Spine Publish Ahead of Print

DOI: 10.1097/BRS.0b013e3181f7cfaa

New Method of Scoliosis Assessment:

Rozilene Maria Cota Aroeira, MSc

Jefferson Soares Leal, MD
Antnio Eustquio de Melo Pertence, PhD

TE
D

Preliminary Results using Computerized Photogrammetry

EP

From the Federal University of the State of Minas Gerais, Belo Horizonte, Brazil.

The manuscript submitted does not contain information about medical device(s)/ drug(s). No

funds were received in support of this work. No benefits in any form have been or will be
received from a commercial party related directly or indirectly to the subject of this manuscript.

This research was approved by Research Ethics Committee of Federal University of the State of

Minas Gerais on the 14th of February, 2008 (ETIC 579/07).

Address correspondence and reprint requests to Jefferson Soares Leal, MD, rua Padre Rolim,
815/ 107, Vertebra Clinic, Santa Efignia, Belo Horizonte, Minas Gerais, Brazil. Zip Code:
30130-090.
E-mail: jeffersonleal@terra.com.br

Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Study Design. A new method for non-radiographic evaluation of scoliosis was independently
compared with the Cobb radiographic method, for the quantification of scoliotic curvature.

Objectives. To develop a protocol for computerized photogrammetry, as a non-radiographic

method, for the quantification of scoliosis, and to mathematically relate this proposed method

TE
D

with the Cobb radiographic method.

Summary of Background Data. Repeated exposure to radiation of children can be harmful to

their health. Nevertheless, no non-radiographic method until now proposed has gained

popularity as a routine method for evaluation, mainly due to a low correspondence to the Cobb

EP

radiographic method.

Methods. Patients undergoing standing posteroanterior full length spine radiographs, who were
willing to participate in this study, were submitted to dorsal digital photography in the orthostatic

position with special surface markers over the spinous process, specifically the vertebrae C7 to
L5. The radiographic and photographic images were sent separately for independent analysis to

two examiners, trained in quantification of scoliosis for the types of images received. The
scoliosis curvature angles obtained through computerized photogrammetry (the new method)

were compared to those obtained through the Cobb radiographic method.

Results. Sixteen individuals were evaluated (14 female and 2 male). All presented idiopathic
scoliosis, and were between 21.4 6.1 years of age; 52.9 5.8Kg in weight; 1.63 0.05m in

height, with a body mass index (BMI) of 19.8 0.2. There was no statistically significant
difference between the scoliosis angle measurements obtained in the comparative analysis of
both methods, and a mathematical relationship was formulated between both methods.
Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Conclusion.
methods.

The preliminary results presented demonstrate equivalence between the two

More studies are needed to firmly assess the potential of this new method as a co-

TE
D

adjuvant tool in the routine following of scoliosis treatment.

MINI ABSTRACT

A new method of non-radiographic evaluation of scoliotic curve using computerized

photogrammetry is described. The new method was easy to apply, with low cost of
implementation and results similar to those of the Cobb method.
KEY POINTS

Computerized photogrammetry is presented as a new method in the non-radiographic

EP

evaluation of scoliotic curves.

The proposed method is described in detail.

In the presented sample, the values of the curves determined through computerized

photogrammetry method were similar to those determined through the Cobb radiographic

method.

Introduction

Scoliosis has been defined as a lateral curvature of the spinal column superior to 10

Cobb, generally associated to vertebral rotation1. Monitoring of this angle has been one of the
principal parameters used in defining the type of treatment to be instituted in young patients who
are still growing 2. The most trustworthy method to accompany the evolution of the curvature

Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

has been the standing posteroanterior full length spine radiograph, with curvature measurement
using the Cobb method 3. However, over the course of follow-up, this can result in taking more
than 25 radiographs 4. Nearly 15% of patients in one study had undergone 50 or more
radiographic examinations, and approximately 17% had received an estimated cumulative
radiation dose of 20 cGy or greater 4. This high number of radiographs can expose patients to

TE
D

relatively high doses of ionizing radiation. Various studies have shown that repeated exposure to
radiation in children can be harmful to their health 4-10.

Many non-radiographic methods of scoliosis accompaniment have been proposed as an

alternative to radiographic evaluation 11-15. Nevertheless, the majority have not demonstrated a
good correlation with the Cobb method 16.

Photogrammetry can be regarded as the science and technology of obtaining spatial

EP

measurements, and other geometrically reliable information, derived from photographs 17. The
computerized photogrammetry method proposed in this study can be considered one of many
uses of photogrammetry. It has been used in different fields, such as: cartography, architecture,

engineering, quality control, and 3-D modeling. However, its use in the evaluation of scoliosis
has not been well studied. To the authors knowledge, no data is presently available in the

literature which proposes a protocol for photogrammetric measurement of scoliosis which is

applicable to clinical practice.

The goals of this investigation were to develop a protocol for computerized

photogrammetry method for the quantification of scoliosis curvature as well as to compare the
results with those from the Cobb radiographic method in the group of volunteers. Our hypothesis
is that both methods yield similar results in obtaining the scoliosis curvature angle.

Materials and Methods

Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Field Studies

After approval from the Research Ethics Committee of our hospital, all patients who were
undergoing conservative treatment for idiopathic scoliosis in the spine unit between March and
October of 2008, and who had received the medical requisition for radiological monitoring of

TE
D

their deformity, were invited to participate in this study. The only exclusion criteria adopted for
this phase was the refusal of the patient or of his parents. All accepted the invitation and gave
their formal informed consent.

Radiological Exam

EP

Full-length standing posteroanterior spine radiographs were obtained using an X-ray

generator TOSHIBA KXO 15R (Tokyo, Japan) and a digital radiography Agfa Health System
(Mortsel, Belgium). The images were acquired in a single exposure for each incidence, and then

digitally obtained images were printed on a single sheet of 35.6 x 43.2 cm film. Only the
posteroanterior incidence for each patient was chosen for radiographic analysis. The scoliotic

curvature and the apical vertebra were determined for each patient by one examiner. Five
separate measurements were made at different times for each exam, and the average of these five

values was utilized for statistical analysis.

Computerized Photogrammetry Exam

Immediately after the radiological exam, a single trained, experienced examiner initiated
palpation and marking on the spinous process with surface markers, for all vertebrae localized
between C7 and L5. An Anatomical Landmarks Type Vector (ALTV), 45mm in length with a
Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

metallic base of 8mm, was developed and standardized as a surface marker specifically for this
study. Following this, each patient was photographed in standing, dorsal, lateral (right side
visualization 90) and oblique (left side 45) positions, utilizing a digital camera, Sony 7.1
megapixel (Manaus, Amazonas, Brazil), at definition 3072 x 2304, mounted on a Greika
WT3750 (So Paulo, SP, Brazil) tripod, set to a height of 1.10m, with a focal distance of 1.30m.

TE
D

A Carci Simetograph (Americanpolis, SP, Brazil), 2.02m in height and 0.72m in width, was
used to provide a base metric reference. The images were imported and analyzed using

CorelDraw13 software (Corel Corporation, Canada), and the angle of deviation of scoliotic
curvature was calculated. The protocol for the deviation calculation consisted, initially, of the
identification of the apical vertebra and the superior limit vertebra of the upper curve. This
procedure was done by tracing two vertical lines (using the free hand tool in the software

EP

toolbar), one line tangential to the convex face of the curve, and the other crossing through the
vertical axis of the C7 vertebra (Figure 1). The apical vertebra was defined as that furthest from
the vertical axis of C7 and touching the tangential line.

The second step of the calculation consisted of measurement of the angle of deviation of
each segment between the superior limit vertebra and the apical vertebra marked with the ALTV,

using the dimension tool located in CorelDraw13 (Corel Corporation, Canada). The sum of
the angles of deviation of the previous segments determined the final angle calculated through

computerized photogrammetry, which is called angle MR (Figure 2).

Definition of Angle MR

Angle MR was defined as the angular quantification calculated for a specific scoliotic
curve obtained through computerized photogrammetry in the frontal plane. It was calculated
through the sum of the lateral deviation of each segment between the superior limit vertebra and
Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

the apical vertebra. Its radiological correspondent is the Cobb angle as obtained through the
radiographic method.

Calculation of Angle MR

TE
D

The deviation of each segment was denominated as angle Rn (R1, R2, R3, etc). Angle Rn
was that formed between the lines traced between the centers of the MASV in two vertebrae of
the segment (adjacent vertebrae) with the line of the vertical axis Y (Figure 2). The sum of these
angle Rn measurements for each segment between the superior limit vertebra and the apical
vertebra determined the Angle MR.

For each volunteer in this study, five measurements were taken of each angle Rn, at

EP

distinct intervals, by the same examiner. The average of the sum of these angle measurements
was considered to be the angular value of the deformity.

Analysis of Data

The radiographic and photogrammetric measurements were performed separately by an

orthopedic spine surgeon, and a physiotherapist experienced in photogrammetry of the spine,

respectively.

To determine the profile of the study participants, a frequency distribution was performed

for the variables gender and type of curvature, along with calculation of the central tendency
for the variables of age and body mass index (BMI).
In order to verify the differences in measurements of thoracic and lumbar scoliosis
curvatures between the Cobb and computerized photogrammetry methods, the Wilcoxon test was
realized.
Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

The localization of the apical vertebra was compared between the two methods through
concordance calculation using Kappa statistics. The non-parametric test was used, considering
the size and the characteristics of the sample. Concordance in the localization of the apical
vertebra in this study was defined as the identification of the same vertebra, or at most, one
adjacent vertebra. This definition was established considering that the computerized

TE
D

photogrammetry method actually detects the tip of the spinous process rather than the vertebral
body of the apical vertebra, and also considering that the real apex curve detected through
radiographic method could be localized in the intervertebral space. The significance level
adopted for all tests was 5%.

EP

Results

Sixteen volunteers were evaluated using the two methods. All presented idiopathic
scoliosis. The average age was 21.4 years. For weight, height and body mass index, the averages

were 52.9kg, 1.63m and 19.8kg/m2, respectively. There was predominance of females (87.5%),
and with a female to male ratio of 7:1. In relation to curvature type, a double thoracic/lumbar

curve was the most frequent, with 12 cases observed (75%), followed by a single lumbar curve
(18.7%), and a single thoracic curve (6.2%). The general characteristics and the measurements

using Cobb and computerized photogrammetry methods of each individual included in the study
are presented in table 1.
The values of the thoracic curve measurements obtained using the Cobb method varied

between 17.8 and 73.0, with an average of 36.1, while the values of the curve measurements
obtained using the computerized photogrammetry method varied between 19.6 and 74.2, with
and average of 36.4, with no significant difference noted between the two groups (Table 2 and
Figure 3).
Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

The average difference between the angle values obtained through the Cobb method and
the computerized photogrammetry method was 2.9 for thoracic curves.
In the lumbar region, the average of the curvature evaluated through the Cobb method
was 27.2, with a variation of 14 to 48.6, while the average through the computerized
photogrammetry method was 25.7, with a variation of 10 to 55.0 (Table 2 and Figure 3). The

TE
D

average difference between the two methods in the lumbar region was 5.1.
There was no significant difference between the average curve measurement values
obtained through the Cobb and computerized photogrammetry methods for all curves analyzed
together. The average of the differences between the methods for all curves was 4.1.

The average Kappa index for evaluation of intensity of concordance in the localization of
the apical vertebra between the methods for the thoracic region was 0.92. For the lumbar region

EP

the Kappa index was 0.82.

Discussion

Various non-radiographic methods for the quantification of scoliosis have been proposed

in the last thirty years 11-15. Unfortunately, they have not survived the test of time, due to a
variety of possible reasons, notably: 1) non-radiographic techniques cannot entirely replace

radiographs; 2) there are important direct costs in the implementation of new technology as well
as recurrent costs with training, salary of the technician and others; 3) theses techniques take
more time than the radiological exam; and 4) finally, the majority have demonstrated low
correlation to the Cobb method 16.
The radiological exam with frontal plane measurement of the Cobb angle has been

considered to be the gold-standard for the quantification and monitoring of scoliosis1, 2. In

addition to the determination of the Cobb angle, it provides information as to skeletal maturity,
Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

the flexibility of the curve, the presence of congenital bone malformations, and the general
balance of the spine, which is all indispensable to good clinical practice in scoliotic patients.
However, once these data have been established, further monitoring of the Cobb angle may be
considered the main reason for the repetition of radiological examination 3. Along with the
important benefits of X-rays, there are risks that need to be considered as well. Brenner et al 18,

TE
D

in a study conducted by a large group of epidemiologists and radiation scientists, concluded that
there is good epidemiological evidence for an increased cancer risk for some doses of radiograph
exposition. Doody et al 4, analyzing breast cancer mortality in 5573 women, reported a 69%
excess in breast cancer mortality among women with scoliosis. And consistent with radiation as
a causative factor, they showed that the risk of dying of breast cancer increased significantly with
number of radiograph exposures and with cumulative radiation dose. Sadetzki and Mandelzweig
, in a recent paper, stated that children constitute a subgroup at greater risk from exposure to

EP

19

develop radiation-induced cancer. Uncomfortably, patients with scoliosis continue to be exposed

to ionizing radiation, practically the same as they were thirty years ago. Although the risks

involved may be relatively small, the ancient concept expressed by Hippocrates of Primum non
nocere (First, do no harm) should serve as a guide when assessing the appropriateness of

radiation-based imaging procedures in pediatric patients.

Mathematical relationship between the Cobb and the photogrammetry methods

The scoliotic curve obtained through the Cobb method (MC), is measured by the arc,

defined by the superior plateau of the uppermost inclined vertebra, and by the inferior plateau of
the lowermost inclined vertebra (Figure 4). The angular measurement of the scoliosis curve
obtained through computerized photogrammetry method (MR) was calculated by the sum of the
deviation angles of axis Y, denominated R1, R2, R3 and R4 (the sum of the deviation angles of
Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

each segment between the superior limit vertebra and the apical vertebra). Thus, the deviation
angle is considered to be that encountered between the secant lines drawn from two consecutive
spinous processes, and the Y axis. Additionally, considering the interval used in the Cobb
method and the interval used in computerized photogrammetry method, a similarity between MC
and MR angles can be shown (Figure 4).

TE
D

In our study, it was possible to establish a mathematical correlation between curve

measurements obtained through the methods taking into account some conditions. Defining the
scoliotic curve as being constituted of perfect segments of a circular arc, as indicated in figure 4,
then the MC measurement is equal to the sum of the angles C1, C2, C3 and C4 obtained between
the consecutive spinous processes found at the same interval of this measurement. Thus, if
angles C1, C2, C3 and C4 are all equal, and equal to C, we will find that the sum of these angles

EP

obtained at the same interval through the Cobb method will be equal to 4C.

Naturally, the scoliotic curve is not a perfect segment of an arc, so a perfect extrapolation
between the curve measurements through the Cobb method and the computerized

photogrammetry method should not easily be made. However, the failure of complete adherence
to mathematical premises for the establishment of a relationship between the two methods did

not result in significant differences in measurements, and appeared to be valid for practical use

within acceptable margins.

Equations

Further considering that the scoliotic curvature is constituted of segments of circular arcs,
the triangles BD, DE, GFH, and HFJ (Figure 5) will be isosceles triangles.
Therefore, angle X will be indicated by equation (1):
X= (180- C) / 2

(1)

Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

In addition, there is a relationship between the angles of deviation of axis Y (R1, R2, R3
e R4) obtained through computerized photogrammetry, with the angles (C) of each vertebral
segment as obtained using the Cobb method, as indicated in equations (2), (3), (4), (5):
(2)

R2= R1 + C = C/2 + C = 3C/2

(3)

R3= R4 + C = C/2 + C = 3C/2

(4)

R4= 90 X = 90 - (180- C)/2 = C/2

TE
D

R1=90 - X = 90 - (180- C)/2 = C/2

(5)

Thus, the MR measurement will be indicated by equation (6):

MR= R1 + R2 +R3 + R4 = C/2 + 3C/2 + 3C/2 + C/2 = 4C

(6)

As a result, mathematically, the value of the measurement obtained through computerized

photogrammetry will be equal to that obtained through the Cobb method.

EP

The mathematical proof proposal considered the scoliotic curve as a polynomial arc of
degree 2. It is important to bear in mind that the prior demonstration does not represent a general
solution. However, it makes it easier to understand, as observed in clinical practice, that there is

equivalence between the measurements obtained in this study and those obtained through the

Cobb method.

Results

The curvature values obtained using the computerized photogrammetry method were

similar to the values obtained using the Cobb radiographic method. The proposed method
showed less difference in the thoracic region than in the lumbar region. The total average of
difference of 4.1 between the methods, which includes all thoracic and lumbar curves, was
compatible with the intrinsic error of the Cobb method in inter- and intraobserver analysis 20-22.
Measurements of the Cobb angle bear both an inter- and intraobserver variability of
Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

approximately 4 to 8 21, 23-29. The average for the individual measurement differences between
the methods in the thoracic region was 2.9, while in the lumbar region the average difference
was 5.1, influenced by the differences observed in two volunteers. Unfortunately, the limited
number of cases in this initial study did not allow for inferences to be made about the possible
causes involved.

TE
D

The identification of the apical vertebra was an important step of the computerized
photogrammetry method, and the use of ALTV marks helped in this task. The vertebra with the
largest variation in the analysis of the three-dimensional coordinates from the base to the tip of
the ALTV coincided, generally, with the identified apical vertebra. The concordance in the
localization of the apical vertebra between the computerized photogrammetry method and the
Cobb method , as estimated through Kappa index, was excellent, according to criteria based on

EP

the interpretation of Landis and Koch (Kappa > 0.80 = excellent concordance) 30. Besides, with
the use of these marks, virtual images of the spatial behavior of the spinal column could be
generated using SolidWorks software (Santa Monica, CA, USA)(Figure 6).

Dhnert and Tomasi attempted, without success, to detect mild scoliosis using
photogrammetric principles31. The protocol used by them was different from the protocol applied

in this study. They chose an analysis protocol using surface marks only in C7, T9 and L5. Saad
et al

32

also did not find a linear relationship between the photogrammetric and radiographic

methods using a similar protocol to that used by Dhnert and Tomasi. We believe that the
protocol used in both studies does not allow adequate angular quantification of scoliotic
curvature, because a mathematical relation of similarity is not possible with the Cobb method
using their methodology and parameters.
Although this study did not have as an objective the evaluation of costs, a few
considerations on the topic are pertinent. The total cost of the photogrammetric equipment
(computer, software license, simetograph, digital camera, tripod, surface markers) was US$
Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

1,826. Strictly for use as a comparison, the cost of installed radiographic equipment used in this
study was US$ 216,238. Basic skills needed for the use of this method were: primary knowledge
on spinal anatomy, photography and imaging software. The time spent with the radiographic
method on a typical patient was 13 minutes (positioning, photographic exposure, and one curve
measurement), while the time spent using the manual photogrammetric measurement was 28

TE
D

minutes (surface marking, photographic exposure and one curve measurement). We considered
this longer time period the principal drawback of the photogrammetric method. On the other
hand, we believe that the automation of the photogrammetric curve measurement with the
development of software specific to this task is feasible and, hypothetically, may reduce the time
necessary to obtain the curvature angle.

This study presents limitations. In this consecutive series, the majority of volunteers

EP

presented a BMI below or within normal variation (18.5 25 kg/m2), which may have facilitated
a more precise localization of the spinous process and, presumably, have positively influenced
the performance of this method, having the Cobb method as a reference. Beyond body weight,

factors such as lumbar curves, previous spine surgery, and the ability of the examiner, may
constitute obstacles which could restrict the application of the method.

Radiography currently is and probably will remain the principal examination method for

children with spinal deformities. The method presented in this study does not seek or intend to

replace the radiological exam. It introduces a new approach in the non-radiographic evaluation
of scoliosis, with potential application in monitoring the curvature. Further studies are needed to
firmly assess the potential of this method as a co-adjuvant tool in the routine following of

scoliosis treatment. Important points arising from this preliminary evaluation, such as refinement
of the photogrammetric technique, costs, and validation of a new diagnostic test are now under
current investigation in our institution.

Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

1. Reamy BV, Slakey JB. Adolescent idiopathic scoliosis: review and current concepts. Am
Fam Physician 2001;164:111-6.
2. Wills BPD, Auerbach JD, Zhu X, et al. Comparison of Cobb angle measurement of
scoliosis radiographs with preselected end vertebrae. Spine 2007;32:98-105.
3. Klos SS, Liu XC, Lyon RM, et al. Reliability of a functional classification system in the

TE
D

monitoring of patients with idiopathic scoliosis. Spine 2007;32:1662-6.

4. Doody M, Lonstein JE, Stovall M, et al. Breast cancer mortality after diagnostic

radiography: finding from the U.S. scoliosis cohort study. Spine 2000;25:2052-63.
5. Levy AR, Goldberg MS, Hanley JA, et al. Projecting the lifetime risk of cancer from
exposure to diagnostic ionizing radiation for adolescent idiopathic scoliosis. Health Phys
1994;66:621-33.

EP

6. Levy AR, Goldberg MS, Hanley JA, et al. Reducing the lifetime risk of cancer from
spinal radiographs among people with adolescent idiopathic scoliosis. Spine
1996;21:1540-7.

7. Goldberg MS, Mayo NE, Levy AR, et al. Adverse reproductive outcomes among women
exposed to low levels of ionizing radiation from diagnostic radiography for adolescent

idiopathic scoliosis. Epidemiology 1998;9:271-8.

8. Bone CM, Hsieh GH. The risk of carcinogenesis from radiographs to pediatric orthopedic

patients. J Pediatr Orthop 2000;20:251-4.

9. Ron E. Cancer risks from medical radiation. Health Phys 2003;85:47-59.

10. Berrington de Gonzales A, Darby S. Risk of cancer from diagnostic X-rays: estimates for
the UK and 14 other countries. Lancet 2004;363:345-51.
11. Ovadia D, Bar-On E, Fragnire B, et al. Radiation-free quantitative assessment of
scoliosis: a multi center prospective study. Eur Spine J 2007;16:97-105.

Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

12. Zsidai A, Kocsis L. Ultrasound-based spinal column examination systems. Facta Univ
Phys Educ Sport 2001;8:1-12.
13. Years AM, Pea RG, Junco R. Moir topography: alternative technique in health care.
Opt Lasers Eng 2003;40:105-16.
14. Goldberg CJ, Kalizer M, Moore DP, et al. Surface topography, Cobb angles and cosmetic

TE
D

change in scoliosis. Spine 2001;26:E55-E63.

15. Willner S. Moire topography for the diagnosis and documentation of scoliosis. Acta
Orthop Scand 1979; 50:295-302.17

16. Sahlstrand T. The clinical value of moir topography in the management of scoliosis.
Spine 1986;11:409-17.

17. Lillesand TM, Kiefer RW, Wiley J. Remote sensing and image interpretation. New York:

EP

Wiley & sons, 2000.

18. Brenner DJ, Doll R, Goodhead DT, et al. Cancer risks attributable to low doses of
ionizing radiation: assessing what we really know. Proc Natl Acad Sci USA

2003;100:137616.

19. Sadetzki S, Mandelzweig L. Childhood exposure to external ionising radiation and solid

cancer risk. Br J Cancer 2009;100:1021-5.

20. Pruijs JEH, Hageman MAPE, Keessen W, et al. Variation in Cobb angle measurements

in scoliosis. Skeletal Radiol 1994;23:517-20.

21. Carman DL, Browne RH, Birch JG. Measurement of scoliosis and kyphosis: radiographs
intraobserver and interobserver variation. J Bone Joint Surg Am 1990;72:32-3.

22. Mok JM, Berven SH, Diab M, et al. Comparison of observer variation in conventional
and three digital radiographic methods used in the evaluation of patients with adolescent
idiopathic scoliosis. Spine 2008;33:681-6.

Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

23. Morrissy RT, Goldsmith GS, Hall EC, et al. Measurement of the Cobb angle on
radiographs of patients who have scoliosis: evaluation of intrinsic error. J Bone Joint
Surg Am 1990;72:320-7.
24. Cheung J, Wever DJ, Veldhuizen AG, et al. The reliability of quantitative analysis on
digital images of the scoliotic spine. Eur Spine J 2002;11:535-42.

TE
D

25. Zmurko MG, Mooney JF 3rd, Podeszwa, DA, et al. Inter- and intraobserver variance of
Cobb angle measurements with digital radiographs. J Surg Orthop Adv 2003;12:208-13.
26. Gstoettner M, Sekyra K, Walochnik N, et al. Inter-and intraobserver reliability

assessment of the Cobb angle: manual versus digital measurement tools. Eur Spine J
2007;16:1587-92.

27. Geijer H, Beckman K, Jonsson B, et al. Digital radiography of scoliosis with a scanning

EP

method: initial evaluation. Radiology 2001;218:40210.

28. Geijer H, Verdonck B, Beckman KW, et al. Digital radiography of scoliosis with a
scanning method: radiation dose optimization. Eur Radiol 2003;13:54351.

29. Gross C, Gross M, Kuschner S. Error analysis of scoliosis curvature measurement. Bull
Hosp Jt Dis Orthop Inst 1983;43:1717.

30. Landis JR, Koch GG. The measurement of observer agreement for categorical data.
Biometrics 1977;33:159-74.

31. Dhnert MB, Tomasi E. Validade da fotogrametria computadorizada na deteco de

escoliose idioptica do adolescente. Rev bras fisioter 2008;12:290-7.

32. Saad KR, MA, Colombo AS, Joo SMA. Reliability and validity of the photogrammetry
for scoliosis evaluation: a cross-sectional prospective study. J Manipulative Physiol Ther
2009;32:423-30.

Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

TE
D

Figure 1. The left line (white) identifies the apical vertebra through the tangent of the vertical

EP

line and convexity of the curve (T10). The right line (grey) identifies the first vertebra to

interrupt the vertical alignment (T4) as the superior limit vertebra.

Figure 2. Measurement of angle Rn for each segment localized between the superior limit
vertebra (T4) and the apical vertebra (T10): R1 = 10, R2 = 4, R3 = 8, R4 = 12, R5 = 10, R6 = 9
and R7 = 4. Angle MR = sum Rn = 57 degrees.

Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

TE
D
EP

Figure 3. Comparison between the curve measurements obtained through the Cobb method and

the computerized photogrammetry method for thoracic and lumbar curves.

Figure 4. Angle of scoliosis curvature obtained through the Cobb method (MC=C1+C2+C3+C4),
and obtained through the computerized photogrammetry method (MR=R1+R2+R3+R4).
Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

TE
D
EP

Figure 5. Demonstration of the isosceles triangles formed by the circular arcs of each vertebral

segment (BD, DE, GFH and HFJ) and their relationship with angles R1, R2, R3 and R4.

Figure 6. The figures represent the spatial behavior of a scoliotic curve of a volunteer, using
anatomical landmark type vector (ALTV) processed in three distinct planes.
Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Table1. General characteristics of each individual included in the study and their measurements using

Lumbar

TAV

LAV

Thoracic

Lumbar

20.8 T and L

59.8

26.0

T10

L4

57.4

17.6

COBB

15 M 17.1 T and L

32.0

15.0

22

18.3 Thoracic

27.0

13

25.5 T and L

30.4

32.0

12

21.7 T and L

19

22.9 T and L

16

22.2 T and L

14

17.4 T and L

11

10 25

LAV
L3

T10

L2-L3

34.4

17.2

T9

L3

T9-T10

25.8

T9

T7-T8

L2

22.6

24.0

T8

L2

33.8

T8

L2

40.8

10.0

T7

L2

28.2

38.4

T8

L1

22.4

35.4

T8

L2

42.2

22.8

T9

47.0

22.8

T7

L2

22.4

27.0

T9

L2-L3

23.0

30.0

T9

L3

39.6

T and L

50.4

48.6

T7

L2

51.0

55.0

T7

L2

19.8 T and L

21.4

18.8

T8

L2

26.4

16.8

T9

L3

19

TAV

T9-T10

EP

COMPUTERIZED
PHOTOGRAMMETRY

TE
D

Gender

39

Scoliosis level

Age (years)

Thoracic

BMI

Subjects

Cobb method and computerized photogrammetry method for thoracic and lumbar regions.

23.4 T and L

25.6

26.4

T7

L2

29.0

27.8

T7

L3

12 46

18.9 T and L

73.0

41.8

T6-T7

L2

74.2

39.4

T7

L2

L3

11 45

13 14

17.8

19.8

L2-L3

22.0

14 13

15.9

14.0

L1

26.0

T12

15 15 M 16.9
16 25

23.1 T and L

18.4

L1

17.8

L1

17.8

25.5

T5

L2

19.6

23.8

T5

L1

F= Female; M=Male; BMI= Body Mass Index; T= Thoracic; L= Lumbar. TAV= Thoracic Apical
Vertebra;
Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

LAV= Lumbar Apical Vertebra.

Table 2. Comparison between the curve measurements obtained through the Cobb
method and the computerized photogrammetry method for thoracic and lumbar
curves.

TE
D

standard

N minimum maximum average deviation p-value

Thoracic Cobb

13

17.80

73.00

36.14

16.38

Thoracic Photogrammetry 13

19.60

74.20

36.43

16.67

Lumbar Photogrammetry 15

10.00

55.00

25.71

11.04

Lumbar Cobb

14.00

48.60

27.20

10.05

0.615

EP

15

0.529

Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.