Analysis of three-dimensional spine growth for vertebral body tethering patients at 2 and 5 years post operatively.

PURPOSE: Scoliosis can be treated with vertebral body tethering (VBT) as a motion-sparing procedure. However, the knowledge of how growth is affected by a tether spanning multiple levels is unclear in the literature. Three-dimensional true spine length (3D-TSL) is a validated assessment technique that accounts for the shape of the spine in both the coronal and sagittal planes. This study aimed to assess if 3D-TSL increases over a five-year period after VBT implantation in thoracic curves for idiopathic scoliosis. METHODS: Prospectively collected radiographic data from an international pediatric spine registry was analyzed. Complete radiographic data over three visits (post-operative, 2 years, and 5 years) was available for 53 patients who underwent VBT. RESULTS: The mean age at instrumentation of this cohort was 12.2 (9-15) years. The average number of vertebrae instrumented was 7.3 (SD 0.7). Maximum Cobb angles were 50° pre-op, which improved to 26° post-op (p < 0.001) and was maintained at 5 years (30°; p = 0.543). Instrumented Cobb angle was 22° at 5 years (p < 0.001 vs 5-year maximum Cobb angle). An accentuation was seen in global kyphosis from 29° pre-operative to 41° at 5 years (p < 0.05). The global spine length (T1-S1 3D-TSL) started at 40.6 cm; measured 42.8 cm at 2 years; and 44.0 cm at the final visit (all p < 0.05). At 5 years, patients reached an average T1-S1 length that is comparable to a normal population at maturity. Immediate mean post-operative instrumented 3D-TSL (top of UIV-top of LIV) was 13.8 cm two-year length was 14.3 cm; and five-year length was 14.6 cm (all p < 0.05). The mean growth of 0.09 cm per instrumented level at 2 years was approximately 50% of normal thoracic growth. Patients who grew more than 0.5 cm at 2 years had a significantly lower BMI (17.0 vs 19.0, p < 0.05) and smaller pre-operative scoliosis (48° vs 53°, p < 0.05). Other subgroup analyses were not significant for age, skeletal maturity, Cobb angles or number of spanned vertebras as contributing factors. CONCLUSIONS: This series demonstrates that 3D-TSL increased significantly over the thoracic instrumented levels after VBT surgery for idiopathic scoliosis. This represented approximately 50% of expected normal thoracic growth over 2 years.

How Long Can You Delay? Curve Progression While Awaiting Vertebral Body Tethering Surgery.

Background: The implications of delaying surgical intervention for patients with adolescent idiopathic scoliosis (AIS) wishing to undergo vertebral body tethering (VBT) have not yet been explored. It is important to understand how these delays can impact surgical planning and patient outcomes. Methods: This was a retrospective review that analyzed all AIS patients treated between 2015 and 2021 at a single tertiary center. Time to surgery from initial surgical consultation and ultimate surgical plan were assessed. Patient characteristics, potential risk factors associated with increased curve progression, and reasons for delay were also analyzed. Results: 174 patients were evaluated and 95 were scheduled for VBT. Four patients later required a change to posterior spinal fusion (PSF) due to excessive curve progression. Patients requiring PSF were shown to have significantly longer delays than those who received VBT. Additionally, longer delays, younger age, greater curve progression, and lower skeletal maturity were correlated with significant curve progression (≥5 degrees). Conclusions: Surgical delays for AIS patients awaiting VBT may lead to significant curve progression and necessitate more invasive procedures. Patients with longer delays experienced an increased risk of needing PSF instead of VBT. Of those requiring PSF, the majority were due to insurance denials. Optimizing surgical timing and shared decision-making among patients, families, and healthcare providers are essential for achieving the best outcomes.

Evaluating compliance with the best practice guidelines for wrong-level surgery prevention in high-risk pediatric spine surgery.

PURPOSE: In 2018, Best Practice Guidelines (BPGs) were published for preventing wrong-level surgery in pediatric spinal deformity, but successful implementation has not been established. The purpose of this study was to evaluate BPG compliance 5 years after publication. We hypothesized higher compliance among BPG authors and among surgeons with more experience, higher caseload, and awareness of the BPGs. METHODS: We queried North American and European surgeons, authors and nonauthors, and members of pediatric spinal study groups on adherence to BPGs using an anonymous survey consisting of 18 Likert scale questions. Respondents provided years in practice, yearly caseload, and guideline awareness. Mean compliance scores (MCS) were developed by correlating Likert responses with MCS scores ("None of the time" = no compliance = MCS 0, "Sometimes" = weak to moderate = MCS 1, "Most of the time" = high = MCS 2, and "All the time" = perfect = MCS 3). RESULTS: Of the 134 respondents, 81.5% reported high or perfect compliance. Average MCS for all guidelines was 2.4 ± 0.4. North American and European surgeons showed no compliance differences (2.4 vs. 2.3, p = 0.07). Authors and nonauthors showed significantly different compliance scores (2.8 vs 2.4, p < 0.001), as did surgeons with and without knowledge of the BPGs (2.5 vs 2.2, p < 0.001). BPG awareness and compliance showed a moderate positive correlation (r = 0.48, p < 0.001), with non-significant associations between compliance and both years in practice (r = 0.41, p = 0.64) and yearly caseload (r = 0.02, p = 0.87). CONCLUSION: Surgeons reported high or perfect compliance 81.5% of the time with BPGs for preventing wrong-level surgery. Authorship and BPG awareness showed increased compliance. Location, study group membership, years in practice, and yearly caseload did not affect compliance. LEVEL OF EVIDENCE: Level V-expert opinion.

Coronal decompensation following thoracic vertebral body tethering in idiopathic scoliosis.

PURPOSE: Post-operative coronal decompensation (CD) continues to be a challenge in the treatment of adolescent idiopathic scoliosis (AIS). CD following selective spinal fusion has been studied. However, there is currently little information regarding CD following Vertebral Body Tethering (VBT). Thus, the goal of this study is to better understand the incidence and risk factors for CD after VBT. METHODS: Retrospective review of a prospective multicenter database was used for analysis. Inclusion criteria were patients undergoing thoracic VBT, a minimum 2-year follow-up, LIV was L1 or above, skeletally immature (Risser ≤ 1), and available preoperative and final follow-up AP and lateral upright radiographs. Radiographic parameters including major and minor Cobb angles, curve type, LIV tilt/translation, L4 tilt, and coronal balance were measured. CD was defined as the distance between C7PL and CSVL > 2 cm. Multiple logistic regression model was used to identify significant predictors of CD. RESULTS: Out of 136 patients undergoing VBT, 94 patients (86 female and 6 male) met the inclusion criteria. The mean age at surgery was 12.1 (9-16) and mean follow-up period was 3.4 years (2-5 years). Major and minor curves, AVR, coronal balance, LIV translation, LIV tilt, L4 tilt were significantly improved after surgery. CD occurred in 11% at final follow-up. Lenke 1A-R (24%) and 1C (26%) had greater incidence of CD compared to 1A-L (4%), 2 (0%), and 3 (0%). LIV selection was not associated with CD. Multivariate logistic regression analysis yielded 1A-R and 1C curves as a predictor of CD with the odds ratio being 17.0. CONCLUSION: CD occurred in 11% of our thoracic VBT patients. Lenke 1A-R and 1C curve types were predictors for CD in patients treated with VBT. There were no other preoperative predictors associated with CD.

Institutional Variability in Anesthesia Time for Mehta Casting in Early-Onset Scoliosis (EOS).

PURPOSE: Mehta casting is a potentially curative intervention for early-onset scoliosis (EOS) that typically requires multiple anesthetics. The Food and Drug Administration (FDA) reported that >3 hours of anesthesia under the age of 3 years old may alter brain development; however, no standard exists for the duration of anesthesia during casting. The purpose of this study is to quantify the variability in anesthesia during Mehta casting. We hypothesize that significant institutional variability exists and may be attributed to modifiable factors. METHODS: An EOS registry was used to identify patients who underwent at least one Mehta casting procedure. Anesthesia exposure was quantified, and site variability was assessed by patient characteristics, cast placement, procedure type, and equipment used. RESULTS: Our cohort consisted of 208 patients from 5 institutions (age 2.6±1.4 y). There were 1097 Mehta casting procedures, with 5.4±3.6 castings per patient. Of these patients, 106 (51%) were female, with an average age of 2.11±1.29 years old at the time casting was initiated. Patient etiologies included 154 idiopathic (74.0%), 22 syndromic (10.6%), 18 congenital (8.7%), 11 neuromuscular (5.3%), and 3 unknown (1.4%). Anesthesia time was 69±31 minutes and varied significantly between sites (59±14 to 117±46 min; P <0.001). Cumulative anesthesia time for patients under 3 years was 320±197 minutes with 120/161 (74.5%) patients exceeding 3 hours. Anesthesia time was lower after the FDA warning in 2016 compared with pre-2016 (71±30 vs. 66±32, P =0.008). CONCLUSIONS: Patients undergoing Mehta casting are at significant risk of exceeding 3 hours of anesthesia, which the FDA has stated may be harmful for children <3 years. Significant site variability indicates that standardization protocols should be developed to encourage best practices and minimize anesthetic times. LEVEL OF EVIDENCE: Prognostic Level II.

Artificial intelligence-based three-dimensional templating for total joint arthroplasty planning: a scoping review.

PURPOSE: The purpose of this review is to evaluate the current status of research on the application of artificial intelligence (AI)-based three-dimensional (3D) templating in preoperative planning of total joint arthroplasty. METHODS: This scoping review followed the PRISMA, PRISMA-ScR guidelines, and five stage methodological framework for scoping reviews. Studies of patients undergoing primary or revision joint arthroplasty surgery that utilised AI-based 3D templating for surgical planning were included. Outcome measures included dataset and model development characteristics, AI performance metrics, and time performance. After AI-based 3D planning, the accuracy of component size and placement estimation and postoperative outcome data were collected. RESULTS: Nine studies satisfied inclusion criteria including a focus on computed tomography (CT) or magnetic resonance imaging (MRI)-based AI templating for use in hip or knee arthroplasty. AI-based 3D templating systems reduced surgical planning time and improved implant size/position and imaging feature estimation compared to conventional radiographic templating. Several components of data processing and model development and testing were insufficiently covered in the studies included in this scoping review. CONCLUSIONS: AI-based 3D templating systems have the potential to improve preoperative planning for joint arthroplasty surgery. This technology offers more accurate and personalized preoperative planning, which has potential to improve functional outcomes for patients. However, deficiencies in several key areas, including data handling, model development, and testing, can potentially hinder the reproducibility and reliability of the methods proposed. As such, further research is needed to definitively evaluate the efficacy and feasibility of these systems.

The Effect of Implant Density on Adolescent Idiopathic Scoliosis Fusion: Results of the Minimize Implants Maximize Outcomes Randomized Clinical Trial.

BACKGROUND: Severe adolescent idiopathic scoliosis (AIS) can be treated with instrumented fusion, but the number of anchors needed for optimal correction is controversial. METHODS: We conducted a multicenter, randomized study that included patients undergoing spinal fusion for single thoracic curves between 45° and 65°, the most common form of operatively treated AIS. Of the 211 patients randomized, 108 were assigned to a high-density screw pattern and 103, to a low-density screw pattern. Surgeons were instructed to use ≥1.8 implants per spinal level fused for patients in the high-implant-density group or ≤1.4 implants per spinal level fused for patients in the low-implant-density group. The primary outcome measure was the percent correction of the coronal curve at the 2-year follow-up. The power analysis for this trial required 174 patients to show equivalence, defined as a 95% confidence interval (CI) within a ±10% correction margin with a probability of 90%. RESULTS: In the intention-to-treat analysis, the mean percent correction of the coronal curve was equivalent between the high-density and low-density groups at the 2-year follow-up (67.6% versus 65.7%; difference, -1.9% [95% CI: -6.1%, 2.2%]). In the per-protocol cohorts, the mean percent correction of the coronal curve was also equivalent between the 2 groups at the 2-year follow-up (65.0% versus 66.1%; difference, 1.1% [95% CI: -3.0%, 5.2%]). A total of 6 patients in the low-density group and 5 patients in the high-density group required reoperation (p = 1.0). CONCLUSIONS: In the setting of spinal fusion for primary thoracic AIS curves between 45° and 65°, the percent coronal curve correction obtained with use of a low-implant-density construct and that obtained with use of a high-implant-density construct were equivalent. LEVEL OF EVIDENCE: Therapeutic Level I . See Instructions for Authors for a complete description of levels of evidence.

Non-Fusion Versus Fusion Surgery in Pediatric Idiopathic Scoliosis: What Trade-Offs in Outcomes Are Acceptable for the Patient and Family?

BACKGROUND: Vertebral body tethering and other non-fusion techniques for the treatment of pediatric idiopathic scoliosis are increasing in popularity. There is limited physician consensus on this topic as the result of a paucity of published data regarding which patients most benefit from non-fusion strategies. Thus, much of the decision-making is left to patients and parents, who must select a treatment based on their goals and values and the information available from health-care providers, the internet, and social media. We sought to understand patient and family preferences regarding the attributes of fusion versus non-fusion surgery that drive these choices. METHODS: Patients and families were recruited from 7 pediatric spine centers and were asked to complete a survey-based choice experiment that had been jointly developed with the U.S. Food and Drug Administration (FDA) to evaluate patient preferences. Choices between experimentally designed alternatives were analyzed to estimate the relative importance of outcomes and requirements associated with the choice options (attributes). The attributes included appearance, confidence in the planned correction, spinal motion, device failure, reoperation, and recovery period. The inclusion criteria were (1) an age of 10 to 21 years and (2) a diagnosis of adolescent idiopathic scoliosis in patients who were considering, or who had already undergone, treatment with fusion or non-fusion surgery. Preference weights were estimated from the expected changes in choice given changes in the attributes. RESULTS: A total of 344 respondents (124 patients, 92 parents, and 128 parent/patient dyads) completed the survey. One hundred and seventy-three patients were enrolled prior to surgery, and 171 were enrolled after surgery. Appearance and motion were found to be the most important drivers of choice. For the entire cohort, fusion was preferred over non-fusion. For patients who were considering surgery, the most important attributes were preservation of spinal motion and appearance. CONCLUSIONS: Patients and families seeking treatment for idiopathic scoliosis value appearance and preservation of spinal motion and, to a lesser extent, reoperation rates when considering fusion versus non-fusion surgery.

Potential endpoints for assessment of bone health in persons with neurofibromatosis type 1.

Neurofibromatosis type 1 is a genetic syndrome characterized by a wide variety of tumor and non-tumor manifestations. Bone-related issues, such as scoliosis, tibial dysplasia, and low bone mineral density, are a significant source of morbidity for this population with limited treatment options. Some of the challenges to developing such treatments include the lack of consensus regarding the optimal methods to assess bone health in neurofibromatosis type 1 and limited data regarding the natural history of these manifestations. In this review, the Functional Committee of the Response Evaluation in Neurofibromatosis and Schwannomatosis International Collaboration: (1) presents the available techniques for measuring overall bone health and metabolism in persons with neurofibromatosis type 1, (2) reviews data for use of each of these measures in the neurofibromatosis type 1 population, and (3) describes the strengths and limitations for each method as they might be used in clinical trials targeting neurofibromatosis type 1 bone manifestations. The Response Evaluation in Neurofibromatosis and Schwannomatosis International Collaboration supports the development of a prospective, longitudinal natural history study focusing on the bone-related manifestations and relevant biomarkers of neurofibromatosis type 1. In addition, we suggest that the neurofibromatosis type 1 research community consider adding the less burdensome measurements of bone health as exploratory endpoints in ongoing or planned clinical trials for other neurofibromatosis type 1 manifestations to expand knowledge in the field.

Automated measurements of interscrew angles in vertebral body tethering patients with deep learning.

BACKGROUND CONTEXT: Vertebral body tethering is the most popular nonfusion treatment for adolescent idiopathic scoliosis. The effect of the tether cord on the spine can be segmentally assessed by comparing the angle between two adjacent screws (interscrew angle) over time. Tether breakage has historically been assessed radiographically by a change in adjacent interscrew angle by greater than 5° between two sets of imaging. A threshold for growth modulation has not yet been established in the literature. These angle measurements are time consuming and prone to interobserver variability. PURPOSE: The purpose of this study was to develop an automated deep learning algorithm for measuring the interscrew angle following VBT surgery. STUDY DESIGN/SETTING: Single institution analysis of medical images. PATIENT SAMPLE: We analyzed 229 standing or bending AP or PA radiographs from 100 patients who had undergone VBT at our institution. OUTCOME MEASURES: Physiologic Measures: An image processing algorithm was used to measure interscrew angles. METHODS: A total of 229 standing or bending AP or PA radiographs from 100 VBT patients with vertebral body tethers were identified. Vertebral body screws were segmented by hand for all images and interscrew angles measured manually for 60 of the included images. A U-Net deep learning model was developed to automatically segment the vertebral body screws. Screw label maps were used to develop and tune an image processing algorithm which measures interscrew angles. Finally, the completed model and algorithm pipeline was tested on a 30-image test set. Dice score and absolute error were used to measure performance. RESULTS: Inter- and Intra-rater reliability for manual angle measurements were assessed with ICC and were both 0.99. The segmentation model Dice score against manually segmented ground truth across the 30-image test set was 0.96. The average interscrew angle absolute error between the algorithm and manually measured ground truth was 0.66° and ranged from 0° to 2.67° in non-overlapping screws (N=206). The primary modes of failure for the model were overlapping screws on a right thoracic/left lumbar construct with two screws in one vertebra and overexposed images. An algorithm step which determines whether an overlapping screw was present correctly identified all overlapping screws, with no false positives. CONCLUSION: We developed and validated an algorithm which measures interscrew angles for radiographs of vertebral body tether patients with an accuracy of within 1° for the majority of interscrew angles. The algorithm can process five images per second on a standard computer, leading to substantial time savings. This algorithm may be used for rapid processing of large radiographic databases of tether patients and could enable more rigorous definitions of growth modulation and cord breakage to be established.

Evaluating compliance with the 2013 best practice guidelines for preventing surgical site infection in high-risk pediatric spine surgery.

PURPOSE: Best Practice Guidelines (BPGs) were published one decade ago to decrease surgical site infection (SSI) in pediatric spinal deformity. Successful implementation has not been established. This study evaluated surgeon compliance with items on the BPG. We hypothesized that BPG authors and surgeons with more experience, higher caseload, and awareness of the BPG would have higher compliance. METHODS: We queried North American and European surgeons, authors and non-authors, and members of various spine study groups on adherence to BPGs using an anonymous survey. Mean compliance scores (MCSs) were developed by correlating Likert responses with MCSs ("None of the time" = no compliance = MCS 0, "Sometimes" = weak to moderate = MCS 1, "Most of the time" = high = MCS 2, "All the time" = perfect = MCS 3). RESULTS: Of the 142 respondents, 73.7% reported high or perfect compliance. Average compliance scores for all guidelines was 2.2 ± 0.4. There were significantly different compliance scores between North American and European surgeons (2.3 vs 1.8, p < 0.001), authors and non-authors (2.5 vs. 2.2, p = 0.023), and surgeons with and without knowledge of the BPGs (2.3 vs. 1.8, p < 0.001). There was a weak correlation between BPG awareness and compliance (r = 0.34, p < 0.001) and no correlation between years in practice (r = 0.0, p = 0.37) or yearly caseload (r = 0.2, p = 0.78) with compliance. CONCLUSIONS: Compliance among our cohort of surgeons surveyed was high. North American surgeons, authors of the BPGs and those aware of the guidelines had increased compliance. Participation in a spine study group, years in practice, and yearly caseload were not associated with compliance. LEVEL OF EVIDENCE: Level V-expert opinion.

Radiation exposure in navigated techniques for AIS: is there a difference between pre-operative CT and intraoperative CT?

PURPOSE: Utilization of navigation improves pedicle screw accuracy in adolescent idiopathic scoliosis (AIS). Our center switched from intraoperative CT (ICT) to an optical navigation system that utilizes pre-operative CT (PCT). We aim to evaluate the radiation dose and operative time for low-dose ICT compared to standard and low-dose PCT used for optical navigation in AIS patients undergoing posterior spinal fusion. METHODS: A single-center matched-control cohort study of 38 patients was conducted. Nineteen patients underwent ICT navigation (O-arm) and were matched by sex, age, and weight to 19 patients who underwent PCT for use with an optical-guided navigation (7D, Seaspine). A total of 418 levels were instrumented and reviewed. PCT was either a standard dose (N = 7) or a low dose (N = 12). The mean volume CT dose index, dose-length product, overall effective dose (ED), ED per level instrumented, and operative time per level were compared. RESULTS: ED per level instrumented was 0.061 ± 0.029 mSv in low-dose PCT and 0.14 ± 0.05 mSv in low-dose ICT (p < 0.0001). ED per level instrumented was significantly higher in standard PCT (1.46 ± 0.39 vs. 0.14 ± 0.03 mSv; p < 0.0001). Mean operative time per level was 31 ± 7 min for ICT and 33 ± 3 min for PCT (p = 0.628). CONCLUSION: Low-dose PCT resulted in 0.70 mSv exposure per case and 31 min per level, standard-dose was 16.95 mSv, while ICT resulted in 1.34-1.62 mSv and a similar operative time. Use of a standard-dose PCT involves radiation exposure about 9 times higher than ICT and 23 times higher than low-dose PCT per level instrumented. LEVEL OF EVIDENCE: Level III.

Does radiographic calibration affect linear radiographic measurements in a large pediatric spine registry?

PURPOSE: Large registries are increasingly at the forefront of modern pediatric spine research, with manual, centralized, trained radiographic measurement serving as the gold standard for spine research. However, there is limited data regarding the reliability of registry measurements which may be subject to differences in radiographic calibration. We undertook this study to evaluate reliability of T1-T12 height, L1-S1 height, and coronal balance measurements for a large registry of early onset scoliosis patients. METHODS: Three trained technicians from the Pediatric Spine Study Group measured 43 radiographs for T1-T12, L1-S1, and coronal balance using 3 different calibration techniques. All radiographs were AP views of patients with magnetically controlled growing rods with known diameters. The calibration techniques used a pre-export manually drawn line, a digital automatically generated computerized marker, and the diameter of a magnetically controlled growing rod. The intraclass correlation coefficient (ICC) was calculated to determine reliability. RESULTS: 1161 measurements were performed. For each of the three raters, coronal balance, T1-T12 height and L1-S1 height had excellent agreement regardless of the calibration technique (alpha 0.93-1.0). Among the parameters, coronal balance had the worst inter-rater reliability, whereas there was excellent interrater reliability regarding T1-T12 height and L1-S1 height (alpha 0.91-0.99). CONCLUSION: There was excellent agreement among reviewers and between the 3 different calibration techniques. While calibration using rod diameter served as the gold standard, this data shows that other standard calibration methods were adequate and achieved excellent reliability for registry radiographs.

Anonymizing Radiographs Using an Object Detection Deep Learning Algorithm.

Radiographic markers contain protected health information that must be removed before public release. This work presents a deep learning algorithm that localizes radiographic markers and selectively removes them to enable de-identified data sharing. The authors annotated 2000 hip and pelvic radiographs to train an object detection computer vision model. Data were split into training, validation, and test sets at the patient level. Extracted markers were then characterized using an image processing algorithm, and potentially useful markers (eg, "L" and "R") without identifying information were retained. The model achieved an area under the precision-recall curve of 0.96 on the internal test set. The de-identification accuracy was 100% (400 of 400), with a de-identification false-positive rate of 1% (eight of 632) and a retention accuracy of 93% (359 of 386) for laterality markers. The algorithm was further validated on an external dataset of chest radiographs, achieving a de-identification accuracy of 96% (221 of 231). After fine-tuning the model on 20 images from the external dataset to investigate the potential for improvement, a 99.6% (230 of 231, P = .04) de-identification accuracy and decreased false-positive rate of 5% (26 of 512) were achieved. These results demonstrate the effectiveness of a two-pass approach in image de-identification. Keywords: Conventional Radiography, Skeletal-Axial, Thorax, Experimental Investigations, Supervised Learning, Transfer Learning, Convolutional Neural Network (CNN) Supplemental material is available for this article. © RSNA, 2023 See also the commentary by Chang and Li in this issue.

A Stepwise Approach to Analyzing Musculoskeletal Imaging Data With Artificial Intelligence.

The digitization of medical records and expanding electronic health records has created an era of "Big Data" with an abundance of available information ranging from clinical notes to imaging studies. In the field of rheumatology, medical imaging is used to guide both diagnosis and treatment of a wide variety of rheumatic conditions. Although there is an abundance of data to analyze, traditional methods of image analysis are human resource intensive. Fortunately, the growth of artificial intelligence (AI) may be a solution to handle large datasets. In particular, computer vision is a field within AI that analyzes images and extracts information. Computer vision has impressive capabilities and can be applied to rheumatologic conditions, necessitating a need to understand how computer vision works. In this article, we provide an overview of AI in rheumatology and conclude with a five step process to plan and conduct research in the field of computer vision. The five steps include (1) project definition, (2) data handling, (3) model development, (4) performance evaluation, and (5) deployment into clinical care.

Few-shot biomedical image segmentation using diffusion models: Beyond image generation.

BACKGROUND: Medical image analysis pipelines often involve segmentation, which requires a large amount of annotated training data, which is time-consuming and costly. To address this issue, we proposed leveraging generative models to achieve few-shot image segmentation. METHODS: We trained a denoising diffusion probabilistic model (DDPM) on 480,407 pelvis radiographs to generate 256 âœ• 256 px synthetic images. The DDPM was conditioned on demographic and radiologic characteristics and was rigorously validated by domain experts and objective image quality metrics (Frechet inception distance [FID] and inception score [IS]). For the next step, three landmarks (greater trochanter [GT], lesser trochanter [LT], and obturator foramen [OF]) were annotated on 45 real-patient radiographs; 25 for training and 20 for testing. To extract features, each image was passed through the pre-trained DDPM at three timesteps and for each pass, features from specific blocks were extracted. The features were concatenated with the real image to form an image with 4225 channels. The feature-set was broken into random patches, which were fed to a U-Net. Dice Similarity Coefficient (DSC) was used to compare the performance with a vanilla U-Net trained on radiographs. RESULTS: Expert accuracy was 57.5 % in determining real versus generated images, while the model reached an FID = 7.2 and IS = 210. The segmentation UNet trained on the 20 feature-sets achieved a DSC of 0.90, 0.84, and 0.61 for OF, GT, and LT segmentation, respectively, which was at least 0.30 points higher than the naively trained model. CONCLUSION: We demonstrated the applicability of DDPMs as feature extractors, facilitating medical image segmentation with few annotated samples.

Long-Term Outcomes at Skeletal Maturity of Combined Pelvic and Femoral Osteotomy for the Treatment of Legg-Calve-Perthes Disease.

Surgical treatment for Legg-Calve-Perthes disease (LCPD) is recommended for older children with moderate to severe disease. We sought to determine whether double osteotomies lead to improved radiologic outcomes compared to reported non-operative outcomes. Patients older than 6 years of age diagnosed with LCPD lateral pillar B or C who were treated with pelvic and femoral osteotomies were included. Radiologic outcomes and leg-length discrepancies were assessed using the Stulberg classification and were compared with the current literature. Fifteen hips in fourteen patients were treated with double osteotomy for LCPD, and seven had lateral pillar C disease (47%). The mean age at surgery was 8.6 years (range, 7.2-10.4) and the mean age at follow-up was 20.2 years (range, 14.2-35.6). At a mean 11.6-year follow-up (range: 6.3-25.2), double osteotomy resulted in 40% of patients having Stulberg I/II scores, 27% having Stulberg III scores, and 33% having Stulberg IV/V scores. The mean leg-length discrepancy was 1.4 cm in lateral pillar C patients compared to 0.8 cm in lateral pillar B patients. Four patients underwent additional surgeries, including two who required total hip arthroplasty. Double osteotomy as an alternative surgical procedure for the treatment of LCPD did not show improved outcomes when compared to historic non-operative cohorts.

Biomechanical Computational Study of Pedicle Screw Position and Density in Adolescent Idiopathic Scoliosis Instrumentation.

STUDY DESIGN: Computer simulation of adolescent idiopathic scoliosis instrumentation. OBJECTIVE: To test the hypothesis that different screw densities would result in different apical vertebral rotation (AVR) corrections and bone-screw forces in adolescent idiopathic scoliosis instrumentation. SUMMARY OF BACKGROUND DATA: The "Minimize Implants Maximize Outcomes" Clinical Trial revealed that the use of more versus fewer screws resulted in similar coronal plane correction for Lenke 1A curves. However, the biomechanical impact of screw density on transverse plane correction is still unclear. Further investigation is needed to determine if and how transverse plane correction is correlated with screw density. PATIENTS AND METHODS: We simulated apical vertebral derotation after segmental translation using patient-specific computer models of 30 patients from the "Minimize Implants Maximize Outcomes" Trial. For each case, 10 alternative screw patterns were tested with overall densities ranging between 1.2 and 2 screws per level fused, and local density at the 3 apical levels ranging between 0.7 and 2 (total: 600 simulations). Main thoracic (MT) Cobb angle, thoracic kyphosis (TK), AVR, and bone-screw forces were computed and compared. RESULTS: The presenting MT (62 ± 11°; range: 45° to 86°), TK (27 ± 20°; -5° to 81°), and AVR (14±7°; -2° to 25°) were corrected through segmental translation to 22 ± 7° (10° to 41°), 26 ± 5° (18° to 45°), and 14 ± 7° (-4° to 26°). After apical vertebral derotation, they became 16 ± 8° (1° to 41°), 24 ± 4° (13° to 40°), and 4 ± 5° (-12° to 18°). There was no significant difference in MT among screw patterns; higher screw density had lower bone-screw forces ( P < 0.05). The apical vertebral derotation maneuver reduced AVR by an average of 70%, positively correlated with apical screw density ( r = 0.825, P < 0.05). There was no significant difference in TK. CONCLUSION: Screw density had no significant effect on 3-dimensional correction through the primary segmental translation maneuver. Transverse plane correction through subsequent apical vertebral derotation was positively correlated with screw density at the apical levels ( r = 0.825, P < 0.05). Bone-screw forces were negatively correlated with overall screw density ( P < 0.05).

Minimum 5-Year Results of Elongation Derotation Flexion Casting for Early Onset Scoliosis: The Story Is Not Over Until Skeletal Maturity.

BACKGROUND: Early conservative treatment for patients with idiopathic infantile scoliosis (IIS) with elongation derotation flexion (EDF) casting and subsequent serial bracing has become widely utilized. However, the long-term outcomes of patients treated with EDF casting are limited. METHODS: We performed a retrospective chart review of all patients who had undergone serial elongation derotation flexion casting and subsequent bracing for scoliosis presenting at a single large tertiary center. All patients were followed for a minimum of 5 years or until surgical intervention. RESULTS: Our study included 21 patients diagnosed with IIS and treated with EDF casting. At a mean 7-year follow-up, 13 of the 21 patients were considered successfully treated with a mean final major coronal curvature of 9 degrees compared to a pretreatment coronal curve of 36 degrees. These patients, on average, began casting at 1.3 years old and spent 1 year in a cast. Patients that did not have substantial improvement began casting at mean 4 years old and remained in a cast for 0.8 years. Three patients initially had substantial improvement with the correction to <20 degrees at a mean age of 7; however, their curves worsened in adolescence with poor brace compliance. All 3 patients will require surgical intervention. Of the patients not successfully treated with casting, 7 required surgery at a mean 8.2 years of age, 4.3 years after initiation of casting. A significant predictor of treatment failure was older age of cast initiation ( P <0.001). CONCLUSIONS: EDF casting can be an effective cure for IIS patients if initiated at a young age with 15 of 21 patients successfully treated (76%). However, 3 patients had a recurrence in adolescence resulting in an overall success rate of only 62%. Casting should be initiated early to maximize the likelihood of treatment success and periodic monitoring should be continued through skeletal maturity as recurrence during adolescence can occur.