Moiré topography as a screening and diagnostic tool—A systematic review

Diagnostic investigation can be carried out using non-radiological and non-contact methods. Moiré topography (MT) seems to be a viable alternative to radiographic research in evaluating the spine and/or trunk deviations. The aim of this systematic review was to analyze the current knowledge regarding the reliability and validity of Moiré topography as a screening and diagnostic tool. The systematic review was performed from 2010 until March 2021 in the PubMed, EBSCO, Web of Science, and Scopus databases, according to the eligibility criteria. This review fulfilled the following criteria according to the PICO system: population (children and adolescents), intervention (MT measurement), comparison (repeated MT measurements, MT compared to Cobb angle or scoliometer), outcome (reliability and validity of MT). Eight studies fulfilled the inclusion criteria for further analysis. All the studies were assessed to be of high quality. Included studies found that MT had high repeatability and high intraobserver and interobserver correlation, and correlation between MT parameters and radiographic Cobb angle ranged from moderate to high. The authors reported difficulty in defining the cut-off values for MT parameter (Surface Trunk Rotation—STR), and unsatisfactory sensitivity and specificity of MT examination. The studies did not reveal the advantage of MT as a screening method in the detection of idiopathic scoliosis in comparison to radiograph. Based on the evidence from eight studies, the results indicated moderate evidence for reliability and validity of Moiré topography as a screening and diagnostic tool. There is still no strong evidence for the accuracy of MT.


Introduction
Postural deviations in children and adolescents are an important medical and social issue. Research indicates a disturbing phenomenon of the frequent appearance and progression of irregularities in the body posture of the youngest part of society [1][2][3][4]. Initially, they develop asymptomatically and their consequences can be felt over the next years of life. Examining children in terms of assessing their posture seems justified because serious irregularities can significantly change the quality of life, cause significant deformations in the osteoarticular system, pain, and disorders of the internal organs [5,6]. Consequently, screening is the most important factor preventing deformity from progressing. a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 criteria. All articles were screened to identify relevant studies, first by title, secondly by abstract, and thirdly by full-text screening. The article was considered potentially significant and the full-text was reviewed if, after discussion by two independent reviewers, it could not be clearly excluded on the basis of the title and abstract [20]. Misunderstandings were resolved through discussions between researchers. Only the included studies were submitted for data extraction and methodological quality evaluation. The extracted data included author, year of publication, study population, participant characteristics. Duplicates were deleted with Endnote online.
Our review fulfilled the following criteria according to the PICO system (Population: children and adolescents; Intervention: MT measurement, Comparison: 1) repeated MT measurements, 2) MT compared to Cobb angle or scoliometer; Outcome: reliability and validity of MT. The reliability and validity of MT were assessed by calculating Pearson's correlation coefficient (r), or False-Positive Rate (FPR), or Positive Predictive Value (PPV) and/or Negative Predictive Value (NPV), or the sensitivity and the specificity, or intraobserver and interobserver error, or Interclass Correlation Index. The analysis of the results was limited to a qualitative summary.
Articles were removed if the study did not meet the previously specified selection criteria. The number of articles included and excluded at different phases was presented in a PRISMA flowchart (Fig 1). PRISMA guidelines were followed for this systematic review [21].

Quality assessment
The studies were assessed for methodological quality using a 13-item critical appraisal tool developed by Brink and Louw [22], for assessing validity and reliability of results from studies ( Table 2). Five items were used to assess the methodological quality of both validity and reliability studies, four items were applied for validity studies, and four items for reliability studies. Preceding the final screening, reviewers tested the methodological quality assessment of two similar articles that were not included in this review. Disagreements were resolved through discussion or consultation.  Article  1  2  3  4  5  6  7  8  9  10  11  12  13 High quality � 60% Fugiel and Krynicka, 2010 [24] y n y n/a n/a n/a n n/a y y y n y 67% (6/9) Ueno et al., 2011 [25] y n y n/a n/a n/a n n/a y y y n y 67% (6/9)  [29] y n y n/a n/a n/a n n/a y y y y y 78% (7/9) Each item was rated as yes or no or not applicable. Each study was assigned a total score, which was the sum of all positive ratings according to the methodological criteria. The reviewers considered studies to be of high quality if the methodological quality score was � 60% of the maximum score, as proposed by previous studies [23].

Search result
This literature search yielded a total of 514 articles. After removing duplicates, 477 studies remained. Based on the analysis of the titles and abstracts 27 studies were eligible for assessment by full paper. Following the full-text review of 27 articles, 8 studies [24][25][26][27][28][29][30][31] fulfilled the inclusion criteria for further analysis. The number of articles included and excluded at different phases was presented in a flowchart (Fig 1).

Study characteristics
The total sample size of the eight included studies consisted of 1 141 645 participants, with the age range of 7-21 years. The systematic review showed studies using MT among healthy children [25-27, 30, 31] and with scoliosis [24,[28][29][30]. In all research enrolled as well as girls and boys. The characteristics of the included studies are described in Table 3.

Methodological quality assessment
The average quality of the 8 studies was 67% (range 62%-78%). All disputes were solution during a consensus meeting. Table 2 shows the results of the methodological quality assessment. All studies were of high quality, scoring � 60% on the critical appraisal tool. The main items with low scores were an evaluator's characterization (75% of studies unreported), a period of time between the test collection (75%), and a description of cases of sample loss (50%).
Validity. Correlation between MT parameters and radiographic Cobb angle ranged from moderate to high ( Table 3). The values of Spearman correlation indicated high or moderate correlations (0.61-0.78) taking into account particular spinal segments [24]. High correlations were observed in thoracic and lumbar scoliosis, and moderate correlation in thoracolumbar scoliosis [24]. Two studies reported high or moderate and significant correlations (p � 0.01 between the Cobb angle with DHOPI and POTSI) [28,29].
The validity of the MT examination assessed by False Positive Rate (FPR) reported the screening test's false-positive rate ranged from 32.4% [25] to 66.7% [27] participants. The validity of the MT examination assessed by Positive Predictive Value (PPV) or Negative Predictive Value (NPV) reported PPV ranged from 33.3% [27] to 86.17 [30]. One study examined by MT the predictive value of school scoliosis screening (SSS) in order to detection a curve of over 20˚ [31]. The value was respectively 2.1% for fifth-grade students and 7.6% for eighthgrade students.
In two studies one of the objectives was to evaluate the accuracy of MT as a screening tool by determining sensitivity and specificity. One study compared the MT method to scoliometer [26] and reported that STR value provided an unsatisfactory sensitivity and specificity (STR � 5: the sensitivity = 64.5%, the specificity = 88%; STR � 4: the sensitivity = 77.4%, the specificity = 71.1%) [26]. Another study reported a perfect sensitivity (92.04%) and allowed specificity (74%) [30].

Discussion
The aim of this systematic review was to analyze the current knowledge regarding the reliability and validity of Moiré topography as a screening and diagnostic tool. It is known and used mainly in Japan (the most advanced country in using this method) [25,27,31], Poland [24,26] and Spain [28][29][30]. This systematic review shows that MT has many advantages. Studies showed that MT had high repeatability and high intraobserver and interobserver correlation, and correlation between MT parameters and radiographic Cobb angle ranged from moderate to high.
On the other hand, studies showed the following disadvantages of the MT method in scoliosis screening (Table 4). One study reported hardship in determining the cut-off values for the STR parameter and low sensitivity and specificity of MT method. Moreover, the studies did not reveal the advantage of MT as a screening method in the detection of idiopathic scoliosis in comparison to radiograph [30] or clinical examination with the use of the scoliometer [26].
A very serious problem is the occurrence of a large number of false positives results. The authors suggested that the high rate of false-positive results was probably due to the fact that the control samples included patients referred for possible scoliosis, and most of them had some kind of asymmetry in their backs, which was detected by the MT method [30]. Other studies suggested that the problem of false-positive results could be minimalized by rescreening [25] and emphasized the need for further research on reducing the false-positive rate of MT in scoliosis screening [27].
Although MT cannot replace radiography in the diagnosis of scoliosis, the authors concluded that the MT method offers additional quantitative data that can complement the radiological examination [28][29][30]. Studies suggested that a combined examination (e.g., MT combined with Cobb angle) improved the accuracy of screening results. Thanks to the correlation of the combined DHOPI and POTSI variables with the Cobb angle, it is possible to monitor the progression of scoliosis, and to reduce the number of exposures to ionizing radiation, and is a harmless procedure that can be applied repeatedly.
In order to eliminate the risk of measurement error, the same methodological procedure should be developed. Many studies suggested that body posture should be assessed at the same time of day [24,33] by a qualified person (physiotherapist, doctor, radiologist, teacher) with several years of experience [24,[34][35][36]. The room in which the examination is performed should be darkened [35][36][37][38][39]. Only one article drew attention to the fact that the height of the measuring station should be adjusted to the child's height [26]. Several studies indicate that the selection of the best view of the child's body is used to analyze deviations in body posture [24,38,40,41]. In two studies, the authors reported that a dermograph was used to mark bone points on a child's body [42,43].
However, the assessment of the generated topograms on the basis of visual inspection and marking specific points on the patient's bone body, may affect the accuracy of the measurements [7,19,44]. Due to the fact that the topogram is made in strictly defined conditions (i.e., Table 4. Main conclusions from studies included in the systematic review.

Exposure Outcome Main conclusions
Fugiel and Krynicka, 2010 [24] MT and radiograph Validity MT method may be used only for the purpose of screening studies carried out in order to diagnose postural defects.

Validity
The existence of a large number of false positives results is a very serious problem that occurs in most of the described screening programs in schools, as a result of the referral of a relatively large number of children. Primary screening for MT is beneficial because it does not require doctors, and the time that children have to spend in the actual screening process is negligible.

Validity and reliability
Studies have not shown an advantage of surface topography as a screening method to detect idiopathic scoliosis compared to a clinical trial using scoliometer. The lack of STR value provided a satisfactory sensitivity and specificity at the same time.

Validity
The MT school screening test had a high false-positive rate. The study highlighted the need for further research into reducing the false-positive rate of MT in scoliosis screening.
Pino-Almero et al., 2016 [28] MT and radiograph Validity and reliability A significant correlation was found between the changes in DHOPI, POTSI, and the Cobb angle. By obtaining the correlation of variables connected DHOPI and POTSI and the Cobb angle, can be monitored the progression of scoliosis. This would help to reduce the number of exposures to ionizing radiation.
Pino-Almero et al., 2017a [29] MT and radiograph Validity Although the MT method cannot replace radiographs in the diagnosis of scoliosis, the correlations between radiographic and topographic parameters suggest that it offers additional quantitative data that can complement the radiological examination.

Validity and reliability
The MT method may be less sensitive in low-grade scoliosis with a slight rotational component but of little importance.
The MT method may be a useful test in the screening phase of idiopathic scoliosis with a higher sensitivity than the Adams test and similar specificity. constant distance, the exact position of the level, constants lighting conditions, the same parameters of the optical system) [8], this measurement should be repeatable. When analyzing the accuracy of the measurement, attention should be paid to the specificity of the object being tested, which is the human body. Practically, it is not possible to show physiological points on the patient's skin with an accuracy greater than 5 mm. Taking this into account, the parameter, e.g., the difference in the height of the blade angles, is affected by a random error of 1 cm. For this reason, it should be assumed that the value of 1 cm is the limit of the accuracy of the method due to physiological features. On the other hand, the accuracy of the apparatus is 1 mm (resolution and the value of the parameters calculated), so it is 10 times better than that necessary from the point of view of physiology. This is of practical importance, as it allows one to detect the first features of curvatures still invisible to the naked eye [19,44,45]. MT has been shown to be more effective than simply reviewing posture for verification symmetries of the cervical and lumbar spine [46]. Moreover, the studies showed the evolution of the MT method, reflecting the increasing efforts to improve the accuracy and precision of the method [19,44].
Despite the advantages of using this technique, there is a high probability of errors in evaluating a large number of topograms qualitatively. The evaluation of the topograms produced is based on visual inspection of the images. It can be a difficult task. Fatigue caused by conducting many subjective evaluations can interfere with the assessment of the images. Moreover, the difficulty may lie in the different ways of assessing posture, the method of recording the results, using measuring tools with a different structure, and the definition of the parameters measured. Standardization should therefore include the necessary measurement conditions, the unification of measurement tools, and typing the definition of the measured attitude traits to overcome terminological barriers.
A literature review showed that the consistency assessment regarding the asymmetry in the sagittal plane was slightly smaller than in the other planes. These less reliable measurements of the spine in the sagittal plane can be justified, as most indirect methods are concerned with the univariate assessment and spine lesions are usually three-dimensional [45,47]. Although the MT does perform a three-dimensional analysis, it may be difficult for the assessor to see asymmetries in the sagittal plane due to the smaller sharpness of the fringes, especially in the cervical and lumbar regions [7].
Scoliosis screening is the most important factor preventing deformity progression. MT has been applied in school screening programs [25,27,28]. Scoliosis is studied by looking at asymmetric patterns in the Moiré photo. However, it only provides qualitative results, that is, whether the patient has scoliosis or not, and the exact Cobb angle cannot be inferred from the fringe patterns. It was not possible to estimate the value of the Cobb angle despite the high accuracy (AUC = 0.929) of the application of the diagnostic system based on artificial intelligence to screening scoliosis using the Moiré image [44].
Studies suggested that MT was confirmed to be a method for the easy detection of low angular value scoliosis (Cobb angle < 10˚) [8] and more accurate results are obtained with small spine deformities than large spine deformities [44]. Most studies of the prevalence of scoliosis detected scoliosis with a higher angular value (higher than 10˚), which limits early intervention by healthcare professionals, because it determines the scoliotic process is the vertebral rotation [8]. The authors claimed that MT is the most sensitive to this factor. However, the Scoliosis Research Society (SRS) suggests confirming the diagnosis at a Cobb angle of 10˚ [32]. Therefore, it can be said that the authors' methodological assumption was incorrect.
This systematic review showed that there is moderate evidence for the reliability and validity of MT as a screening and diagnostic tool. Nowadays alternatives to MT are methods as raster stereography (the Formetric 4D) [14], 3D ultrasound imaging (the Scolioscan system) [16], and Infrared Thermography (IR thermography) [17]. Studies showed that these methods had a good reproducibility [14,15,17] and had good to excellent correlation comparison with radiography [14,15]. However, the authors suggest that further studies are required to demonstrate their clinical values with a larger number of scoliosis patients with different types of curvature [14,15].

Strengths and limitations
This is the first review focused only on one non-radiographic method of measuring body posture. This systematic review looked at the sensitivity of only high quality research However, there are some limitations to this review. Studies with significant heterogeneity were summarized. We found heterogeneity between studies with regard to aspects such as study population, exposure assessment methods, and data presentation that may limit the final conclusions. A quantitative meta-analysis could not be performed. Finally, the search strategy was limited only to full-text articles in English.

Conclusion
Based on the evidence from eight studies, there is moderate evidence for the reliability and validity of Moiré topography as a screening and diagnostic tool. MT is an alternative method of examining the deformity of the spine and trunk but there is still no strong evidence for the accuracy of MT. Moreover, the methodology of MT should be standardized in order to use it as an accurate screening tool. Therefore, researchers should be careful in drawing conclusions from studies using the MT measurement.