are the coordinates in the 3D space of the markers.


2.3 Protocol for the evaluation of the kinematics of movements


A cross-sectional descriptive study of 33 patients (16 men and 17 women) with clinical-neurological diagnosis of spinocerebellar ataxia was performed [20]. The sample also included 28 patients with CVA, 10 men and 18 women with clinical and imaging diagnoses of occlusive episode in the middle cerebral artery, attended at the General Hospital Dr Juan Bruno Zayas Alfons, from January 2000 to December 2016.


Figure 3 Device for metrological characterization and configuration of the cluster of markers used.


Table 1 Groups of patients included in the study.


Number of


16 Men SCA

10 Men CVA

18 Healthy Men

17 Women SCA

18 Women CVA

33 Healthy Women














Age (years)













Height (m)













Weight (kg)














Figure 4 Evaluated movement structures, (a) gait, (b) sitting-standing up-sitting of a chair, (c) plant- extension with the tip of the foot.


Likewise, it was included a sample of 51 healthy individuals older than 18 years (18 men and 33 women, with an average age of 51 and 54 years, respectively), randomly chosen, with no family and consanguineous relationship with the patients, without a history of having suffered from diseases related to movement disorders or accidents (Table 1).


For the quantitative evaluation of kinematic disorders, the participants (healthy and sick) were filmed during the realization of 3 structures of movement, gait (a), sitting-standing up-sitting (b) and plant-extension with the tip of the foot (c) (Figure 4).


Participants were asked to perform the 3 motor tasks routinely on 30 occasions. The gait was carried out in a straight path of 10 m in both directions and a regulated chair was provided to guarantee a 90 bending angle in the knee during the sitting position.


Likewise, spherical markers of 0.020 m in diameter (covered with reflective material) were placed in different locations (sacrum, iliac crests, mid-lateral portions of the thighs and triceps sural, external condyles of the femur, lateral malleoli, posterior back) of both calcaneus, fifth metatarsal and greater toe of both feet) [21] [22] [23] [24] [25] [26] [27].


Patients who required technical aids to perform the movements, had heart and respiratory affections, musculoskeletal conditions and a proven history of accidents and falls, were excluded from the study.


2.4 Protocol for the classification of normal and pathological motor skills


A multivariate cross-sectional study was conducted using Principal Component Analysis (PCA) as a dimensional reduction technique, the results of which allowed the objective identification of the motor activity that contained most of the kinematic disorder, reflected in the greater amount of altered kinematic variables (differed significantly) with respect to normal behavior.

From the kinematic data of the selected motor activity, the classification models for each age group were obtained, as well as for the 2 pathologies, they were subjected to a cross validation with data of healthy and sick individuals who were not part of the initial design of obtaining said models [28].


Dimensionality reduction and the classification was made using Minitab R16 software and Mathlab R2010b software, and for the acquisition, pre-processing, calculation and filtering of the kinematic registers, the Human software of HMA Technologies Inc. was used.


3. Results and Discussion


3.1 Metrological characterization of videographic technique


The metrological characterization of the videographic technique used for the analysis of movements was made based on the analysis of measurement errors in 3 laboratories.


The precision validated with the videographic technique poses error magnitudes of less than 12 % with respect to the variable of lesser metric magnitude, the step width, of approximately 110 mm. Similarly, the fact that the low rotation speeds of the marker cluster influenced the increase in precision (fast speed - errors of 0.0053 m; slow speed - errors of 0.0049 m for the inter-marker distance of 0.3 m), corresponds to the results described by other authors.


As the speed increased, there was a loss of resolution of the marker contour, resulting in poor location of the marker center with the pattern recognition tools of the automatic digitalization software and of the technician when manual digitalization was used [19].   A similar behavior to the mean of the errors (2 to 3 mm) was observed between the Canadian laboratories for 2 speeds of movements (fast 72 rpm and slow 27 rpm) and in the case of the laboratory of the General Hospital "Dr Juan Bruno Zayas Alfonso" a difference of around 5 mm was obtained between the mean of the errors, registering for the slow speed 10 mm and 15 mm for the fast one, (figure 5).


Likewise, the study of the distribution of errors within the study volume (2.5 m x 1 m x 1 m), proposed for the biomechanical evaluation of the locomotor apparatus of patients with SCA and CVA, showed that the central portion (with respect to the orientation in the x-axis) of the study volume of the laboratory of the General Hospital of Santiago de Cuba corresponds to the region of highest precision of the measurements, which ranged between 10 mm and 12 mm for inter-marker distances that varied between 0.3 m and 0.1 m, (Figure 6).

Figure 5 Mean values of errors for factor interaction, laboratory and speed, 0.1 m inter-marker distance. (In graphic: X-Axis: Average errors x 10-3 (m).

Y-Axis: Laboratories. Conventions: Speed, Fast/Slow)



Figure 6 Distribution of the Mean values of errors on the [x] axis (0.3m, 0.2m, 0.1m) of the capture volume of the Santiago de Cuba laboratory, for 3 inter-marker distances, 0.3 m, 0.2 m and 0.1 m. (In graphic: Axis (X) x 10-3 (m). Y-Axis: Mean of the errors x 10-3 (m)).



3.2 Biomechanical characterization of movement disorders in healthy subjects and patients with SCA and CVA.


For the 3 movement structures evaluated, the results of the descriptive study and the comparison of kinematic patterns in men and women, healthy and sick with the 2 diseases showed a decrease in movement speeds (gait, ascent and descent), the cycle and step lengths with an increase in gait width and lift base, lateral oscillations of the hip and of the times (of support, of cycle and of double support). See Figure 7 for results of locomotion in SCA.


In the 3 structures of movements evaluated, the results of the descriptive study and comparison of kinematic patterns in men and women, healthy and sick with SCA and CVA, corroborated the need of patients to practice a slow and insecure gait,

Figure 7 Behavior of kinematic parameters of gait, mean and confidential interval (95%) in men and women, healthy and sick with SCA. (In graphic: 1st graphic. X-Axis: Mean speed of the march, Speed range of the march. Y-Axis: MEAN-IC, speed (m / s). 2nd graphic. X-Axis: Cycle length, Step length. Y-Axis: MEAN-IC, length (m). 3rd graphic. X-Axis: Support, Swinging, Cycle, Double support. Y-Axis: MEAN-IC, time (s)

This gait is characterized by the generation of small ranges of movements and an exaggerated increase in spatial parameters related to stability, aspects that coincide with that reported in the literature [20].


Once the PCA technique was applied in the groups of patients, it was evidenced that the gait was the physical activity that best described the movement disorders present in the groups of men and women, healthy and sick with SCA and CVA.


In the 3 structures of movements evaluated, the results of the descriptive study and comparison of kinematic patterns in men and women, healthy and sick with SCA and CVA, corroborated the need of patients to practice a slow and insecure gait.


This gait is characterized by the generation of small ranges of movements and an exaggerated increase in spatial parameters related to stability, aspects that coincide with that reported in the literature [20].


Once the PCA technique was applied in the groups of patients, it was evidenced that the gait was the physical activity that best described the movement disorders present in the groups of men and women, healthy and sick with SCA and CVA.


3.3 Behavior of discriminant linear models


CSick = -749.16-2085.72*VM+899.64*LC 519.08*LP+626.12*ROYC334.0*APie+2.34*RRR+2789.52*Cad+758.63*TA+495.73*TDA                            



CHealthy = -749.92-2019.83*VM+894.16*LC-551.02*LP+595.75*ROYC-220.43*APie+2.37*RR+2741.66*Cad+772.93*TA+465.02*TDA                                                                                                                                                              (3)

In the group of healthy and sick men with SCA the model allowed the classification of 94.12% of the individuals, with a 0.775% correlation and a high level of statistical significance (p<0.05).


Male sex SCA:


CSick = -443.338-811.980*VM +300.775*LC+2054.474*Cad+1352.176*TDA                               



CHealthy = -501.479-853.921*VM +318.902*LC+2191.133*Cad+1446.782*TDA                            



Also, in the group of women with CVA, a discriminant model was obtained that allowed the classification of 100% of healthy and sick women with CVA, with a 0.921% correlation and high level of statistical significance (p<0.05).


CSick = -21.5391-2.2032*VM +195.2354*RVM+208.5983*APie                                   (6)


CHealthy = -44.9967+71.6720*VM +41.5522*RVM+411.3186*APie                                     (7)


In the group of healthy and sick men with CVA, the model allowed the classification of 100% of the individuals, with a 0.97% correlation and a high level of statistical significance (p<0.05).


Male sex with CVA:


CSick = -134.745+280.693*VM +698.333*RVM-388.638*APie+853.374*Cad                                            (8)


CHealthy = -270.512+529.014*VM +935.271*RVM-685.671*APie+1167.551*Cad                                                      



The validation of the discriminant models was made from the kinematic parameters of the gait obtained from 3 patients with SCA (2 women and 1 man), randomly selected, whose kinematic parameters were not used to obtain the models.


Similarly, 5 subjects with CVA (2 women and 3 men) were randomly selected whose kinematic parameters once introduced in the corresponding models allowed the correct classification of all of them.


4. Conclusions


The study conducted, based on the use of the videographic technique, the evaluation of kinematic parameters of the locomotor system, related to the execution of 3 physical activities of daily life, allowing the analysis of kinematic parameters in samples of healthy and sick individuals with SCA and CVA.


The results of the study of the 3 movements in the 2 groups of patients with SCA and CVA, revealed that the movement disorders were characterized in a general way, by the decrease of the spatial parameters of the partial realizations. The most important parameters are step length, cycle length and the maximum height of the food, combined with the increase in the execution times of each of the particular phases.


The metrological characterization of the laboratory for the analysis of human movements, allowed to obtain the dimensions and distribution of the errors within the volume of study. The validation of the method was carried out using the theoretical model obtained in clinical protocols oriented to the study of 2 neurological diseases.


Videographic measurements of the gait allowed to obtain the discriminant functions for the classification of groups of healthy and sick people with SCA and CVA, which has made it possible to have the first quantitative tools aimed at complementing the evolutionary monitoring of motor skills.




[1]      Ilg W, Bastian AJ, Boesch S, Burciu RG, Celnik P, Claaben J, et al. Consensus paper: management of degenerative cerebellar disorders. Cerebellum. 2014; 13(2):248-68.

[2]      Velazquez Perez L, Rodriguez Labrada R, Canales Ochoa N, Medrano Montero J, Sanchez Cruz G, Aguilera Rodriguez R, et al. Progression of early features of spinocerebellar ataxia type 2 in individuals at risk: a longitudinal study. The Lancet Neurol. 2014; 13(5):482–9.   

[3]      Cakrt O, Vyhnalek M, Slabý K, Funda T, Vuillerme N, Kola P, et al. Balance rehabilitation therapy by tongue electrotactile biofeedback in patients with degenerative cerebellar disease. NeuroRehabilitation. 2012; 31(4):429-34.

[4]      Keller LJ, Bastian AJ. A home balance exercise program improves walking in people with cerebellar ataxia. NeuroRehabil Neural Repair. 2014 [citado 21 Abr 2015]; 28(8). Disponible en:

[5]      Miyai I, Ito M, Hattori N, Mihara M, Hatakenaka M, Yagura H, et al. Cerebellar ataxia rehabilitation trial in degenerative cerebellar diseases. NeuroRehabil Neural Repair. 2012; 26:515-3.

[6]      Saute Morales AJ, Donis Carvalho K, Munuera Serrano C, Genis D, Ramirez Torrez L, Mazzetti P, et al. Ataxia rating scales-psychometric profiles, natural history and their application in clinical trials. Cerebellum. 2012; 11(12):488-504.

[7]      Morton SM, Bastian AJ. Relative contributions of balance and voluntary leg coordination deficits to cerebellar gait ataxia. J Neurophysiol. 2003; 89(4):1844-56.

[8]      Palliyath S, Hallett M, Thomas SL, Lebiedowska MK. Gait in patients with cerebellar ataxia. Mov Disord.1998; 13:958–64.

[9]      Deuschl G, Stolze H, Petersen G, Raethjen J. The gait disorder of advanced essential tremor. Brain. 2001; 124:2278-86.

[10]    Stolze H, Klebe S, Petersen G, Raethjen J, Wenzelburger R, Witt K, et al. Typical features of cerebellar ataxic gait. J Neurol Neurosurg Psychiatry. 2002; 73:310-12.

[11]    Ministry of Public Health. National Directorate of Medical Records and Health Statistics. Statistical Yearbook of Health 2011. Havana: MINSAP; 2012. quoted 12 Jun 2012]. Available at:

[12]    Llibre-Guerra J.C., Valhuerdi Cepero A., Fernandez Concepcion O., Llibre-Guerra J.J., Gutierrez R.F. Llibre-Rodriguez J.J. Stroke incidence and risk factors in Havana and Matanzas, Cuba Neurology. 2015; 30(8):488—495.

[13]    Ricci A. N, Ferrarias P., G., Molina I. K., Dib M. P., Alouche R. S. Gait speed and falls self-efficacy in individuals with hemiparesis after stroke. Fisioter Pesq. 2015; 22(2):191-196.

[14]    Walsh M., Galvin R., Horgan Frances N. Fall-related experiences of stroke survivors: a meta-ethnography J. Disability and Rehabilitation. 2016; 39:631–641.

[15]    Metz G, Coppard N, Cooper JM, Delatycki MB, Drr A, Di Prospero NA, Giunti P, Lynch DR, Schulz JB, Rummey C, Meier T. Rating disease progression of Friedreich's ataxia by the International Cooperative Ataxia Rating Scale: analysis of a 603-patient database. Brain. 2013; 136(1):259-68.

[16]    Mansour Y. O., Megahed M. M., Abd Elghany E. Acute ischemic stroke prognostication, comparison between Glasgow Coma Score, NIHS Scale and Full Outline of UnResponsiveness Score in intensive care unit. Alexandria Journal of Medicine. 2015; 51:247–253.

[17]    Yew S. K., Cheng M. E. Diagnosis of Acute Stroke. Am. Fam. Physician. 2015; 91(8):528-536.

18]     [Online] Medical Biophysics Center: Forging "healthy" dreams [Access 14 June 2017] Available on the Web:

[19]    Novo Diaz C, Alharbi S, Fox M, Ouellette E, Biden E, Tingley M, et al. The impact of technical parameters such as video sensor technology, system configuration, marker size and speed on the accuracy of motion analysis systems. Ing Mec, Tecnol Des. 2014; 5:265-71.

[20]    Novo Diaz C, Rodriguez Gamez O., Padron Montoya A., Zulueta Vigil A. I., Matamoros Zamora L. Kinematic description of gait in patients with spinocerebellar ataxia type 2. MEDISAN. 2016; 20(12):6017.

[21]    Broche Vazquez L, Torres Quezada M, Novo Diaz C, Perez Bonne P, Sagaro Zamora R. Influence of gait asymmetry on biomechanical behaviour of hip joints in patients with transfemoral prostheses. Ingeniare. Rev Chil Ing. 2015; 23(2): 312-22.  

[22]    Miyares O., A, Vazquez B., L, Novo Diaz C, Castillo G., L, Zamora S., R. Analysis of the functionality of transfemoral orthopedic prostheses. Rev Cubana Orttop Traumatol. 2011; 25(2):102-16.

[23]    Miyares Olivares l., A, Zamora Sagaro R, Martinez Rodriguez C, Mojena R., A, Novo Diaz C. Comprehensive evaluation of transfemoral orthopedic prostheses. Engineering 2012; 47:17-24.   

[24]    Palasi Ferrer A, Diez Aguilera A., T, Novo Diaz C. Methodology used for the cinematic study of the execution of the chest stroke with land ligatures for the development of the special strength of swimmers. EFDeportes 2013 [cited 20 Apr 2017]; 176(17). Available at: cinematico-de-la-brazada-de-pecho.htm

[25]    Novo Diaz C, Rios Lopez N, Pedron Montoya A, Fals Carvajal H. Preliminary evaluation of gait in healthy individuals. University of Science and Technology 2007; 11(44):135-140.

[26]    Rodriguez Delisle D, Rio Lopez N, Naranjo Garcia J C, Novo Diaz C. Digital module for portable electromyograph Universidad Ciencia y Tecnologia 2007; 11(45):173-176.

[27]    Novo Diaz C. Application of the videographic method for biomechanical classification in spinocerebellar ataxias and cerebrovascular accidents [doctoral thesis]. Santiago de Cuba: Universidad de Oriente; 2013. 137 p.

[28]    [Online] Engineering and Rehabilitation are in a Research scenario [Access June 15, 2017]. Available at:

[29]    [Online] Module 6 Biomechanical Analysis in Patients with Neurological Dysfunction [Access June 10, 2017]. Available at:"ct=clnk

[30]    [Online] International Congress of Electronics, Control and Telecommunications - CIECT XII [Access 15 April 2017]. Available at:

[31]    Quintero, J., et al Design and development of transcutaneous electrical stimulation equipment for neuromuscular rehabilitation in individuals with facial palsy, TECCIENCIA, Vol. 9 No. 16., 43-49, 2014, DOI:



  • There are currently no refbacks.

Copyright (c) 2019 TECCIENCIA