Deep Learning Automates Measurement of Spinopelvic Parameters on Lateral Lumbar Radiographs

Anatomical landmark detection on bi-planar radiographs for predicting spinopelvic parameters

INTRODUCTION

Accurate landmark detection is essential for precise analysis of anatomical structures, supporting diagnosis, treatment planning, and monitoring in patients with spinal deformities. Conventional methods rely on laborious landmark identification by medical experts, which motivates automation. The proposed deep learning pipeline processes bi-planar radiographs to determine spinopelvic parameters and Cobb angles without manual supervision.

METHODS

The dataset used for training and evaluation consisted of 555 bi-planar radiographs from un-instrumented patients, which were manually annotated by medical professionals. The pipeline performed a pre-processing step to determine regions of interest, including the cervical spine, thoracolumbar spine, sacrum, and pelvis. For each ROI, a segmentation network was trained to identify vertebral bodies and pelvic landmarks. The U-Net architecture was trained on 455 bi-planar radiographs using binary cross-entropy loss. The post-processing algorithm determined spinal alignment and angular parameters based on the segmentation output. We evaluated the pipeline on a test set of 100 previously unseen bi-planar radiographs, using the mean absolute difference between annotated and predicted landmarks as the performance metric. The spinopelvic parameter predictions of the pipeline were compared to the measurements of two experienced medical professionals using intraclass correlation coefficient (ICC) and mean absolute deviation (MAD).

RESULTS

The pipeline was able to successfully predict the Cobb angles in 61% of all test cases and achieved mean absolute differences of 3.3° (3.6°) and averaged ICC of 0.88. For thoracic kyphosis, lumbar lordosis, sagittal vertical axis, sacral slope, pelvic tilt, and pelvic incidence, the pipeline produced reasonable outputs in 69%, 58%, 86%, 85%, 84%, and 84% of the cases. The MAD was 5.6° (7.8°), 4.7° (4.3°), 2.8 mm (3.0 mm), 4.5° (7.2°), 1.8° (1.8°), and 5.3° (7.7°), while the ICC was measured at 0.69, 0.82, 0.99, 0.61, 0.96, and 0.70, respectively.

CONCLUSION

Despite limitations in patients with severe pathologies and high BMI, the pipeline automatically predicted coronal and sagittal spinopelvic parameters, which has the potential to simplify clinical routines and large-scale retrospective data analysis.

Automated spinopelvic measurements on radiographs with artificial intelligence: a multi-reader study

PURPOSE

To develop an artificial intelligence (AI) algorithm for automated measurements of spinopelvic parameters on lateral radiographs and compare its performance to multiple experienced radiologists and surgeons.

METHODS

On lateral full-spine radiographs of 295 consecutive patients, a two-staged region-based convolutional neural network (R-CNN) was trained to detect anatomical landmarks and calculate thoracic kyphosis (TK), lumbar lordosis (LL), sacral slope (SS), and sagittal vertical axis (SVA). Performance was evaluated on 65 radiographs not used for training, which were measured independently by 6 readers (3 radiologists, 3 surgeons), and the median per measurement was set as the reference standard. Intraclass correlation coefficient (ICC), mean absolute error (MAE), and standard deviation (SD) were used for statistical analysis; while, ANOVA was used to search for significant differences between the AI and human readers.

RESULTS

Automatic measurements (AI) showed excellent correlation with the reference standard, with all ICCs within the range of the readers (TK: 0.92 [AI] vs. 0.85-0.96 [readers]; LL: 0.95 vs. 0.87-0.98; SS: 0.93 vs. 0.89-0.98; SVA: 1.00 vs. 0.99-1.00; all p < 0.001). Analysis of the MAE (± SD) revealed comparable results to the six readers (TK: 3.71° (± 4.24) [AI] v.s 1.86-5.88° (± 3.48-6.17) [readers]; LL: 4.53° ± 4.68 vs. 2.21-5.34° (± 2.60-7.38); SS: 4.56° (± 6.10) vs. 2.20-4.76° (± 3.15-7.37); SVA: 2.44 mm (± 3.93) vs. 1.22-2.79 mm (± 2.42-7.11)); while, ANOVA confirmed no significant difference between the errors of the AI and any human reader (all p > 0.05). Human reading time was on average 139 s per case (range: 86-231 s).

CONCLUSION

Our AI algorithm provides spinopelvic measurements accurate within the variability of experienced readers, but with the potential to save time and increase reproducibility.

Artificial intelligence in musculoskeletal imaging: realistic clinical applications in the next decade

This article will provide a perspective review of the most extensively investigated deep learning (DL) applications for musculoskeletal disease detection that have the best potential to translate into routine clinical practice over the next decade. Deep learning methods for detecting fractures, estimating pediatric bone age, calculating bone measurements such as lower extremity alignment and Cobb angle, and grading osteoarthritis on radiographs have been shown to have high diagnostic performance with many of these applications now commercially available for use in clinical practice. Many studies have also documented the feasibility of using DL methods for detecting joint pathology and characterizing bone tumors on magnetic resonance imaging (MRI). However, musculoskeletal disease detection on MRI is difficult as it requires multi-task, multi-class detection of complex abnormalities on multiple image slices with different tissue contrasts. The generalizability of DL methods for musculoskeletal disease detection on MRI is also challenging due to fluctuations in image quality caused by the wide variety of scanners and pulse sequences used in routine MRI protocols. The diagnostic performance of current DL methods for musculoskeletal disease detection must be further evaluated in well-designed prospective studies using large image datasets acquired at different institutions with different imaging parameters and imaging hardware before they can be fully implemented in clinical practice. Future studies must also investigate the true clinical benefits of current DL methods and determine whether they could enhance quality, reduce error rates, improve workflow, and decrease radiologist fatigue and burnout with all of this weighed against the costs.

Radiographic Findings Associated With Mild Hip Dysplasia in 3869 Patients Using a Deep Learning Measurement Tool

Background

Hip dysplasia is considered one of the leading etiologies contributing to hip degeneration and the eventual need for total hip arthroplasty (THA). We validated a deep learning (DL) algorithm to measure angles relevant to hip dysplasia and applied this algorithm to determine the prevalence of dysplasia in a large population based on incremental radiographic cutoffs.

Methods

Patients from the Osteoarthritis Initiative with anteroposterior pelvis radiographs and without previous THAs were included. A DL algorithm automated 3 angles associated with hip dysplasia: modified lateral center-edge angle (LCEA), Tönnis angle, and modified Sharp angle. The algorithm was validated against manual measurements, and all angles were measured in a cohort of 3869 patients (61.2 ± 9.2 years, 57.1% female). The percentile distributions and prevalence of dysplastic hips were analyzed using each angle.

Results

The algorithm had no significant difference (P > .05) in measurements (paired difference: 0.3°-0.7°) against readers and had excellent agreement for dysplasia classification (kappa = 0.78-0.88). In 140 minutes, 23,214 measurements were automated for 3869 patients. LCEA and Sharp angles were higher and the Tönnis angle was lower (P < .01) in females. The dysplastic hip prevalence varied from 2.5% to 20% utilizing the following cutoffs: 17.3°-25.5° (LCEA), 9.4°-15.6° (Tönnis), and 41.3°-45.9° (Sharp).

Conclusions

A DL algorithm was developed to measure and classify hips with mild hip dysplasia. The reported prevalence of dysplasia in a large patient cohort was dependent on both the measurement and threshold, with 12.4% of patients having dysplasia radiographic indices indicative of higher THA risk.

Artificial intelligence automatic measurement technology of lumbosacral radiographic parameters

Background

Currently, manual measurement of lumbosacral radiological parameters is time-consuming and laborious, and inevitably produces considerable variability. This study aimed to develop and evaluate a deep learning-based model for automatically measuring lumbosacral radiographic parameters on lateral lumbar radiographs.

Methods

We retrospectively collected 1,240 lateral lumbar radiographs to train the model. The included images were randomly divided into training, validation, and test sets in a ratio of approximately 8:1:1 for model training, fine-tuning, and performance evaluation, respectively. The parameters measured in this study were lumbar lordosis (LL), sacral horizontal angle (SHA), intervertebral space angle (ISA) at L4-L5 and L5-S1 segments, and the percentage of lumbar spondylolisthesis (PLS) at L4-L5 and L5-S1 segments. The model identified key points using image segmentation results and calculated measurements. The average results of key points annotated by the three spine surgeons were used as the reference standard. The model's performance was evaluated using the percentage of correct key points (PCK), intra-class correlation coefficient (ICC), Pearson correlation coefficient (r), mean absolute error (MAE), root mean square error (RMSE), and box plots.

Results

The model's mean differences from the reference standard for LL, SHA, ISA (L4-L5), ISA (L5-S1), PLS (L4-L5), and PLS (L5-S1) were 1.69°, 1.36°, 1.55°, 1.90°, 1.60%, and 2.43%, respectively. When compared with the reference standard, the measurements of the model had better correlation and consistency (LL, SHA, and ISA: ICC = 0.91-0.97, r = 0.91-0.96, MAE = 1.89-2.47, RMSE = 2.32-3.12; PLS: ICC = 0.90-0.92, r = 0.90-0.91, MAE = 1.95-2.93, RMSE = 2.52-3.70), and the differences between them were not statistically significant (p > 0.05).

Conclusion

The model developed in this study could correctly identify key vertebral points on lateral lumbar radiographs and automatically calculate lumbosacral radiographic parameters. The measurement results of the model had good consistency and reliability compared to manual measurements. With additional training and optimization, this technology holds promise for future measurements in clinical practice and analysis of large datasets.

Intra-Examiner Reliability and Validity of Sagittal Cervical Spine Mensuration Methods Using Deep Convolutional Neural Networks

Background: The biomechanical analysis of spine and postural misalignments is important for surgical and non-surgical treatment of spinal pain. We investigated the examiner reliability of sagittal cervical alignment variables compared to the reliability and concurrent validity of computer vision algorithms used in the PostureRay® software 2024. Methods: A retrospective database of 254 lateral cervical radiographs of patients between the ages of 11 and 86 is studied. The radiographs include clearly visualized C1-C7 vertebrae that were evaluated by a human using the software. To evaluate examiner reliability and the concurrent validity of the trained CNN performance, two blinded trials of radiographic digitization were performed by an extensively trained expert user (US) clinician with a two-week interval between trials. Then, the same clinician used the trained CNN twice to reproduce the same measures within a 2-week interval on the same 254 radiographs. Measured variables included segmental angles as relative rotation angles (RRA) C1-C7, Cobb angles C2-C7, relative segmental translations (RT) C1-C7, anterior translation C2-C7, and absolute rotation angle (ARA) C2-C7. Data were remotely extracted from the examiner's PostureRay® system for data collection and sorted based on gender and stratification of degenerative changes. Reliability was assessed via intra-class correlations (ICC), root mean squared error (RMSE), and R2 values. Results: In comparing repeated measures of the CNN network to itself, perfect reliability was found for the ICC (1.0), RMSE (0), and R2 (1). The reliability of the trained expert US was in the excellent range for all variables, where 12/18 variables had ICCs ≥ 0.9 and 6/18 variables were 0.84 ≤ ICCs ≤ 0.89. Similarly, for the expert US, all R2 values were in the excellent range (R2 ≥ 0.7), and all RMSEs were small, being 0.42 ≤ RMSEs ≤ 3.27. Construct validity between the expert US and the CNN network was found to be in the excellent range with 18/18 ICCs in the excellent range (ICCs ≥ 0.8), 16/18 R2 values in the strong to excellent range (R2 ≥ 0.7), and 2/18 in the good to moderate range (R2 RT C6/C7 = 0.57 and R2 Cobb C6/C7 = 0.64. The RMSEs for expert US vs. the CNN network were small, being 0.37 ≤ RMSEs ≤ 2.89. Conclusions: A comparison of repeated measures within the computer vision CNN network and expert human found exceptional reliability and excellent construct validity when comparing the computer vision to the human observer.

A Convolutional Neural Network for Automated Detection of Cervical Ossification of the Posterior Longitudinal Ligament using Magnetic Resonance Imaging

STUDY DESIGN

Retrospective cohort study.

OBJECTIVE

We aimed to develop and validate a convolutional neural network (CNN) model to distinguish between cervical ossification of posterior longitudinal ligament (OPLL) and multilevel degenerative spinal stenosis using Magnetic Resonance Imaging (MRI) and to compare the diagnostic ability with spine surgeons.

SUMMARY OF BACKGROUND DATA

Some artificial intelligence models have been applied in spinal image analysis and many of promising results were obtained; however, there was still no study attempted to develop a deep learning model in detecting cervical OPLL using MRI images.

MATERIALS AND METHODS

In this retrospective study, 272 cervical OPLL and 412 degenerative patients underwent surgical treatment were enrolled and divided into the training (513 cases) and test dataset (171 cases). CNN models applying ResNet architecture with 34, 50, and 101 layers of residual blocks were constructed and trained with the sagittal MRI images from the training dataset. To evaluate the performance of CNN, the receiver operating characteristic curves of 3 ResNet models were plotted and the area under the curve were calculated on the test dataset. The accuracy, sensitivity, and specificity of the diagnosis by the CNN were calculated and compared with 3 senior spine surgeons.

RESULTS

The diagnostic accuracies of our ResNet34, ResNet50, and ResNet101 models were 92.98%, 95.32%, and 97.66%, respectively; the area under the curve of receiver operating characteristic curves of these models were 0.914, 0.942, and 0.971, respectively. The accuracies and specificities of ResNet50 and ResNet101 models were significantly higher than all spine surgeons; for the sensitivity, ResNet101 model achieved better values than that of the 2 surgeons.

CONCLUSION

The performance of our ResNet model in differentiating cervical OPLL from degenerative spinal stenosis using MRI is promising, better results were achieved with more layers of residual blocks applied.

Automatic Segmentation and Radiologic Measurement of Distal Radius Fractures Using Deep Learning

Background

Recently, deep learning techniques have been used in medical imaging studies. We present an algorithm that measures radiologic parameters of distal radius fractures using a deep learning technique and compares the predicted parameters with those measured by an orthopedic hand surgeon.

Methods

We collected anteroposterior (AP) and lateral X-ray images of 634 wrists in 624 patients with distal radius fractures treated conservatively with a follow-up of at least 2 months. We allocated 507 AP and 507 lateral images to the training set (80% of the images were used to train the model, and 20% were utilized for validation) and 127 AP and 127 lateral images to the test set. The margins of the radius and ulna were annotated for ground truth, and the scaphoid in the lateral views was annotated in the box configuration to determine the volar side of the images. Radius segmentation was performed using attention U-Net, and the volar/dorsal side was identified using a detection and classification model based on RetinaNet. The proposed algorithm measures the radial inclination, dorsal or volar tilt, and radial height by index axes and points from the segmented radius and ulna.

Results

The segmentation model for the radius exhibited an accuracy of 99.98% and a Dice similarity coefficient (DSC) of 98.07% for AP images, and an accuracy of 99.75% and a DSC of 94.84% for lateral images. The segmentation model for the ulna showed an accuracy of 99.84% and a DSC of 96.48%. Based on the comparison of the radial inclinations measured by the algorithm and the manual method, the Pearson correlation coefficient was 0.952, and the intraclass correlation coefficient was 0.975. For dorsal/volar tilt, the correlation coefficient was 0.940, and the intraclass correlation coefficient was 0.968. For radial height, it was 0.768 and 0.868, respectively.

Conclusions

The deep learning-based algorithm demonstrated excellent segmentation of the distal radius and ulna in AP and lateral radiographs of the wrist with distal radius fractures and afforded automatic measurements of radiologic parameters.

Machine Learning-Based Measurement of Regional and Global Spinal Parameters Using the Concept of Incidence Angle of Inflection Points

This study delves into the application of convolutional neural networks (CNNs) in evaluating spinal sagittal alignment, introducing the innovative concept of incidence angles of inflection points (IAIPs) as intuitive parameters to capture the interplay between pelvic and spinal alignment. Pioneering the fusion of IAIPs with machine learning for sagittal alignment analysis, this research scrutinized whole-spine lateral radiographs from hundreds of patients who visited a single institution, utilizing high-quality images for parameter assessments. Noteworthy findings revealed robust success rates for certain parameters, including pelvic and C2 incidence angles, but comparatively lower rates for sacral slope and L1 incidence. The proposed CNN-based machine learning method demonstrated remarkable efficiency, achieving an impressive 80 percent detection rate for various spinal angles, such as lumbar lordosis and thoracic kyphosis, with a precise error threshold of 3.5°. Further bolstering the study's credibility, measurements derived from the novel formula closely aligned with those directly extracted from the CNN model. In conclusion, this research underscores the utility of the CNN-based deep learning algorithm in delivering precise measurements of spinal sagittal parameters, and highlights the potential for integrating machine learning with the IAIP concept for comprehensive data accumulation in the domain of sagittal spinal alignment analysis, thus advancing our understanding of spinal health.

Predicting female pelvic tilt and lumbar angle using machine learning in case of urinary incontinence and sexual dysfunction

Urinary incontinence (UI) is defined as any uncontrolled urine leakage. Pelvic floor muscles (PFM) appear to be a crucial aspect of trunk and lumbo-pelvic stability, and UI is one indication of pelvic floor dysfunction. The evaluation of pelvic tilt and lumbar angle is critical in assessing the alignment and posture of the spine in the lower back region and pelvis, and both of these variables are directly related to female dysfunction in the pelvic floor. UI affects a significant number of women worldwide and can have a major impact on their quality of life. However, traditional methods of assessing these parameters involve manual measurements, which are time-consuming and prone to variability. The rehabilitation programs for pelvic floor dysfunction (FSD) in physical therapy often focus on pelvic floor muscles (PFMs), while other core muscles are overlooked. Therefore, this study aimed to predict the activity of various core muscles in multiparous women with FSD using multiple scales instead of relying on Ultrasound imaging. Decision tree, SVM, random forest, and AdaBoost models were applied to predict pelvic tilt and lumbar angle using the train set. Performance was evaluated on the test set using MSE, RMSE, MAE, and R2. Pelvic tilt prediction achieved R2 values > 0.9, with AdaBoost (R2 = 0.944) performing best. Lumbar angle prediction performed slightly lower with decision tree achieving the highest R2 of 0.976. Developing a machine learning model to predict pelvic tilt and lumbar angle has the potential to revolutionize the assessment and management of this condition, providing faster, more accurate, and more objective assessments than traditional methods.

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).

Comparison of tibial alignment parameters based on clinically relevant anatomical landmarks : a deep learning radiological analysis

AIMS

Accurate identification of the ankle joint centre is critical for estimating tibial coronal alignment in total knee arthroplasty (TKA). The purpose of the current study was to leverage artificial intelligence (AI) to determine the accuracy and effect of using different radiological anatomical landmarks to quantify mechanical alignment in relation to a traditionally defined radiological ankle centre.

METHODS

Patients with full-limb radiographs from the Osteoarthritis Initiative were included. A sub-cohort of 250 radiographs were annotated for landmarks relevant to knee alignment and used to train a deep learning (U-Net) workflow for angle calculation on the entire database. The radiological ankle centre was defined as the midpoint of the superior talus edge/tibial plafond. Knee alignment (hip-knee-ankle angle) was compared against 1) midpoint of the most prominent malleoli points, 2) midpoint of the soft-tissue overlying malleoli, and 3) midpoint of the soft-tissue sulcus above the malleoli.

RESULTS

A total of 932 bilateral full-limb radiographs (1,864 knees) were measured at a rate of 20.63 seconds/image. The knee alignment using the radiological ankle centre was accurate against ground truth radiologist measurements (inter-class correlation coefficient (ICC) = 0.99 (0.98 to 0.99)). Compared to the radiological ankle centre, the mean midpoint of the malleoli was 2.3 mm (SD 1.3) lateral and 5.2 mm (SD 2.4) distal, shifting alignment by 0.34^o (SD 2.4^o) valgus, whereas the midpoint of the soft-tissue sulcus was 4.69 mm (SD 3.55) lateral and 32.4 mm (SD 12.4) proximal, shifting alignment by 0.65^o (SD 0.55^o) valgus. On the intermalleolar line, measuring a point at 46% (SD 2%) of the intermalleolar width from the medial malleoli (2.38 mm medial adjustment from midpoint) resulted in knee alignment identical to using the radiological ankle centre.

CONCLUSION

The current study leveraged AI to create a consistent and objective model that can estimate patient-specific adjustments necessary for optimal landmark usage in extramedullary and computer-guided navigation for tibial coronal alignment to match radiological planning.Cite this article: Bone Jt Open 2022;3(10):767-776.

Development of artificial intelligence for automated measurement of cervical lordosis on lateral radiographs

Cervical sagittal alignment is an essential parameter for the evaluation of spine disorders. Manual measurement is time-consuming and burdensome to measurers. Artificial intelligence (AI) in the form of convolutional neural networks has begun to be used to measure x-rays. This study aimed to develop AI for automated measurement of lordosis on lateral cervical x-rays. We included 4546 cervical x-rays from 1674 patients. For all x-rays, the caudal endplates of C2 and C7 were labeled based on consensus among well-experienced spine surgeons, the data for which were used as ground truth. This ground truth was split into training data and test data, and the AI model learned the training data. The absolute error of the AI measurements relative to the ground truth for 4546 x-rays was determined by fivefold cross-validation. Additionally, the absolute error of AI measurements was compared with the error of other 2 surgeons’ measurements on 415 radiographs of 168 randomly selected patients. In fivefold cross-validation, the absolute error of the AI model was 3.3° in the average and 2.2° in the median. For comparison of other surgeons, the mean absolute error for measurement of 168 patients was 3.1° ± 3.4° for the AI model, 3.9° ± 3.4° for Surgeon 1, and 3.8° ± 4.7° for Surgeon 2. The AI model had a significantly smaller error than Surgeon 1 and Surgeon 2 (P = 0.002 and 0.036). This algorithm is available at (https://ykszk.github.io/c2c7demo/). The AI model measured cervical spine alignment with better accuracy than surgeons. AI can assist in routine medical care and can be helpful in research that measures large numbers of images. However, because of the large errors in rare cases such as highly deformed ones, AI may, in principle, be limited to assisting humans.

Can artificial intelligence support or even replace physicians in measuring sagittal balance? A validation study on preoperative and postoperative full spine images of 170 patients

PURPOSE

Sagittal balance (SB) plays an important role in the surgical treatment of spinal disorders. The aim of this research study is to provide a detailed evaluation of a new, fully automated algorithm based on artificial intelligence (AI) for the determination of SB parameters on a large number of patients with and without instrumentation.

METHODS

Pre- and postoperative sagittal full body radiographs of 170 patients were measured by two human raters, twice by one rater and by the AI algorithm which determined: pelvic incidence, pelvic tilt, sacral slope, L1-S1 lordosis, T4-T12 thoracic kyphosis (TK) and the spino-sacral angle (SSA). To evaluate the agreement between human raters and AI, the mean error (95% confidence interval (CI)), standard deviation and an intra- and inter-rater reliability was conducted using intra-class correlation (ICC) coefficients.

RESULTS

ICC values for the assessment of the intra- (range: 0.88-0.97) and inter-rater (0.86-0.97) reliability of human raters are excellent. The algorithm is able to determine all parameters in 95% of all pre- and in 91% of all postoperative images with excellent ICC values (PreOP-range: 0.83-0.91, PostOP: 0.72-0.89). Mean errors are smallest for the SSA (PreOP: -0.1° (95%-CI: -0.9°-0.6°); PostOP: -0.5° (-1.4°-0.4°)) and largest for TK (7.0° (6.1°-7.8°); 7.1° (6.1°-8.1°)).

CONCLUSION

A new, fully automated algorithm that determines SB parameters has excellent reliability and agreement with human raters, particularly on preoperative full spine images. The presented solution will relieve physicians from time-consuming routine work of measuring SB parameters and allow the analysis of large databases efficiently.

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.

Image Quality Control in Lumbar Spine Radiography Using Enhanced U-Net Neural Networks

Purpose

To standardize the radiography imaging procedure, an image quality control framework using the deep learning technique was developed to segment and evaluate lumbar spine x-ray images according to a defined quality control standard.

Materials and Methods

A dataset comprising anteroposterior, lateral, and oblique position lumbar spine x-ray images from 1,389 patients was analyzed in this study. The training set consisted of digital radiography images of 1,070 patients (800, 798, and 623 images of the anteroposterior, lateral, and oblique position, respectively) and the validation set included 319 patients (200, 205, and 156 images of the anteroposterior, lateral, and oblique position, respectively). The quality control standard for lumbar spine x-ray radiography in this study was defined using textbook guidelines of as a reference. An enhanced encoder-decoder fully convolutional network with U-net as the backbone was implemented to segment the anatomical structures in the x-ray images. The segmentations were used to build an automatic assessment method to detect unqualified images. The dice similarity coefficient was used to evaluate segmentation performance.

Results

The dice similarity coefficient of the anteroposterior position images ranged from 0.82 to 0.96 (mean 0.91 ± 0.06); the dice similarity coefficient of the lateral position images ranged from 0.71 to 0.95 (mean 0.87 ± 0.10); the dice similarity coefficient of the oblique position images ranged from 0.66 to 0.93 (mean 0.80 ± 0.14). The accuracy, sensitivity, and specificity of the assessment method on the validation set were 0.971-0.990 (mean 0.98 ± 0.10), 0.714-0.933 (mean 0.86 ± 0.13), and 0.995-1.000 (mean 0.99 ± 0.12) for the three positions, respectively.

Conclusion

This deep learning-based algorithm achieves accurate segmentation of lumbar spine x-ray images. It provides a reliable and efficient method to identify the shape of the lumbar spine while automatically determining the radiographic image quality.

Vertebral Deformity Measurements at MRI, CT, and Radiography Using Deep Learning

PURPOSE

To construct and evaluate the efficacy of a deep learning system to rapidly and automatically locate six vertebral landmarks, which are used to measure vertebral body heights, and to output spine angle measurements (lumbar lordosis angles [LLAs]) across multiple modalities.

MATERIALS AND METHODS

In this retrospective study, MR (n = 1123), CT (n = 137), and radiographic (n = 484) images were used from a wide variety of patient populations, ages, disease stages, bone densities, and interventions (n = 1744 total patients, 64 years ± 8, 76.8% women; images acquired 2005-2020). Trained annotators assessed images and generated data necessary for deformity analysis and for model development. A neural network model was then trained to output vertebral body landmarks for vertebral height measurement. The network was trained and validated on 898 MR, 110 CT, and 387 radiographic images and was then evaluated or tested on the remaining images for measuring deformities and LLAs. The Pearson correlation coefficient was used in reporting LLA measurements.

RESULTS

On the holdout testing dataset (225 MR, 27 CT, and 97 radiographic images), the network was able to measure vertebral heights (mean height percentage of error ± 1 standard deviation: MR images, 1.5% ± 0.3; CT scans, 1.9% ± 0.2; radiographs, 1.7% ± 0.4) and produce other measures such as the LLA (mean absolute error: MR images, 2.90°; CT scans, 2.26°; radiographs, 3.60°) in less than 1.7 seconds across MR, CT, and radiographic imaging studies.

CONCLUSION

The developed network was able to rapidly measure morphometric quantities in vertebral bodies and output LLAs across multiple modalities.Keywords: Computer Aided Diagnosis (CAD), MRI, CT, Spine, Demineralization-Bone, Feature Detection Supplemental material is available for this article. © RSNA, 2021.

Clinical Artificial Intelligence Applications: Musculoskeletal

We present an overview of current clinical musculoskeletal imaging applications for artificial intelligence, as well as potential future applications and techniques.

Emerging Technologies in the Treatment of Adult Spinal Deformity

Outcomes for adult spinal deformity continue to improve as new technologies become integrated into clinical practice. Machine learning, robot-guided spinal surgery, and patient-specific rods are tools that are being used to improve preoperative planning and patient satisfaction. Machine learning can be used to predict complications, readmissions, and generate postoperative radiographs which can be shown to patients to guide discussions about surgery. Robot-guided spinal surgery is a rapidly growing field showing signs of greater accuracy in screw placement during surgery. Patient-specific rods offer improved outcomes through higher correction rates and decreased rates of rod breakage while decreasing operative time. The objective of this review is to evaluate trends in the literature about machine learning, robot-guided spinal surgery, and patient-specific rods in the treatment of adult spinal deformity.

Using Artificial Intelligence for Automatic Segmentation of CT Lung Images in Acute Respiratory Distress Syndrome

Knowledge of gas volume, tissue mass and recruitability measured by the quantitative CT scan analysis (CT-qa) is important when setting the mechanical ventilation in acute respiratory distress syndrome (ARDS). Yet, the manual segmentation of the lung requires a considerable workload. Our goal was to provide an automatic, clinically applicable and reliable lung segmentation procedure. Therefore, a convolutional neural network (CNN) was used to train an artificial intelligence (AI) algorithm on 15 healthy subjects (1,302 slices), 100 ARDS patients (12,279 slices), and 20 COVID-19 (1,817 slices). Eighty percent of this populations was used for training, 20% for testing. The AI and manual segmentation at slice level were compared by intersection over union (IoU). The CT-qa variables were compared by regression and Bland Altman analysis. The AI-segmentation of a single patient required 5–10 s vs. 1–2 h of the manual. At slice level, the algorithm showed on the test set an IOU across all CT slices of 91.3 ± 10.0, 85.2 ± 13.9, and 84.7 ± 14.0%, and across all lung volumes of 96.3 ± 0.6, 88.9 ± 3.1, and 86.3 ± 6.5% for normal lungs, ARDS and COVID-19, respectively, with a U-shape in the performance: better in the lung middle region, worse at the apex and base. At patient level, on the test set, the total lung volume measured by AI and manual segmentation had a R² of 0.99 and a bias −9.8 ml [CI: +56.0/−75.7 ml]. The recruitability measured with manual and AI-segmentation, as change in non-aerated tissue fraction had a bias of +0.3% [CI: +6.2/−5.5%] and −0.5% [CI: +2.3/−3.3%] expressed as change in well-aerated tissue fraction. The AI-powered lung segmentation provided fast and clinically reliable results. It is able to segment the lungs of seriously ill ARDS patients fully automatically.