A review of methods for quantitative evaluation of spinal curvature

Artificial intelligence automated measurements of spinopelvic parameters in adult spinal deformity-a systematic review

PURPOSE

This review evaluates advances made in deep learning (DL) applications to automatic spinopelvic parameter estimation, comparing their accuracy to manual measurements performed by surgeons.

METHODS

The PubMed database was queried for studies on DL measurement of adult spinopelvic parameters between 2014 and 2024. Studies were excluded if they focused on pediatric patients, non-deformity-related conditions, non-human subjects, or if they lacked sufficient quantitative data comparing DL models to human measurements. Included studies were assessed based on model architecture, patient demographics, training, validation, testing methods, and sample sizes, as well as performance compared to manual methods.

RESULTS

Of 442 screened articles, 16 were included, with sample sizes ranging from 15 to 9,832 radiograph images and reporting interclass correlation coefficients (ICCs) of 0.56 to 1.00. Measurements of pelvic tilt, pelvic incidence, T4-T12 kyphosis, L1-L4 lordosis, and SVA showed consistently high ICCs (>0.80) and low mean absolute deviations (MADs <6°), with substantial number of studies reporting pelvic tilt achieving an excellent ICC of 0.90 or greater. In contrast, T1-T12 kyphosis and L4-S1 lordosis exhibited lower ICCs and higher measurement errors. Overall, most DL models demonstrated strong correlations (>0.80) with clinician measurements and minimal differences compared to manual references, except for T1-T12 kyphosis (average Pearson correlation: 0.68), L1-L4 lordosis (average Pearson correlation: 0.75), and L4-S1 lordosis (average Pearson correlation: 0.65).

CONCLUSION

Novel computer vision algorithms show promising accuracy in measuring spinopelvic parameters, comparable to manual surgeon measurements. Future research should focus on external validation, additional imaging modalities, and the feasibility of integration in clinical settings to assess model reliability and predictive capacity.

A new human spine model for use in cinematographic gait analysis

PURPOSE

The aim of this study was to compare the accuracy of two spine models: the broken curve model and a new four tangent circles model. The modification concerns the adaptation of data acquisition to kinematic methods used in, e.g., gait and running analysis.

METHOD

Plastic, movable spine model of human with flexible intervertebral disks (manufactured by Erler Zimmer GE3014) was used as the study material. Markers with a diameter of 5 mm were glued to each spinous process (from C7 to L5). The recording was performed with a 6-camera Vicon system. Two spine models were created: a broken curve model used, among others, in the Diers scanner, and an own model of 4 circles, similar to the model of circles used in X-ray and CT analysis.

RESULTS

The errors in the position of the spinous processes were significantly smaller in the 4-circle model than in the broken curve model. They ranged from 0.01 to 6.5 mm in the lumbar section, from 0.004 to 3.1 mm in the thoracic section. The practical possibilities of using the four-circle model during the cinematographic analysis of gait and run should be checked.

CONCLUSION

The four-circle model is more accurate than the broken curve model and can be used in the cinematographic analysis of the human spine movement.

Convolutional Neural Network-Based Approach for Cobb Angle Measurement Using Mask R-CNN

Background: Scoliosis is a disorder characterized by an abnormal spinal curvature, which can lead to negative effects on patients, affecting their quality of life. Given its progressive nature, the classification of the scoliosis severity requires an accurate diagnosis and effective monitoring. The Cobb angle measurement method has been widely considered as the gold standard for a scoliosis assessment. Commonly, an expert assesses scoliosis severity manually by identifying the most tilted vertebrae of the spine. However, this method requires time, effort, and presents limitations in measurement accuracy, such as the intra- and inter-observer variability. Artificial intelligence provides more objective tools that are less sensitive to manual intervention aiming to transform the diagnosis of scoliosis. Objectives: The objective of this study was to address three key research questions regarding automated Cobb angle quantification: "Where is the spine in this radiograph?", "What is its exact shape?", and "Is the proposed method accurate?". We propose the use of Mask R-CNN architecture for spine detection and segmentation in response to the first two questions, and a set of algorithms to tackle the third. Methods: The network's detection and segmentation performance was evaluated through various metrics. An automated workflow for Cobb angle quantification and severity classification was developed. Finally, statistical methods provided the agreement between manual and automated measurements. Results: A high segmentation accuracy was achieved, highlighting the following: mIoU of 0.8012, and a mean precision of 0.9145. MAE was 2.96° ± 2.60° demonstrating a high agreement. Conclusions: The results obtained in this study demonstrate the potential of the proposed automated approach in clinical scenarios, which provides experts with a clear visualization of each stage in the scoliosis assessment by overlaying the results onto the X-ray image.

Adjacent point aided vertebral landmark detection and Cobb angle measurement for automated AIS diagnosis

Adolescent Idiopathic Scoliosis (AIS) is a prevalent structural deformity disease of human spine, and accurate assessment of spinal anatomical parameters is essential for clinical diagnosis and treatment planning. In recent years, significant progress has been made in automatic AIS diagnosis based on deep learning methods. However, effectively utilizing spinal structure information to improve the parameter measurement and diagnosis accuracy from spinal X-ray images remains challenging. This paper proposes a novel spine keypoint detection framework to complete the intelligent diagnosis of AIS, with the assistance of spine rigid structure information. Specifically, a deep learning architecture called Landmark and Adjacent offset Detection (LAD-Net) is designed to predict spine centre and corner points as well as their related offset vectors, based on which error-detected landmarks can be effectively corrected via the proposed Adjacent Centre Iterative Correction (ACIC) and Corner Feature Optimization and Fusion (CFOF) modules. Based on the detected spine landmarks, spine key parameters (i.e. Cobb angles) can be computed to finish the AIS Lenke diagnosis. Experimental results demonstrate the superiority of the proposed framework on spine landmark detection and Lenke classification, providing strong support for AIS diagnosis and treatment.

Capturing the cervical spine shape: Angular measurements versus geometric morphometric methods

The cervical spine manifests a wide shape variation. However, the traditional methods to evaluate the cervical spine curve were never tested against its actual shape. The study's main aim was to determine whether the shape classification of the cervical spine, based on traditional angular measurements, coincides with each other and with the shape captured by the 2D landmark-based geometric morphometric method. The study's second aim was to reveal the associations between the cervical spine shape and the demographic parameters, the head's position, and the spine's sagittal balance. CT scans of the cervical spine of 163 individuals were evaluated to achieve these goals. The shape was assessed by measuring the C2-C7 Cobb angle (CA), the C2-C7 posterior tangent angle (PTA), the curvedness of the arch, and by a 2D landmark-based geometric morphometric method. The position of the head and the sagittal balance of the spine were evaluated by measuring the foramen magnum-C2 Cobb angle (FMCA) and the T1 slope angle (T1SA), respectively. Based on the size of the angle measured, each individual was classified into one of the three cervical 'shape groups' (lordotic, straight, and kyphotic). We found that cervical lordosis was the dominant shape regardless of the measuring methods utilized (46.6%-54.6%), followed by straight neck (28.2%-30.1%), and kyphosis (15.3%-25.2%); however, about a third of the 163 individuals were classified into a different shape group using the CA and PTA methods. The cervical spine angle was sex-independent and age-dependent. The T1SA was significantly correlated with CA and PTA (r = 0.640 and r = 0.585, respectively; p < 0.001). In conclusion, the cervical spine shape evaluation is method-dependent and varies with age.

Ultrasound vs. x-ray: a new way for clinicians to track scoliosis progression?

This retrospective study, utilising prospectively collected data, investigates the use of spine ultrasound as an alternative method for assessing scoliosis, with the aim of reducing radiation exposure. We included 92 patients aged 10 to 16 years with suspected idiopathic scoliosis. Exclusion criteria were weight over 150 kg, metal implants, pre-existing conditions, secondary deformities, and cognitive impairments. Each patient underwent clinical assessment and full spine radiographs, followed by spine ultrasound using the Scolioscan® system. Unprocessed B-mode ultrasound images were analysed using automatic measurements. The correlation between Ultrasound Coronal Angle (UCA) and Radiographic Cobb Angle (RCA) was evaluated at initial and follow-up visits. Strong correlations were found between UCA and RCA, with correlation coefficients ranging from 0.786 to 0.903 (p<0.001). The regression formula showed good predictive accuracy for curve progression on follow-up radiographs. The best results were observed in females and in primary thoracic curves (r = 0.936, p<0.001). Although only four patients exhibited true progression (≥5° increase in Cobb angle), changes in scoliotic angles were effectively detected using ultrasound. This study confirms the feasibility of unprocessed spine ultrasound for scoliosis monitoring in clinical settings. Automatic measurements without 3D reconstruction make ultrasound a practical tool for tracking progression. The regression model shows potential for predicting curve progression, although further validation is needed. These findings suggest spine ultrasound could reduce the need for radiographs, benefiting patients by minimising radiation exposure while providing reliable monitoring of scoliosis progression and treatment outcomes.

Three-dimensional (3D) ultrasound imaging for quantitative assessment of frontal cobb angles in patients with idiopathic scoliosis - a systematic review and meta-analysis

BACKGROUND

Measurement of Cobb angle in the frontal plane from radiographs is the gold standard of quantifying spinal deformity in adolescent idiopathic scoliosis (AIS). As a radiation free alternative, ultrasonography (USG) for quantitative measurement of frontal cobb angles has been reported. However, a systematic review and meta-analysis on the reliability of ultrasound comparing with the gold standard have not yet been reported.

OBJECTIVES

This systematic review and meta-analysis aimed to evaluate (1) the reliability of ultrasound imaging compared with radiographs in measuring frontal cobb angle for screening or monitoring in AIS patients; (2) whether the performance of USG differ when using different anatomical landmarks for measurement of frontal cobb angles.

METHODS

Systematic search was performed on MEDLINE, EMBASE, CINAHL, and CENTRAL databases for relevant studies. QUADAS-2 was adopted for quality assessment. The intra- and inter-rater reliability of ultrasound measurement in terms of intra-class correlation coefficient (ICC) was recorded. Mean Absolute Difference (MAD) and Pearson correlation coefficients between frontal cobb angle measured from USG and radiographic measurements, were extracted with meta-analysis performed.

RESULTS AND DISCUSSION

Nineteen studies were included with a total of 2318 patients. The risk of bias of included studies were unclear or high. Pooled MAD of frontal cobb angle measured between USG and radiography was 4.02 degrees (95% CI: 3.28-4.76) with a pooled correlation coefficient of 0.91 (95% CI: 0.87-0.93). Subgroup analyses show that pooled correlation was > 0.87 across using various USG landmarks for measurement of frontal cobb angles. There was a high level of heterogeneity between results of the included studies with I2 > 90%. Potential sources of heterogeneity include curve severity, curve types, location of apex, scanning postures, patient demographics, equipment, and operator experience. Despite being the "gold standard", intrinsic errors in quantifying spinal deformities with radiographs may also be a source of inconsistency.

CONCLUSION

The current systematic review indicated that there is evidence in favor of using USG for quantitative evaluation of frontal cobb angle in AIS. However, the quality of evidence is low due to high risk of bias and heterogeneity between existing studies. Current literature is insufficient to support the use of USG as a screening and/or follow-up method for AIS. Further investigation addressing the limitations identified in this review is required before USG could be adapted for further clinical use.

Which Method of the Radiologic Measurements of the Angle of Curvature in Idiopathic Scoliosis is the Most Reliable for an Inexperienced Researcher?

Aim

The aim of the examination was to determine which of the three measurement methods Cobb (CB), Ferguson (FR), and Centroid (CN) has the best repeatability and reliability when the measurements are made by inexperienced researchers.

Methods

Three researchers (from the student research group) measured the angle of spine curvature on X-rays of the entire spine in standing anteroposterior view in 50 patients with severe idiopathic scoliosis qualified for surgery. Cobb, Ferguson, and Centroid methods were used. One of the researchers repeated all examinations twice at 3-week intervals. The measurements were compared with each other using the intraclass correlation coefficient (ICC) method. Values less than 0.5 are indicative of poor reliability, values between 0.5 and 0.75 indicate moderate reliability, values between 0.75 and 0.9 indicate good reliability and values greater than 0.90 indicate excellent reliability.

Results

The ICC (inter-rater) between the researchers' measurements was 0.9387 for CB, 0.9169 for FR, and 0.9061 for CR. Whereas the ICC (intra-rater) between measurements taken by a single researcher was 0.9824 for CB, 0.9088 for FR, and 0.9546 for CR.

Conclusions

The above results show that Cobb angle measurement method is the most reliable for measuring the curvature angle of the spine for novice researchers. Although it seems to be difficult to measure, it provides the most repeatable results.

Changes in the shape of the lumbar curve during growth : a geometric morphometric approach

Aims

The development of lumbar lordosis has been traditionally examined using angular measurements of the spine to reflect its shape. While studies agree regarding the increase in the angles during growth, the growth rate is understudied, and sexual dimorphism is debated. In this study, we used a novel method to estimate the shape of the lumbar curve (LC) using the landmark-based geometric morphometric method to explore changes in LC during growth, examine the effect of size and sex on LC shape, and examine the associations between angular measurements and shape.

Methods

The study population included 258 children aged between 0 and 20 years (divided into five age groups) who underwent a CT scan between the years 2009 and 2019. The landmark-based geometric morphometric method was used to capture the LC shape in a sagittal view. Additionally, the lordosis was measured via Cobb and sacral slope angles. Multivariate and univariate statistical analyses were carried out to examine differences in shape between males and females and between the age groups.

Results

The overall shape of the LC overlapped between males and females in most age groups, except for the nine- to 12-year age group. However, size did not affect LC shape. LC shape changed significantly during growth from straight to curved, reaching its mature shape earlier in females. This corresponded with the results obtained by the lordosis and sacral slope angles. A significant positive correlation was found between the LC shape and angles, although the angles demonstrated poor distinction between age groups, as opposed to the LC shape.

Conclusion

New insights into LC shape development were achieved using the geometrical morphometric method. The LC shape was sex-independent in most age groups. However, the LC reached its mature shape earlier in females than males. The method and data of this study are beneficial for future studies examining aetiological factors for spinal pathologies and maldevelopment.

Intelligent measurement of adolescent idiopathic scoliosis x-ray coronal imaging parameters based on VB-Net neural network: a retrospective analysis of 2092 cases

BACKGROUND

Adolescent idiopathic scoliosis (AIS) is a complex three-dimensional deformity, and up to now, there has been no literature reporting the analysis of a large sample of X-ray imaging parameters based on artificial intelligence (AI) for it. This study is based on the accurate and rapid measurement of x-ray coronal imaging parameters in AIS patients by AI, to explore the differences and correlations, and to further investigate the risk factors in different groups, so as to provide a theoretical basis for the diagnosis and surgical treatment of AIS.

METHODS

Retrospective analysis of 3192 patients aged 8-18 years who had a full-length orthopantomogram of the spine and were diagnosed with AIS at the First Affiliated Hospital of Zhengzhou University from January 2019 to March 2024. After screened 2092 cases were finally included. The uAI DR scoliosis analysis system with multi-resolution VB-Net convolution network architecture was used to measure CA, CBD, CV, RSH, T1 Tilt, PT, LLD, SS, AVT, and TS parameters. The results were organized and analyzed by using R Studio 4.2.3 software.

RESULTS

The differences in CA, CBD, CV, RSH, TI tilt, PT, LLD and SS were statistically significant between male and female genders (p < 0.05); Differences in CA, CBD, T1 Tilt, PT, SS, AVT and TS were statistically significant in patients with AIS of different severity (p < 0.001), and T1 Tilt, AVT, TS were risk factors; Differences in CA, CBD, CV, RSH, T1 Tilt, PT, LLD, SS, AVT and TS were statistically significant (p < 0.05) in patients with AIS of different curve types, and TS was a risk factor; Analyzing the correlation between parameters revealed a highly linear correlation between CV and RSH (r = 0.826, p < 0.001), and a significant linear correlation between CBD and TS, and PT and SS (r = 0.561, p < 0.001; r = 0.637, p < 0.001).

CONCLUSION

Measurements based on VB-Net neural network found that x-ray coronal imaging parameters varied among AIS patients with different curve types and severities. In clinical practice, it is recommended to consider the discrepancy in parameters to enable a more accurate diagnosis and a personalized treatment plan.

Machine learning identifies clusters of the normal adolescent spine based on sagittal balance

PURPOSE

This study applied a machine learning semi-supervised clustering approach to radiographs of adolescent sagittal spines from a single pediatric institution to identify patterns of sagittal alignment in the normal adolescent spine. We sought to explore the inherent variability found in adolescent sagittal alignment using machine learning to remove bias and determine whether clusters of sagittal alignment exist.

METHODS

Multiple semi-supervised machine learning clustering algorithms were applied to 111 normal adolescent sagittal spines. Sagittal parameters for resultant clusters were determined.

RESULTS

Machine learning analysis found that the spines did cluster into distinct groups with an optimal number of clusters ranging from 3 to 5. We performed an analysis on both 3 and 5-cluster groups. The 3-cluster groups analysis found good consistency between methods with 96 of 111, while the analysis of 5-cluster groups found consistency with 105 of 111 spines. When assessing for differences in sagittal parameters between the groups for both analyses, there were differences in T4-12 TK, L1-S1 LL, SS, SVA, PI-LL mismatch, and TPA. However, the only parameter that was statistically different for all groups was SVA.

CONCLUSIONS

Based on machine learning, the adolescent sagittal spine alignments do cluster into distinct groups. While there were distinguishing features with TK and LL, the most important parameter distinguishing these groups was SVA. Further studies may help to understand these findings in relation to spinal deformities.

Normative vertebral deformity measurements in a clinically relevant population using magnetic resonance imaging

BACKGROUND

Osteoporosis is the leading cause of vertebral fractures. Dual-energy X-ray absorptiometry (DXA) and radiographs are traditionally used to detect osteoporosis and vertebral fractures/deformities. Magnetic resonance imaging (MRI) can be utilized to detect the relative severity of vertebral deformities using three-dimensional information not available in traditional DXA and lateral two-dimensional radiography imaging techniques.

AIM

To generate normative vertebral parameters in women using MRI and DXA scans, determine the correlations between MRI-calculated vertebral deformities and age, DXA T-scores, and DXA Z-scores, and compare MRI vertebral deformity values with radiography values previously published in the literature.

METHODS

This study is a retrospective vertebral morphometric analysis conducted at our institution. The patient sample included MR images from 1638 female patients who underwent both MR and DXA imaging between 2005 and 2014. Biconcavity, wedge, crush, anterior height (Ha)/posterior height (Hp), and middle height (Hm)/posterior height values were calculated from the MR images of the patient's vertebrae. Associations between vertebral deformity values, patient age, and DXA T-scores were analyzed using Spearman correlation. The MRI-derived measurements were compared with radiograph-based calculations from population-based data compiled from multiple studies.

RESULTS

Age was positively correlated with lumbar Hm/Hp (P = 0.04) and thoracic wedge (P = 0.03) and biconcavity (P = 0.001) and negatively correlated with thoracic Ha/Hp (P = 0.002) and Hm/Hp (P = 0.001) values. DXA T-scores correlated positively with lumbar Hm/Hp (P < 0.0001) and negatively with lumbar wedge (P = 0.046), biconcavity (P < 0.0001), and Ha/Hp (P = 0.046) values. Qualitative analysis revealed that Ha/Hp differed between MRI and radiography population-based data by no more than 0.3 and Hm/Hp by a maximum of 1.2.

CONCLUSION

Compared with traditional imaging techniques, MRI detects vertebral deformities with high accuracy and reliability. It may be a sensitive, ionizing, radiation-free tool for use in clinical settings.

Superior performance of a center-point AI model over VFLDNet in automated cobb angle estimation for scoliosis assessment

PURPOSE

Aims to establish the superiority of our proposed model over the state-of-the-art vertebra-focused landmark detection network (VFLDNet) in automating Cobb angle estimation from spinal radiographs.

METHODS

Utilizing a private dataset for external validation, we compared the performance of our center-point detection-based vertebra localization and tilt estimation network (VLTENet) with the key-point detection-based VFLDNet. Both models' Cobb angle predictions were rigorously evaluated against manual consensus score using metrics such as mean absolute error (MAE), correlation coefficient, intraclass correlation coefficient (ICC), Fleiss' kappa, Bland-Altman analysis, and classification metrics [sensitivity (SN), specificity, accuracy] focusing on major curve estimation and scoliosis severity classification.

RESULTS

A retrospective analysis of 118 cases with 342 Cobb angle measurements revealed that our model achieved a MAE of 2.15° for total Cobb angles and 1.89° for the major curve, significantly outperforming VFLDNet's MAE of 2.80°and 2.57°, respectively. Both models demonstrated robust correlation and ICC, but our model excelled in classification consistency, particularly in predicting major curve magnitude (ours: kappa = 0.83; VFLDNet: kappa = 0.67). In subgroup analyses by scoliosis severity, our model consistently surpassed VFLDNet, displaying superior mean (SD) differences, narrower limits of agreement, and higher SN, specificity, and accuracy, most notably in moderate (ours: SN = 86.84%; VFLDNet: SN = 83.16%) to severe (ours: SN = 92.86%; VFLDNet: SN = 85.71%) scoliosis.

CONCLUSION

Our model emerges as the superior choice for automated Cobb angle estimation, particularly in assessing major curve and moderate to severe scoliosis, underscoring its potential to revolutionize clinical workflows and enhance patient care.

The efficacy of combined physiotherapeutic scoliosis-specific exercises and manual therapy in adolescent idiopathic scoliosis

BACKGROUND

Adolescent idiopathic scoliosis (AIS) is a pathological condition characterized by vertebral curvature and associated trunk deformities in adolescents. The clinical efficacy of conservative treatment in alleviating spinal curvature of AIS remains a topic of ongoing debate. The objective of this study was to investigate the impact of combined physiotherapeutic scoliosis-specific exercises (PSSE) and manual therapy (MT) on trunk deformity, spinal function, mobility, and mental health in patients with AIS.

METHODS

Thirty-one participants who were diagnosed with AIS whose Cobb angle was between 10-45°were enrolled in the study. Participants in the intervention group received 50 min of PSSE combined with 10 min of MT, while the control group performed 50 min of PSSE as their home exercise program. Both treatments were implemented three times a week for four weeks. Cobb angle, spinal mobility, trunk morphology (vertebral rotation angle, apical deviation, pelvic obliquity distance and angle), movement capability, and quality of life (QOL) were assessed at baseline and post intervention. The treatment effects between the intervention and control groups were analyzed using a two-way repeated measures ANOVA.

RESULTS

Following a 4-week treatment period, Cobb angle was significantly reduced from 21.58° to 18.58° in intervention group and increased from 18.00° at baseline and 19.14° post intervention in the control group. Significant improvements were also observed in spinal mobility, movement capability, quality of life, and some of the trunk morphology indices in the intervention group compared to baseline (p < 0.05). Improvements were significantly higher in the intervention group than the control group.

CONCLUSION

Combining PSSE and MT shows potential benefits in alleviating AIS symptoms and improving QOL. Further studies to substantiate these findings are warranted.

TRIAL REGISTRATION

The trial was retrospectively registered in the Chinese Clinical Trial Registry ( https://www.chictr.org.cn ) with the registration number: ChiCTR2300071357, (Date: 12/05/2023).

Correlation between sagittal balance and thoracolumbar elastic energy parameters in 42 spines subject to spondylolisthesis or spinal stenosis and 21 normal spines

The curvature of the lumbar spine plays a critical role in maintaining spinal function, stability, weight distribution, and load transfer. We have developed a mathematical model of the lumbar spine curve by introducing a novel mechanism: minimization of the elastic bending energy of the spine with respect to two biomechanical parameters: dimensionless lumbosacral spinal curvature c LS and dimensionless curvature increment along the spine CI. While most of the biomechanical studies focus on a particular segment of the spine, the distinction of the presented model is that it describes the shape of the thoracolumbar spine by considering it as a whole (non-locally) and thus includes interactions between the different spinal levels in a holistic approach. From radiographs, we have assessed standard geometrical parameters: lumbar lordosis LL, pelvic incidence PI, pelvic tilt PT, sacral slope ψ0 and sagittal balance parameter SB = sagittal vertical axis (SVA)/sacrum-bicoxofemoral distance (SFD) of 42 patients with lumbar spinal stenosis (SS) or degenerative spondylolisthesis (SL) and 21 radiologically normal subjects. SB statistically significantly correlated with model parameters c L5 (r = -0.34, p = 0.009) and -CI (r = 0.33, p = 0.012) but not with standard geometrical parameters. A statistically significant difference with sufficient statistical power between the patients and the normal groups was obtained for c LS, CI, and SB but not for standard geometrical parameters. The model provides a possibility to predict changes in the thoracolumbar spine shape in surgery planning and in assessment of different spine pathologies.

Reliability of two smartphone inclinometer apps in the measurement of dorsal kyphosis in three different positions

BACKGROUND

The exploration of the spine in the sagittal plane is of great relevance in the health area. The inclinometer has proved to be reliable in the assessment of this sagittal plane. Due to the growth of technology, a number of smartphone apps which simulates the action of the inclinometer have been developed.

OBJECTIVES

to analyze the reliability of two Smartphone Inclinometer Apps in the assessment of the dorsal kyphosis of the spine in three different positions when comparing it to a traditional inclinometer and determine if there are differences according to gender.

DESIGN

observational, cross-sectional study.

METHODS

25 healthy voluntaries were included. A traditional inclinometer and two smartphone apps, Clinometer® and Rotating Sphere Inclinometer® were used. Dorsal kyphosis was measured in three different positions: relaxed standing, relaxing seated position and standing anterior flexion.

RESULTS

no statistical differences were found between the measurements, and an excellent correlation was obtained comparing the tree devices, observing no differences according to gender CONCLUSION: The two apps are reliable and comparable to a traditional inclinometer as measuring sagittal curves of the spine. Besides, no variations in reliability according to sex were observed.

Comparison of three artificial intelligence algorithms for automatic cobb angle measurement using teaching data specific to three disease groups

Spinal deformities, including adolescent idiopathic scoliosis (AIS) and adult spinal deformity (ASD), affect many patients. The measurement of the Cobb angle on coronal radiographs is essential for their diagnosis and treatment planning. To enhance the precision of Cobb angle measurements for both AIS and ASD, we developed three distinct artificial intelligence (AI) algorithms: AIS/ASD-trained AI (trained with both AIS and ASD cases); AIS-trained AI (trained solely on AIS cases); ASD-trained AI (trained solely on ASD cases). We used 1612 whole-spine radiographs, including 1029 AIS and 583 ASD cases with variable postures, as teaching data. We measured the major and two minor curves. To assess the accuracy, we used 285 radiographs (159 AIS and 126 ASD) as a test set and calculated the mean absolute error (MAE) and intraclass correlation coefficient (ICC) between each AI algorithm and the average of manual measurements by four spine experts. The AIS/ASD-trained AI showed the highest accuracy among the three AI algorithms. This result suggested that learning across multiple diseases rather than disease-specific training may be an efficient AI learning method. The presented AI algorithm has the potential to reduce errors in Cobb angle measurements and improve the quality of clinical practice.

A multipurpose, adolescent idiopathic scoliosis-specific, short MRI protocol: A feasibility study in volunteers

INTRODUCTION

Visualization of scoliosis typically requires ionizing radiation (radiography and CT) to visualize bony anatomy. MRI is often additionally performed to screen for neural axis abnormalities. We propose a 14-minutes radiation-free scoliosis-specific MRI protocol, which combines MRI and MRI-based synthetic CT images to visualize soft and osseous structures in one examination. We assess the ability of the protocol to visualize landmarks needed to detect 3D patho-anatomical changes, screen for neural axis abnormalities, and perform surgical planning and navigation.

METHODS

18 adult volunteers were scanned on 1.5 T MR-scanner using 3D T2-weighted and synthetic CT sequences. A predefined checklist of relevant landmarks was used for the parameter assessment by three readers. Parameters included Cobb angles, rotation, torsion, segmental height, area and centroids of Nucleus Pulposus and Intervertebral Disc. Precision, reliability and agreement between the readers measurements were evaluated.

RESULTS

91 % of Likert-based questions scored ≥ 4, indicating moderate to high confidence. Precision of 3D dot positioning was 1.0 mm. Precision of angle measurement was 0.6° (ICC 0.98). Precision of vertebral and IVD height measurements was 0.4 mm (ICC 0.99). Precision of area measurement for NP was 8 mm2 (ICC 0.55) and for IVD 18 mm2 (ICC 0.62) for IVD. Precision of centroid measurement for NP was 1.3 mm (ICC 0.88-0.92) and for IVD 1.1 mm (ICC 0.88-91).

CONCLUSIONS

The proposed MRI protocol with synthetic CT reconstructions, has high precision, reliability and agreement between the readers for multiple scoliosis-specific measurements. It can be used to study scoliosis etiopathogenesis and to assess 3D spinal morphology.

Pisa Syndrome Secondary to Drugs: A Scope Review

BACKGROUND

Pisa syndrome, also known as pleurothotonus, is a neurological condition characterized by more than ten degrees of constant lateral curvature of the spine when upright. In this way, the present manuscript aims to systematically review Pisa syndrome secondary to drugs.

METHODS

Two reviewers identified and assessed relevant reports in six databases without language restriction between January 1990 and June 2024.

RESULTS

The prevalence of Pisa syndrome varied from 0.037 to 9.3%. We found 109 articles containing 191 cases of drug-induced Pisa syndrome reported in the literature. The mean and median ages were 59.70 (SD = 19.02) and 67 (range = 12-98 years). The most prevalent sex was female, 56.91% (107/188). The most frequent medications associated with Pisa syndrome were acetylcholinesterase inhibitors in 87 individuals. Of 112 individuals in which the onset time from the medication to the movement disorder occurrence was reported, 59 took place within a month. In this way, a return to baseline was observed in 45.50% of the cases, and partial recovery was observed in 14.28%.

CONCLUSION

We proposed new diagnostic criteria for Pisa syndrome based on previous findings in the literature. Moreover, multiple mechanisms are probably involved in balance control and the development of lateral trunk flexions.

Explainable Deep-Learning-Based Gait Analysis of Hip-Knee Cyclogram for the Prediction of Adolescent Idiopathic Scoliosis Progression

Accurate prediction of scoliotic curve progression is crucial for guiding treatment decisions in adolescent idiopathic scoliosis (AIS). Traditional methods of assessing the likelihood of AIS progression are limited by variability and rely on static measurements. This study developed and validated machine learning models for classifying progressive and non-progressive scoliotic curves based on gait analysis using wearable inertial sensors. Gait data from 38 AIS patients were collected using seven inertial measurement unit (IMU) sensors, and hip-knee (HK) cyclograms representing inter-joint coordination were generated. Various machine learning algorithms, including support vector machine (SVM), random forest (RF), and novel deep convolutional neural network (DCNN) models utilizing multi-plane HK cyclograms, were developed and evaluated using 10-fold cross-validation. The DCNN model incorporating multi-plane HK cyclograms and clinical factors achieved an accuracy of 92% in predicting curve progression, outperforming SVM (55% accuracy) and RF (52% accuracy) models using handcrafted gait features. Gradient-based class activation mapping revealed that the DCNN model focused on the swing phase of the gait cycle to make predictions. This study demonstrates the potential of deep learning techniques, and DCNNs in particular, in accurately classifying scoliotic curve progression using gait data from wearable IMU sensors.

Sagittal balance parameters measurement on cervical spine MR images based on superpixel segmentation

Introduction: Magnetic Resonance Imaging (MRI) is essential in diagnosing cervical spondylosis, providing detailed visualization of osseous and soft tissue structures in the cervical spine. However, manual measurements hinder the assessment of cervical spine sagittal balance, leading to time-consuming and error-prone processes. This study presents the Pyramid DBSCAN Simple Linear Iterative Cluster (PDB-SLIC), an automated segmentation algorithm for vertebral bodies in T2-weighted MR images, aiming to streamline sagittal balance assessment for spinal surgeons. Method: PDB-SLIC combines the SLIC superpixel segmentation algorithm with DBSCAN clustering and underwent rigorous testing using an extensive dataset of T2-weighted mid-sagittal MR images from 4,258 patients across ten hospitals in China. The efficacy of PDB-SLIC was compared against other algorithms and networks in terms of superpixel segmentation quality and vertebral body segmentation accuracy. Validation included a comparative analysis of manual and automated measurements of cervical sagittal parameters and scrutiny of PDB-SLIC's measurement stability across diverse hospital settings and MR scanning machines. Result: PDB-SLIC outperforms other algorithms in vertebral body segmentation quality, with high accuracy, recall, and Jaccard index. Minimal error deviation was observed compared to manual measurements, with correlation coefficients exceeding 95%. PDB-SLIC demonstrated commendable performance in processing cervical spine T2-weighted MR images from various hospital settings, MRI machines, and patient demographics. Discussion: The PDB-SLIC algorithm emerges as an accurate, objective, and efficient tool for evaluating cervical spine sagittal balance, providing valuable assistance to spinal surgeons in preoperative assessment, surgical strategy formulation, and prognostic inference. Additionally, it facilitates comprehensive measurement of sagittal balance parameters across diverse patient cohorts, contributing to the establishment of normative standards for cervical spine MR imaging.

Landet: an efficient physics-informed deep learning approach for automatic detection of anatomical landmarks and measurement of spinopelvic alignment

PURPOSE

An efficient physics-informed deep learning approach for extracting spinopelvic measures from X-ray images is introduced and its performance is evaluated against manual annotations.

METHODS

Two datasets, comprising a total of 1470 images, were collected to evaluate the model's performance. We propose a novel method of detecting landmarks as objects, incorporating their relationships as constraints (LanDet). Using this approach, we trained our deep learning model to extract five spine and pelvis measures: Sacrum Slope (SS), Pelvic Tilt (PT), Pelvic Incidence (PI), Lumbar Lordosis (LL), and Sagittal Vertical Axis (SVA). The results were compared to manually labelled test dataset (GT) as well as measures annotated separately by three surgeons.

RESULTS

The LanDet model was evaluated on the two datasets separately and on an extended dataset combining both. The final accuracy for each measure is reported in terms of Mean Absolute Error (MAE), Standard Deviation (SD), and R Pearson correlation coefficient as follows: [ S S ∘ : 3.7 ( 2.7 ) , R = 0.89 ] ,[ P T ∘ : 1.3 ( 1.1 ) , R = 0.98 ] , [ P I ∘ : 4.2 ( 3.1 ) , R = 0.93 ] , [ L L ∘ : 5.1 ( 6.4 ) , R = 0.83 ] , [ S V A ( m m ) : 2.1 ( 1.9 ) , R = 0.96 ] . To assess model reliability and compare it against surgeons, the intraclass correlation coefficient (ICC) metric is used. The model demonstrated better consistency with surgeons with all values over 0.88 compared to what was previously reported in the literature.

CONCLUSION

The LanDet model exhibits competitive performance compared to existing literature. The effectiveness of the physics-informed constraint method, utilized in our landmark detection as object algorithm, is highlighted. Furthermore, we addressed the limitations of heatmap-based methods for anatomical landmark detection and tackled issues related to mis-identifying of similar or adjacent landmarks instead of intended landmark using this novel approach.

Unsupervised local center of mass based scoliosis spinal segmentation and Cobb angle measurement

Scoliosis is a medical condition in which a person's spine has an abnormal curvature and Cobb angle is a measurement used to evaluate the severity of a spinal curvature. Presently, automatic Existing Cobb angle measurement techniques require huge dataset, time-consuming, and needs significant effort. So, it is important to develop an unsupervised method for the measurement of Cobb angle with good accuracy. In this work, an unsupervised local center of mass (LCM) technique is proposed to segment the spine region and further novel Cobb angle measurement method is proposed for accurate measurement. Validation of the proposed method was carried out on 2D X-ray images from the Saudi Arabian population. Segmentation results were compared with GMM-Based Hidden Markov Random Field (GMM-HMRF) segmentation method based on sensitivity, specificity, and dice score. Based on the findings, it can be observed that our proposed segmentation method provides an overall accuracy of 97.3% whereas GMM-HMRF has an accuracy of 89.19%. Also, the proposed method has a higher dice score of 0.54 compared to GMM-HMRF. To further evaluate the effectiveness of the approach in the Cobb angle measurement, the results were compared with Senior Scoliosis Surgeon at Multispecialty Hospital in Saudi Arabia. The findings indicated that the segmentation of the scoliotic spine was nearly flawless, and the Cobb angle measurements obtained through manual examination by the expert and the algorithm were nearly identical, with a discrepancy of only ± 3 degrees. Our proposed method can pave the way for accurate spinal segmentation and Cobb angle measurement among scoliosis patients by reducing observers' variability.

Effects of Pilates exercises on spine deformities and posture: a systematic review

BACKGROUND

Pilates is becoming increasingly popular amongst a wide range of people and is gaining more attention. It is also an effective means of physical rehabilitation. The aim of this systematic review is to explore the effects of Pilates on spinal deformity and posture.

METHOD

This systematic review was conducted using four recognised academic and scientific databases (Scopus, Web of Science, PubMed and Cochrane) to identify articles that met the inclusion criteria. The secondary search used the Google Scholar and the Science Direct search engines. The search for articles for this review began in July 06, 2023 and was concluded on February 01, 2024. The search process for this study was documented using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA 2020). The PEDro scale was used to assess the internal validity and data statistics of the studies included in this systematic review and to evaluate the quality of the studies.

RESULTS

The systematic review included nine studies that met the inclusion criteria from the 651 studies retrieved, involving a total of 643 participants. The PEDro scale scores of the studies included in this systematic review ranged from 3 to 8. The intervention was in the form of Pilates or Pilates combined exercises. The studies included in this review used outcome measures of Cobb angle, angle of trunk rotation (ATR), range of motion (ROM), chest expansion, Scoliosis Research Society Questionnaire (SRS-22r) and postural assessment. Research has shown that Pilates is effective in correcting spinal deformities and posture, as well as improving quality of life, pain relief, function and fitness.

CONCLUSIONS

This systematic review provide substantial evidence that Pilates has a positive impact on improving spinal deformity and posture. However, more research is needed to validate whether Pilates can be used effectively as a physical therapy for spinal deformity rehabilitation. Pilates has considerable potential for public health interventions.

Predicting cervico-thoraco-lumbar vertebra positions from cutaneous markers: Combining local frame and postural predictors improves robustness to posture

Predictions of vertebra positions from external data are required in many fields like motion analysis or for clinical applications. Existing predictions mainly cover the thoraco-lumbar spine, in one posture. The objective of this study was to develop a method offering robust vertebra position predictions in different postures for the whole spine, in the sagittal plane. EOS radiographs were taken in three postures: slouched, erect, and subject's usual sitting posture, using 21 healthy participants pre-equipped with opaque cutaneous markers. Local curvilinear Frenet frames were built on a spline fitted to spinous processes' cutaneous markers. Vertebra positions were expressed as polar coordinates in these frames, defining an angle (α) and distance (d). Multilinear regressions were fitted to explain α and d from anthropometric predictors and predictors presumed to be linked to spinal posture, the predictors' effects being considered both locally and remotely. Anthropometric predictors were the main predictors for d distances, and postural predictors for α angles, with postural predictors still showing a marked influence on d distances for the cervical spine. Vertebra positions were then predicted by cross-validation. The average RMSE on vertebra positions was 11.0 ± 3.7 mm across the entire spine, 13.4 ± 4.1 mm across the cervical spine and 10.1 ± 3.1 mm across the thoraco-lumbar spine for all participants and postures, performances similar to previous models designed for a single posture. Our simple geometrical and statistical model thus appears promising for predicting vertebra positions from external data in several spinal postures and for the whole spine.

Concurrent Validity of the Ergotex Device for Measuring Low Back Posture

Highlighting the crucial role of monitoring and quantifying lumbopelvic rhythm for spinal curvature, the Ergotex IMU, a portable, lightweight, cost-effective, and energy-efficient technology, has been specifically designed for the pelvic and lumbar area. This study investigates the concurrent validity of the Ergotex device in measuring sagittal pelvic tilt angle. We utilized an observational, repeated measures design with healthy adult males (mean age: 39.3 ± 7.6 y, body mass: 82.2 ± 13.0 kg, body height: 179 ± 8 cm), comparing Ergotex with a 3D optical tracking system. Participants performed pelvic tilt movements in anterior, neutral, and posterior conditions. Statistical analysis included paired samples t-tests, Bland-Altman plots, and regression analysis. The findings show minimal systematic error (0.08° overall) and high agreement between the Ergotex and optical tracking, with most data points falling within limits of agreement of Bland-Altman plots (around ±2°). Significant differences were observed only in the anterior condition (0.35°, p < 0.05), with trivial effect sizes (ES = 0.08), indicating that these differences may not be clinically meaningful. The high Pearson's correlation coefficients across conditions underscore a robust linear relationship between devices (r > 0.9 for all conditions). Regression analysis showed a standard error of estimate (SEE) of 1.1° with small effect (standardized SEE < 0.26 for all conditions), meaning that the expected average deviation from the true value is around 1°. These findings validate the Ergotex as an effective, portable, and cost-efficient tool for assessing sagittal pelvic tilt, with practical implications in clinical and sports settings where traditional methods might be impractical or costly.

A Study of Polish Family with Scoliosis and Limb Contractures Expands the MYH3 Disease Spectrum

A disease associated with malfunction of the MYH3 gene is characterised by scoliosis, contractures of the V fingers, knees and elbows, dysplasia of the calf muscles, foot deformity and limb length asymmetry. The aim of this study was to identify the cause of musculoskeletal deformities in a three-generation Polish family by exome sequencing. The segregation of the newly described c.866A>C variant of the MYH3 gene in the family indicates an autosomal dominant model of inheritance. The detected MYH3 variant segregates the disease within the family. The presented results expand the MYH3 disease spectrum and emphasize the clinical diagnostic challenge in syndromes harbouring congenital spine defects and joint contractures.

Immediate Effects of Sprinter-Pattern Exercise on the Lordotic Curve and Abdominal Muscle Activity in Individuals with Hyperlordosis

Background and Objectives: Abdominal muscle exercises with limb movements are more effective for trunk stabilization than traditional exercises involving trunk flexion alone. This study examined the effects of abdominal exercises incorporating sprinter pattern and crunch exercises on changes in the lordotic curve and abdominal muscle activation in individuals with low back pain caused by hyperlordosis resulting from weak abdominal muscles. Materials and Methods: In this single-blind, randomized controlled trial, a total of 40 participants with hyperlordosis were recruited and randomly assigned to perform either sprinter-pattern abdominal exercises or crunch exercises. The participants assigned to each group performed three sets of ten abdominal exercises. The lumbar lordotic angle (LLA) and sacrohorizontal angle (SHA) were assessed prior to and following the intervention, whereas abdominal muscle activity was gauged throughout the intervention period. Changes in the LLA and SHA were measured by radiography. Abdominal muscle activity was measured using electromyography. Results: The LLA and SHA decreased significantly in both groups (p < 0.001), while the sprinter-pattern exercise group showed a statistically significant decrease compared to the crunch exercise group (p < 0.001). In the activity of the abdominal muscles, there was no significant difference in the rectus abdominis muscle between the two groups (p > 0.005). However, a significant difference between the external and internal oblique muscles was observed, and the activities of both muscles were significantly higher in the sprinter-pattern exercise group than in the crunch exercise group (p < 0.005). Conclusions: Abdominal exercise using a sprinter pattern may be effective in reducing lumbar lordosis by strengthening the abdominal muscles in patients with hyperlordosis.

Evaluation of lumbar lordosis in symptomatic individuals and comparative analysis of six different techniques: a retrospective radiologic study

PURPOSE

The aim of this study; evaluate lumbar lordosis (LL) in symptomatic individuals with six different techniques and to examine the techniques comparatively. Thus, to provide an overview of lumbal lordosis and techniques.

METHODS

Cobb L1-L5, Cobb L1-S1, Posterior Tangent, tangential radiologic assessment of lumbar lordosis (TRALL), vertebral centroid measurement of lumbar lordosis (CLL) and Risser Ferguson measurement techniques were used to assess LL from radiographs of 175 symptomatic adults. Correlations between techniques and relationship between the measurements obtained, gender and age were analyzed. Also ınterclass correlation (ICC) analyzed. Bland-Altman plots were performed to compare the techniques with Cobb.

RESULTS

ICC for all methods were greater than 0.96. For each method, no difference in LL was observed with respect to gender or age (p > 0.05). High positive correlation was observed between the Risser Ferguson, Posterior Tangent, Cobb L1-L5, Cobb L1-S1 and CLL techniques (p < 0.001), and moderate positive correlation between TRALL and all other techniques (p < 0.001).

CONCLUSION

In this study, it was found that the mean lumbar lordosis values of symptomatic participants were lower than most of the other asymptomatic studies in the literature and there was no significant difference in lumbar lordosis values in terms of gender and age in symptomatic individuals. Based on statistical findings, Risser Ferguson can be used to assess LL. These results and the data obtained as a result of the comparative examination of techniques according to age groups and gender will benefit clinicians and those working in the field by providing a better understanding LL.

Applying Machine Learning and Point-Set Registration to Automatically Measure the Severity of Spinal Curvature on Radiographs

OBJECTIVE

Measuring the severity of the lateral spinal curvature, or Cobb angle, is critical for monitoring and making treatment decisions for children with adolescent idiopathic scoliosis (AIS). However, manual measurement is time-consuming and subject to human error. Therefore, clinicians seek an automated measurement method to streamline workflow and improve accuracy. This paper reports on a novel machine learning algorithm of cascaded convolutional neural networks (CNN) to measure the Cobb angle on spinal radiographs automatically.

METHODS

The developed method consisted of spinal column segmentation using a CNN, vertebra localization and segmentation using iterative vertebra body location coupled with another CNN, point-set registration to correct vertebra segmentations, and Cobb angle measurement using the final segmentations. Measurement performance was evaluated with the circular mean absolute error (CMAE) and percentage within clinical acceptance ([Formula: see text]) between automatic and manual measurements. Analysis was separated by curve severity to identify any potential systematic biases using independent samples Student's t-tests.

RESULTS

The method detected 346 of the 352 manually measured Cobb angles (98%), with a CMAE of 2.8° and 91% of measurements within the 5° clinical acceptance. No statistically significant differences were found between the CMAEs of mild ([Formula: see text]), moderate (25°-45°), and severe ([Formula: see text]) groups. The average measurement time per radiograph was 17.7±10.2s, improving upon the estimated average of 30s it takes an experienced rater to measure. Additionally, the algorithm outputs segmentations with the measurement, allowing clinicians to interpret measurement results.

DISCUSSION/CONCLUSION

The developed method measured Cobb angles on radiographs automatically with high accuracy, quick measurement time, and interpretability, suggesting clinical feasibility.

Automatic measurement of the Cobb angle for adolescent idiopathic scoliosis using convolutional neural network

This study proposes a convolutional neural network method for automatic vertebrae detection and Cobb angle (CA) measurement on X-ray images for scoliosis. 1021 full-length X-ray images of the whole spine of patients with adolescent idiopathic scoliosis (AIS) were used for training and segmentation. The proposed AI algorithm's results were compared with those of the manual method by six doctors using the intraclass correlation coefficient (ICC). The ICCs recorded by six doctors and AI were excellent or good, with a value of 0.973 for the major curve in the standing position. The mean error between AI and doctors was not affected by the angle size, with AI tending to measure 1.7°-2.2° smaller than that measured by the doctors. The proposed method showed a high correlation with the doctors' measurements, regardless of the CA size, doctors' experience, and patient posture. The proposed method showed excellent reliability, indicating that it is a promising automated method for measuring CA in patients with AIS.

Evaluation of the accuracy of new modalities in the assessment and classification of lumbar lordosis: A comparison to Cobb's angle measurement

Background

Because of the association of lumbar lordosis with some clinical conditions such as low back pain, the chiropractic field has emphasized the significance of evaluating the lumbar lordotic status, by measuring Cobb's angle, regarded as the radiological gold standard, for the assessment of lumbar lordosis, on lateral radiographs. However, research has shown that this technique has some considerable drawbacks, mostly in terms of low accuracy and high variability between clinicians when compared with other radiological modalities. The main objective was to compare the diagnostic accuracy of newly established radiological measurements with one of Cobb's angle methods, for the characterization of lumbar lordosis status in a sample of Lebanese patients aged 15 and above.

Material and methods

This retrospective single-center study consisted of measuring Cobb's L1-S1 and Cobb's L1-L5 angles, along with the novel established measurements which are the derivative and the normalized surface area, on 134 lateral radiographs of the lumbar spine of Lebanese patients aged fifteen years old and above, gotten from the Radiology department at Zahra'a's Hospital in Beirut, performed by two observers using MATLAB. Inter-rater agreement was assessed by calculating the Intra-class correlation coefficients. Spearman correlation was analyzed between both Cobb's angle methods and with the derivative and normalized area respectively. 54 patients of the sample were diagnosed by two radiologists, according to their LL status. ROC curve analysis was performed to compare the diagnostic accuracy of the four techniques used. Data were analyzed with IBM SPSS Statistics 23.0 (NY, USA); P < 0.05 was considered statistically significant.

Results

According to the ROC curve analysis the new methods, which are the derivative and the normalized surface area, displayed lower diagnostic accuracy (AUCderivative = 0.818 and 0.677, AUCsurface area = 0.796 and 0.828) than Cobb's L1-L5 (AUCL1-L5 = 0.924 and 0.929 values) and L1-S1 (AUCL1-S1 = 0.971 and 0.955) angles, in the characterization of hypo and hyperlordotic patients, respectively, in our Lebanese sample consisting of patients aged 15 and above, because of their lower area under the curve's values compared to the traditional Cobb's techniques. The Cobb's L1-S1 has shown to have the highest diagnostic accuracy among the four methods to characterize normal patients from hypo and hyperlordotic ones, by referring to its highest area under the curve's values. However, the sensitivity of Cobb's L1-L5 angle in characterizing hyperlordotic patients was similar to the one of the normalized surface area with a value of 100%.Conclusion: among the four modalities, the new methods didn't show a better diagnostic accuracy compared to the traditional modalities. Cobb's L1-S1 displayed the highest diagnostic accuracy despite its drawbacks. Further prospective studies are needed to validate the cut-offs obtained for Cobb's L1-S1 angle in our sample.

Automatic Spine Segmentation and Parameter Measurement for Radiological Analysis of Whole-Spine Lateral Radiographs Using Deep Learning and Computer Vision

Radiographic examination is essential for diagnosing spinal disorders, and the measurement of spino-pelvic parameters provides important information for the diagnosis and treatment planning of spinal sagittal deformities. While manual measurement methods are the golden standard for measuring parameters, they can be time consuming, inefficient, and rater dependent. Previous studies that have used automatic measurement methods to alleviate the downsides of manual measurements showed low accuracy or could not be applied to general films. We propose a pipeline for automated measurement of spinal parameters by combining a Mask R-CNN model for spine segmentation with computer vision algorithms. This pipeline can be incorporated into clinical workflows to provide clinical utility in diagnosis and treatment planning. A total of 1807 lateral radiographs were used for the training (n = 1607) and validation (n = 200) of the spine segmentation model. An additional 200 radiographs, which were also used for validation, were examined by three surgeons to evaluate the performance of the pipeline. Parameters automatically measured by the algorithm in the test set were statistically compared to parameters measured manually by the three surgeons. The Mask R-CNN model achieved an average precision at 50% intersection over union (AP50) of 96.2% and a Dice score of 92.6% for the spine segmentation task in the test set. The mean absolute error values of the spino-pelvic parameters measurement results were within the range of 0.4° (pelvic tilt) to 3.0° (lumbar lordosis, pelvic incidence), and the standard error of estimate was within the range of 0.5° (pelvic tilt) to 4.0° (pelvic incidence). The intraclass correlation coefficient values ranged from 0.86 (sacral slope) to 0.99 (pelvic tilt, sagittal vertical axis).

Validation of an artificial intelligence-based method to automate Cobb angle measurement on spinal radiographs of children with adolescent idiopathic scoliosis

BACKGROUND

Accurately measuring the Cobb angle on radiographs is crucial for diagnosis and treatment decisions for adolescent idiopathic scoliosis (AIS). However, manual Cobb angle measurement is time-consuming and subject to measurement variation, especially for inexperienced clinicians.

AIM

This study aimed to validate a novel artificial-intelligence-based (AI) algorithm that automatically measures the Cobb angle on radiographs.

DESIGN

This is a retrospective cross-sectional study.

SETTING

The population of patients attended the Stollery Children's Hospital in Alberta, Canada.

POPULATION

Children who: 1) were diagnosed with AIS, 2) were aged between 10 and 18 years old, 3) had no prior surgery, and 4) had a radiograph out of brace, were enrolled.

METHODS

A total of 330 spinal radiographs were used. Among those, 130 were used for AI model development and 200 were used for measurement validation. Automatic Cobb angle measurements were validated by comparing them with manual ones measured by a rater with 20+ years of experience. Analysis was performed using the standard error of measurement (SEM), inter-method intraclass correlation coefficient (ICC2,1), and percentage of measurements within clinical acceptance (≤5°). Subgroup analysis was conducted by severity, region, and X-ray system to identify any systematic biases.

RESULTS

The AI method detected 346 of 352 manually measured curves (mean±standard deviation: 24.7±9.5°), achieving 91% (316/346) of measurements within clinical acceptance. Excellent reliability was obtained with 0.92 ICC and 0.79° SEM. Comparable performance was found throughout all subgroups, and no systematic biases in performance affecting any subgroup were discovered. The algorithm measured each radiograph approximately 18s on average which is slightly faster than the estimated measurement time of an experienced rater. Radiographs taken by the EOS X-ray system were measured more quickly on average than those taken by a conventional digital X-ray system (10s vs. 26s).

CONCLUSIONS

An AI-based algorithm was developed to measure the Cobb angle automatically on radiographs and yielded reliable measurements quickly. The algorithm provides detailed images on how the angles were measured, providing interpretability that can give clinicians confidence in the measurements.

CLINICAL REHABILITATION IMPACT

Employing the algorithm in practice could streamline clinical workflow and optimize measurement accuracy and speed in order to inform AIS treatment decisions.

Integrating biplane information and context for spine landmark detection

Spine landmark detection is of great significance for spinal morphological parameter assessment and three-dimensional reconstruction of the human spine. This detection task generally involves locating spine landmarks in the anterior-posterior (AP) and lateral (LAT) X-rays of the spine. Recently, the two-stage methods for AP spine landmark detection achieve better performance. However, these methods perform poorly in LAT landmark detection because of poor detection accuracy of LAT vertebra due to occlusion. To solve this problem, this paper proposes a new two-stage spine landmark detection method. In the first stage, this paper propose a biplane vertebra detection network for vertebra detection on AP and X-rays simultaneously. Then an epipolar module and a context enhancement module are proposed to assist LAT vertebra detection by using the biplane information and the context information of the vertebrae respectively. In the second stage, the landmarks can be obtained in the detected vertebrae area. Extensive experiment results conducted on a dataset containing 328 pairs of X-rays demonstrate that our method improves the vertebra and landmark detection accuracy.

Creation and Evaluation of a Severity Classification of Hyperkyphosis and Hypolordosis for Exercise Therapy

The rise in the occurrence of musculoskeletal disorders, such as thoracic hyperkyphosis (THK) or lumbar hypolordosis (LHL), is a result of demographic changes. Exercise therapy is an effective approach that can reduce related disabilities and costs. To ensure successful therapy, an individualized exercise program adapted to the severity of the disorder is expedient. Nevertheless, appropriate classification systems are scarce. This project aimed to develop and evaluate a severity classification focused on exercise therapy for patients with THK or LHL. A multilevel severity classification was developed and evaluated by means of an online survey. Reference limits of spinal shape angles were established by data from video rasterstereography of 201 healthy participants. A mean kyphosis angle of 50.03° and an average lordosis angle of 40.72° were calculated as healthy references. The strength of the multilevel classification consisting of the combination of subjective pain and objective spinal shape factors was confirmed by the survey (70% agreement). In particular, the included pain parameters were considered relevant by 78% of the experts. Even though the results of the survey provide important evidence for further analyses and optimization options of the classification system, the current version is still acceptable as therapeutic support.

Effects of 3D Postural Correction and Abdominal Muscle Contraction on the Symmetry of the Transverse Abdominis and Spinal Alignment in Patients with Idiopathic Scoliosis

This study aimed to investigate the effectiveness of 3D postural correction (3DPC) using corrective cushions (CCs) and abdominal muscle contraction (AMC) on the thickness symmetry of the transversus abdominis (TrA) and spinal alignment in patients with idiopathic scoliosis (IS). In the first experiment, ultrasound measurements were taken of the TrA thickness on both the convex and concave sides of the lumbar curve in the supine position during AMC and non-AMC without 3DPC, and during AMC and non-AMC with 3DPC using CCs, in 11 IS patients. In the second experiment, 37 IS patients participated in a four-week 3DPC exercise program that aimed to maintain TrA thickness symmetry based on the results of the first experiment. The study found that TrA thickness symmetry significantly increased after 3DPC using CCs and combined with AMC (p < 0.05). Additionally, the Cobb angles and trunk rotation angles showed significant decreases, and trunk expansion showed a significant increase (p < 0.05). These results indicate that the simultaneous application of 3DPC and AMC is the most effective way to achieve TrA thickness symmetry in IS patients. Therefore, 3DPC and AMC should be considered as crucial elements in exercise interventions for IS patients.

[Development and validation of an automatic diagnostic tool for lumbar stability based on deep learning]

Objective

To develop an automatic diagnostic tool based on deep learning for lumbar spine stability and validate diagnostic accuracy.

Methods

Preoperative lumbar hyper-flexion and hyper-extension X-ray films were collected from 153 patients with lumbar disease. The following 5 key points were marked by 3 orthopedic surgeons: L₄ posteroinferior, anterior inferior angles as well as L₅ posterosuperior, anterior superior, and posterior inferior angles. The labeling results of each surgeon were preserved independently, and a total of three sets of labeling results were obtained. A total of 306 lumbar X-ray films were randomly divided into training (n=156), validation (n=50), and test (n=100) sets in a ratio of 3∶1∶2. A new neural network architecture, Swin-PGNet was proposed, which was trained using annotated radiograph images to automatically locate the lumbar vertebral key points and calculate L_{4, 5} intervertebral Cobb angle and L₄ lumbar sliding distance through the predicted key points. The mean error and intra-class correlation coefficient (ICC) were used as an evaluation index, to compare the differences between surgeons' annotations and Swin-PGNet on the three tasks (key point positioning, Cobb angle measurement, and lumbar sliding distance measurement). Meanwhile, the change of Cobb angle more than 11° was taken as the criterion of lumbar instability, and the lumbar sliding distance more than 3 mm was taken as the criterion of lumbar spondylolisthesis. The accuracy of surgeon annotation and Swin-PGNet in judging lumbar instability was compared.

Results

① Key point: The mean error of key point location by Swin-PGNet was (1.407±0.939) mm, and by different surgeons was (3.034±2.612) mm. ② Cobb angle: The mean error of Swin-PGNet was (2.062±1.352)° and the mean error of surgeons was (3.580±2.338)°. There was no significant difference between Swin-PGNet and surgeons (P>0.05), but there was a significant difference between different surgeons (P<0.05). ③ Lumbar sliding distance: The mean error of Swin-PGNet was (1.656±0.878) mm and the mean error of surgeons was (1.884±1.612) mm. There was no significant difference between Swin-PGNet and surgeons and between different surgeons (P>0.05). The accuracy of lumbar instability diagnosed by surgeons and Swin-PGNet was 75.3% and 84.0%, respectively. The accuracy of lumbar spondylolisthesis diagnosed by surgeons and Swin-PGNet was 70.7% and 71.3%, respectively. There was no significant difference between Swin-PGNet and surgeons, as well as between different surgeons (P>0.05). ④ Consistency of lumbar stability diagnosis: The ICC of Cobb angle among different surgeons was 0.913 [95%CI (0.898, 0.934)] (P<0.05), and the ICC of lumbar sliding distance was 0.741 [95%CI (0.729, 0.796)] (P<0.05). The result showed that the annotating of the three surgeons were consistent. The ICC of Cobb angle between Swin-PGNet and surgeons was 0.922 [95%CI (0.891, 0.938)] (P<0.05), and the ICC of lumbar sliding distance was 0.748 [95%CI(0.726, 0.783)] (P<0.05). The result showed that the annotating of Swin-PGNet were consistent with those of surgeons.

Conclusion

The automatic diagnostic tool for lumbar instability constructed based on deep learning can realize the automatic identification of lumbar instability and spondylolisthesis accurately and conveniently, which can effectively assist clinical diagnosis.

Assessment of spinal alignment in children with unilateral cerebral palsy

BACKGROUND

Children/youths with unilateral cerebral palsy are at high risk for the development of scoliosis and other postural deformities. The purpose of this study was to perform spinal assessment in the frontal and sagittal plane using Spinal Mouse® in children/youths with unilateral cerebral palsy and to compare their spinal shape and angles with typically developing children/youths.

METHODS

25 children/youths with unilateral cerebral palsy and 25 typical children/youths, aged 6-18 years, were included. The subject's frontal (scoliosis) and sagittal plane (kyphosis and lordosis) spinal curvatures were compared by assessing them with Spinal Mouse®.

FINDINGS

Scoliosis was detected in 40% of subjects in the unilateral cerebral palsy group and this rate was considerably higher than that in typical subjects (12%). The median angle of scoliosis was 8° in subjects with unilateral cerebral palsy and 5.3° in typical subjects. While the median angle of scoliosis was higher in subjects with unilateral cerebral palsy than typical subjects (p < 0.001), there was no significant difference in the angles of lordosis and kyphosis between both groups (p > 0.05). Curvature patterns of subjects with unilateral cerebral palsy differed from typical subjects.

INTERPRETATION

Our findings will allow children/youths with unilateral cerebral palsy, who are at risk of developing spinal deformity, to be identified earliest possible and included in the intervention. Children/youths with unilateral cerebral palsy have to be assessed in detail from the earliest period, especially when the possibility of an age-related increase in scoliosis is considered.

Validity of sagittal thoracolumbar curvature measurement using a non-radiographic surface topography method

PURPOSE

To estimate the criterion validity of sagittal thoracolumbar spine measurement using a surface topography method in a clinical population against the gold standard and to estimate concurrent validity against two non-radiographic clinical tools.

METHODS

In this cross-sectional validity study, thoracolumbar curvature was measured in adults with spinal conditions recruited from a specialist orthopaedic hospital. A surface topography method using a Kinect sensor was compared to three other measurement methods: spinal radiograph (gold standard), flexicurve and digital inclinometer. Correlation coefficients and agreement between the measurement tools were analysed.

RESULTS

Twenty-nine participants (79% female) were included in criterion validity analyses and 38 (76% female) in concurrent validity analyses. The surface topography method was moderately correlated with the radiograph (r = .70, p < .001) in the thoracic spine, yet there was no significant correlation with the radiograph in the lumbar spine (r = .32, p = .89). The surface topography method was highly correlated with the flexicurve (rs = .91, p < .001) and digital inclinometer (r = .82, p < .001) in the thoracic spine, and highly correlated with the flexicurve (r = .74, p < .001) and digital inclinometer (r = .74, p < .001) in the lumbar spine.

CONCLUSIONS

The surface topography method showed moderate correlation and agreement in thoracic spine with the radiograph (criterion validity) and high correlation with the flexicurve and digital inclinometer (concurrent validity). Compared with other non-radiographic tools, this surface topography method displayed similar criterion validity for kyphosis curvature measurement.

The efficacy of trunk bracing with an instrumented corrective exercise on spinal deformity, pulmonary function, trunk muscle endurance and quality of life in adolescent idiopathic scoliosis: Protocol for a parallel-groups clinical study (Preprint)

BACKGROUND

Adolescent idiopathic scoliosis is a 3D spine distortion with an unidentified etiology. It results in noticeable trunk deformity, decreased muscle strength and endurance at the trunk, changes in chest volume, breathing issues, and ultimately a decline in the quality of life. Trunk bracing and corrective exercises make up most of the treatment of patients with scoliosis when their deformity is between 20° and 45°, and they have not yet attained skeletal maturity. Evidence suggests that spinal deformity in people with scoliosis may result from improper motor control. Automatic response training is an exercise therapy technique that can modify the pattern of trunk muscle control for supporting the spinal column in normal alignment. An apparatus called a cantilever device is required for this type of exercise, which facilitates training at home. In spite of research showing the benefit of braces and therapeutic exercise in adolescents with scoliosis, less emphasis has been given to the impact of home-based training, especially when this intervention is paired with braces.

OBJECTIVE

We aim to compare the efficacy of bracing and a conventional exercise program to a combination treatment that includes trunk bracing and exercises with a cantilever device performed at home on the degree of spine curvature, pulmonary function, trunk muscular endurance, and quality of life.

METHODS

This study was a 2-arms parallel-group clinical study. A total of 16 adolescents with idiopathic scoliosis and single lumbar and thoracolumbar curves of 20°-45° were recruited and randomly assigned into 2 groups. Group A received a combination of trunk bracing and exercise using an instrument known as a "cantilever." Group B (controls) received trunk bracing and a conventional exercise program (without a tool). The study outcomes were the Cobb angle of the scoliotic curve, pulmonary function, the endurance of the trunk muscles, and quality of life. The study outcomes were measured at 2 time points: before the intervention (T1) and 12 weeks following the start of the intervention (T2; at this time, the intervention period has been completed). Multivariate analysis of variance was used to test between- and within-group differences.

RESULTS

Recruitment for this study began in fall 2022 and is expected to be completed by the end of summer 2023.

CONCLUSIONS

We studied the efficacy of a combined trunk bracing program and postural response exercises using a cantilever device in treating adolescent idiopathic scoliosis and compared it with trunk bracing and conventional home exercises. Exercises performed at home using a cantilever device are anticipated to raise the endurance of trunk muscles, which will help reduce trunk deformity, enhance pulmonary function, and improve the quality of life of participants.

TRIAL REGISTRATION

Iranian Registry of Clinical Trials IRCT20220330054371N1; https://www.irct.ir/trial/62811.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

PRR1-10.2196/43265.

Age-of-cessation of lumbar lordosis development as an assessment parameter

BACKGROUND

In managing paediatric spinal deformities, the currently-in-use growth maturity assessment parameters (clinical or radiological) are based mostly on Caucasian populations. They may be adequate for general treatment planning but may not accurately predict the remaining growth potential. Some therapies (e.g. growing rod distractions or growth modulation surgeries) require more accurate predictions of remaining growth potential and race-specific values. Lumbar lordosis (LL) development ceases at spinal bone maturity. The age-of-cessation seems a more accurate predictor of remaining spinal bone growth potential, compared to currently-in-use growth maturity assessment parameters, but is rarely included in the growth maturity assessment parameters.

AIMS AND OBJECTIVES

As a predictor of remaining spinal growth potential, age-of-cessation of LL development (Race-specific of Black populations) was quantified.

MATERIALS AND METHODS

In archival normal lateral lumbosacral radiographs of patients of a tertiary hospital in South-East Nigeria, LL development across five age groups (Birth- 9, 10-15, 16-20, 21-25 and 26-30 years) was quantified with lumbosacral joint angle (LSJA) in 215 (110 males, 105 females), and lumbosacral angle (LSA) in 238 (119 males, 119 females). Data were analysed with IBM SPSS Statistics 23.0 (NY, USA). P ≤ 0.05 was considered statistically significant.

RESULTS

Both LSJA and LSA age groups' mean values progressively increased with age, and plateaued at 21-25 years range, with LSJA mean of 23.4 ± 1.3 years, and LSA mean 23.5 ± 1.3 years; the means difference was insignificant (P = 0.680).

CONCLUSION

With ageing, there is progressive increment, and later, cessation of LL. Age-of-cessation indirectly infers spinal-maturity-age, and could indirectly be an assessment parameter of spinal-maturity-status.

Semi-automatic method for pre-surgery scoliosis classification on X-ray images using Bending Asymmetry Index

PURPOSE

Bending Asymmetry Index (BAI) has been proposed to characterize the types of scoliotic curve in three-dimensional ultrasound imaging. Scolioscan has demonstrated its validity and reliability in scoliosis assessment with manual assessment-based X-ray imaging. The objective of this study is to investigate the ultrasound-derived BAI method to X-ray imaging of scoliosis, with supplementary information provided for the pre-surgery planning.

METHODS

About 30 pre-surgery scoliosis subjects (9 males and 21 females; Cobb: 50.9 ± 19.7°, range 18°-115°) were investigated retrospectively. Each subject underwent three-posture X-ray scanning supine on a plain mattress on the same day. BAI is an indicator to distinguish structural or non-structural curves through the spine flexibility information obtained from lateral bending spinal profiles. BAI was calculated semi-automatically with manual annotation of vertebral centroids and pelvis level inclination adjustment. BAI classification was validated with the scoliotic curve type and traditional Lenke classification using side-bending Cobb angle measurement (S-Cobb).

RESULTS

82 curves from 30 pre-surgery scoliosis patients were included. The correlation coefficient was R² = 0.730 (p < 0.05) between BAI and S-Cobb. In terms of scoliotic curve type classification, all curves were correctly classified; out of 30 subjects, 1 case was confirmed as misclassified when applying to Lenke classification earlier, thus has been adjusted.

CONCLUSION

BAI method has demonstrated its inter-modality versatility in X-ray imaging application. The curve type classification and the pre-surgery Lenke classification both indicated promising performances upon the exploratory dataset. A fully-automated of BAI measurement is surely an interesting direction to continue our endeavor. Deep learning on the vertebral-level segmentation should be involved in further study.

A comparison of the reliability and vulnerability of 3D sterEOS and 2D EOS when measuring the sagittal spinal alignment of patients with adolescent idiopathic scoliosis

PURPOSE

An essential component of making the diagnosis of adolescent idiopathic scoliosis (AIS) is standing anteroposterior and lateral radiographs. Two-dimensional (2D) radiographs inevitably fail to reflect every plane of the three-dimensional (3D) deformity in scoliosis. We have tested the hypothesis that there is no difference in the assessment of the sagittal plane deformity when measured with either 2D or 3D EOS radiography.

METHODS

A retrospective radiographic analysis was performed on patients diagnosed with AIS, with subdivided into three groups according to the coronal angular deformity (mild group: 45°-69°, moderate group: 70°-89°, and severe group: 90° +). The sagittal parameters were compared between manual measurement with 2D sterEOS and those made using computer-aided 3D reconstruction.

RESULTS

Fifty-two patients were included in each group. The inter-study reliability when measuring the thoracic Kyphosis (TK) and lumbar lordosis (LL) between the two study modalities was excellent in mild group (ICC: 0.90, 95% CI 0.82 ~ 0.94 and ICC: 0.84, 95% CI 0.74 ~ 0.91), excellent in TK and fair in LL in moderate group (ICC: 0.76, 95% CI 0.61 ~ 0.85 and ICC: 0.70, 95% CI 0.53 ~ 0.81), and fair in TK and LL in severe group, respectively (ICC: 0.74, 95% CI 0.57 ~ 0.84 and ICC: 0.65, 95% CI 0.46 ~ 0.84). A Bland-Altman plot showed proportional bias in TK measurements in each group and LL in moderate group, which means the measured value is underestimated in 2D method when the angle is small.

CONCLUSION

3D sterEOS is less vulnerable to the influence of coronal plane than 2D EOS in evaluating the sagittal spinal parameters of patients with a coronal deformity exceeding 70°.

Clinical Validation of the Differences Between Two-Dimensional Radiography and Three-Dimensional Computed Tomography Image Measurements of the Spine in Adolescent Idiopathic Scoliosis

OBJECTIVE

The purpose of this study was to investigate differences between measurements of spine on two-dimensional (2D) radiography and three-dimensional (3D) computed tomography (CT) images taken of patients with adolescent idiopathic scoliosis.

METHODS

Standard preoperative CT images and posteroanterior (PA) and lateral radiography images were collected prospectively from 43 patients with adolescent idiopathic scoliosis in whom selective spinal fusions were performed. The parameters of interest were the thoracic Cobb angle, lumbar Cobb angle, T4-T12 kyphosis angle, and L1-S1 lordosis (LL) angle. The parameters were measured using 3 separate methods: 3D measurement of CT images (3D measurement), 2D measurement of radiography images (2D measurement), and 2D measurement of radiography images generated by the projection of CT images (2D XP measurement). Significant differences among the results were assessed by comparison T test.

RESULTS

The mean difference between the 2D and 2D XP measurements for the thoracic Cobb, lumbar Cobb, T4-T12 kyphotic, and L1-S1 lordotic angles was 8.38°, 7.67°, 8.77°, and 10.18°, respectively. The mean difference between the 2D XP and 3D measurements was -2.81°, -2.78°, -1.29°, and -2.36°, respectively. The mean difference between the 2D and 3D measurements was 5.16°, 4.51°, 6.49°, and 7.37°, respectively. The results showed significant differences (P < 0.05) among the spinal parameters measured using the 2D, 2D XP, and 3D measurement methods on both the coronal and sagittal plane.

CONCLUSIONS

Significant differences among the 2D, 2D XP, and 3D measurement methods were observed on both the sagittal plane and coronal plane of the scoliotic spines as a result of variations in posture during imaging and differences in measurement methods.

A deep learning framework for vertebral morphometry and Cobb angle measurement with external validation

PURPOSE

To propose a fully automated deep learning (DL) framework for the vertebral morphometry and Cobb angle measurement from three-dimensional (3D) computed tomography (CT) images of the spine, and validate the proposed framework on an external database.

METHODS

The vertebrae were first localized and segmented in each 3D CT image using a DL architecture based on an ensemble of U-Nets, and then automated vertebral morphometry in the form of vertebral body (VB) and intervertebral disk (IVD) heights, and spinal curvature measurements in the form of coronal and sagittal Cobb angles (thoracic kyphosis and lumbar lordosis) were performed using dedicated machine learning techniques. The framework was trained on 1725 vertebrae from 160 CT images and validated on an external database of 157 vertebrae from 15 CT images.

RESULTS

The resulting mean absolute errors (± standard deviation) between the obtained DL and corresponding manual measurements were 1.17 ± 0.40 mm for VB heights, 0.54 ± 0.21 mm for IVD heights, and 3.42 ± 1.36° for coronal and sagittal Cobb angles, with respective maximal absolute errors of 2.51 mm, 1.64 mm, and 5.52°. Linear regression revealed excellent agreement, with Pearson's correlation coefficient of 0.943, 0.928, and 0.996, respectively.

CONCLUSION

The obtained results are within the range of values, obtained by existing DL approaches without external validation. The results therefore confirm the scalability of the proposed DL framework from the perspective of application to external data, and time and computational resource consumption required for framework training.

Automatic recognition of whole-spine sagittal alignment and curvature analysis through a deep learning technique

PURPOSE

Artificial intelligence based on deep learning (DL) approaches enables the automatic recognition of anatomic landmarks and subsequent estimation of various spinopelvic parameters. The locations of inflection points (IPs) and apices (APs) in whole-spine lateral radiographs could be mathematically determined by a fully automatic spinal sagittal curvature analysis system.

METHODS

We developed a DL model for automatic spinal curvature analysis of whole-spine lateral plain radiographs by using 1800 annotated images of various spinal disease etiologies. The DL model comprised a landmark localizer to detect 25 vertebral landmarks and a numerical algorithm for the generation of an individualized spinal sagittal curvature. The characteristics of the spinal curvature, including the IPs, APs, and curvature angle, could thus be analyzed using mathematical definitions. The localization error of each landmark was calculated from the predictions of 300 test images to evaluate the performance of the landmark localizer. The interrater reliability among a senior orthopedic surgeon, a radiologist, and the DL model was assessed using the intraclass correlation coefficient (ICC).

RESULTS

The accuracy of the landmark localizer was within an acceptable range (median error: 1.7-4.1 mm), and the interrater reliabilities between the proposed DL model and each expert were good to excellent (all ICCs > 0.85) for the measurement of spinal curvature characteristics.

CONCLUSION

The interrater reliability between the proposed DL model and human experts was good to excellent in predicting the locations of IPs, APs, and curvature angles. Future applications should be explored to validate this system and improve its clinical efficiency.

Spinopelvic measurements of sagittal balance with deep learning: systematic review and critical evaluation

PURPOSE

To summarize and critically evaluate the existing studies for spinopelvic measurements of sagittal balance that are based on deep learning (DL).

METHODS

Three databases (PubMed, WoS and Scopus) were queried for records using keywords related to DL and measurement of sagittal balance. After screening the resulting 529 records that were augmented with specific web search, 34 studies published between 2017 and 2022 were included in the final review, and evaluated from the perspective of the observed sagittal spinopelvic parameters, properties of spine image datasets, applied DL methodology and resulting measurement performance.

RESULTS

Studies reported DL measurement of up to 18 different spinopelvic parameters, but the actual number depended on the image field of view. Image datasets were composed of lateral lumbar spine and whole spine X-rays, biplanar whole spine X-rays and lumbar spine magnetic resonance cross sections, and were increasing in size or enriched by augmentation techniques. Spinopelvic parameter measurement was approached either by landmark detection or structure segmentation, and U-Net was the most frequently applied DL architecture. The latest DL methods achieved excellent performance in terms of mean absolute error against reference manual measurements (~ 2° or ~ 1 mm).

CONCLUSION

Although the application of relatively complex DL architectures resulted in an improved measurement accuracy of sagittal spinopelvic parameters, future methods should focus on multi-institution and multi-observer analyses as well as uncertainty estimation and error handling implementations for integration into the clinical workflow. Further advances will enhance the predictive analytics of DL methods for spinopelvic parameter measurement.

LEVEL OF EVIDENCE I

Diagnostic: individual cross-sectional studies with the consistently applied reference standard and blinding.

Magnetically controlled growing rods in the management of early onset scoliosis: a systematic review

Background

Early onset scoliosis (EOS) presents in patients younger than 10 years. Magnetically controlled growing rods (MCGR) were developed as an outpatient distraction system for EOS, allowing to avoid multiple surgeries. This systematic review investigated the efficacy and feasibility of MCGR in EOS.

Methods

This systematic review was conducted according to the PRISMA guidelines. PubMed, Google scholar, Embase, and Scopus were accessed in May 2022. All the clinical trials which investigate the role of MCGR for early onset scoliosis were accessed. Only studies reporting data in patients younger than 10 years with a preoperative Cobb Angle greater than 40° were eligible. The following data was extracted at baseline and at last follow-up: mean kyphosis angle, overall mean Cobb angle, mean T1–S1 length. Data from complication were also collected.

Results

Data from 23 clinical studies (504 patients) were included in the present study. 56% (282 of 504) were females. The average length of the follow-up was 28.9 ± 16.0 months. The mean age of the patients was 8.7 ± 1.9 years old. The mean BMI was 17.7 ± 7.6 kg/m². The mean kyphosis angle had reduced by the last follow-up (P = 0.04), as did the overall mean Cobb angle (P < 0.0001), while the overall T1–S1 length increased (P = 0.0002). Implant-associated complications, followed by spinal alignment failure, wound healing ailments, pulmonary complications, progressive trunk stiffness, persistent back pain, and fracture.

Conclusion

The management of EOS remains challenging. The current evidence indicates that MCGR may be effective to distract the spine and model the curve in EOS.