Clinical applications of AI-prediction tools in spine surgery: a narrative review.

Spine surgery has grown into a wide, complex field encompassing trauma surgery to deformity to tumours. Artificial intelligence (AI) based technology has been particularly useful in improving imaging-reporting and detection of predictive patterns. The purpose of this narrative review is to present practical approaches towards implementing upcoming AI spine research for clinicians to help improve practices, clinical throughput, and surgical decision-making.

Mobility of the lumbo-pelvic-hip complex (spinopelvic mobility) and sagittal spinal alignment - implications for surgeons performing hip arthroplasty.

INTRODUCTION: The optimal positioning of the hip prosthesis components is influenced by the mobility and balance of the spine. The present study classifies patients with pathology of the spino-pelvic-hip complex, showing possible methods of preventing hip dislocations after arthroplasty. HYPOTHESIS: Hip-Spine Classification helps arthroplasty surgeons to implant components in more patient-specific position. MATERIALS AND METHODS: The group of 100 patients treated with total hip arthroplasty. Antero-posterior (AP) X-rays of the pelvis in a standing position, lateral spine (standing and sitting) and AP of the pelvis (supine after the procedure) were analyzed. We analyzed a change in sacral tilt value when changing from standing to sitting (∆SS), Pelvic Incidence (PI), Lumbar Lordosis (LL) Mismatch, sagittal lumbar pelvic balance (standing position). Patients were classified according to the Hip-Spine Classification. Postoperatively, the inclination and anteversion of the implanted acetabular component were measured. RESULTS: In our study 1 A was diagnosed in 61% of all cases, 1B in 18%, 2 A in 16%, 2B in 5%. 50 out of 61 (82%) in group 1 A were placed within the Levinnek "safe zone". In 1B, 2 A, 2B, the position of the acetabular component was influenced by both the spinopelvic mobility and sagittal spinal balance. The mean inclination was 43.35° and the anteversion was 17.4°. CONCLUSIONS: Categorizing patients according to Hip-Spine Classification one can identify possible consequences the patients at risk. Pathology of the spino-pelvic-hipcomplex can lead to destabilization or dislocation of hip after surgery even though implanted according to Lewinnek's indications. Our findings suggest that Lewinnek safe zone should be abandoned in favor of the concept of functional safe zones.

Deformable registration of preoperative MR and intraoperative long-length tomosynthesis images for guidance of spine surgery via image synthesis.

PURPOSE: Improved integration and use of preoperative imaging during surgery hold significant potential for enhancing treatment planning and instrument guidance through surgical navigation. Despite its prevalent use in diagnostic settings, MR imaging is rarely used for navigation in spine surgery. This study aims to leverage MR imaging for intraoperative visualization of spine anatomy, particularly in cases where CT imaging is unavailable or when minimizing radiation exposure is essential, such as in pediatric surgery. METHODS: This work presents a method for deformable 3D-2D registration of preoperative MR images with a novel intraoperative long-length tomosynthesis imaging modality (viz., Long-Film [LF]). A conditional generative adversarial network is used to translate MR images to an intermediate bone image suitable for registration, followed by a model-based 3D-2D registration algorithm to deformably map the synthesized images to LF images. The algorithm's performance was evaluated on cadaveric specimens with implanted markers and controlled deformation, and in clinical images of patients undergoing spine surgery as part of a large-scale clinical study on LF imaging. RESULTS: The proposed method yielded a median 2D projection distance error of 2.0 mm (interquartile range [IQR]: 1.1-3.3 mm) and a 3D target registration error of 1.5 mm (IQR: 0.8-2.1 mm) in cadaver studies. Notably, the multi-scale approach exhibited significantly higher accuracy compared to rigid solutions and effectively managed the challenges posed by piecewise rigid spine deformation. The robustness and consistency of the method were evaluated on clinical images, yielding no outliers on vertebrae without surgical instrumentation and 3% outliers on vertebrae with instrumentation. CONCLUSIONS: This work constitutes the first reported approach for deformable MR to LF registration based on deep image synthesis. The proposed framework provides access to the preoperative annotations and planning information during surgery and enables surgical navigation within the context of MR images and/or dual-plane LF images.

Spinal Manifestations of Skeletal Dysplasia: A Practical Guide for Clinical Diagnosis.

Skeletal dysplasias are a group of genetic conditions defined by atypical bone or cartilage growth and development. Skeletal abnormalities include short stature, limb deformity, joint contracture, and spinal deformity. Over 90% of disorders have a known genetic mutation that can definitively determine the diagnosis. As patients may present with a primary spinal concern, a careful clinical and radiographic evaluation can allow the physician to develop a working diagnosis to guide additional evaluation. Spinal manifestations include scoliosis and kyphoscoliosis, cervical instability, cervical kyphosis, thoracolumbar kyphosis, spinal stenosis, and atypical vertebral body morphology. An understanding of the affected conditions, prevalence, and natural history of these radiographic findings aids the orthopaedic surgeon in establishing a diagnosis and guides appropriate orthopaedic care.

Automated Detection and Segmentation of Bone Metastases on Spine MRI Using U-Net: A Multicenter Study.

OBJECTIVE: To develop and evaluate a deep learning model for automated segmentation and detection of bone metastasis on spinal MRI. MATERIALS AND METHODS: We included whole spine MRI scans of adult patients with bone metastasis: 662 MRI series from 302 patients (63.5 ± 11.5 years; male:female, 151:151) from three study centers obtained between January 2015 and August 2021 for training and internal testing (random split into 536 and 126 series, respectively) and 49 MRI series from 20 patients (65.9 ± 11.5 years; male:female, 11:9) from another center obtained between January 2018 and August 2020 for external testing. Three sagittal MRI sequences, including non-contrast T1-weighted image (T1), contrast-enhanced T1-weighted Dixon fat-only image (FO), and contrast-enhanced fat-suppressed T1-weighted image (CE), were used. Seven models trained using the 2D and 3D U-Nets were developed with different combinations (T1, FO, CE, T1 + FO, T1 + CE, FO + CE, and T1 + FO + CE). The segmentation performance was evaluated using Dice coefficient, pixel-wise recall, and pixel-wise precision. The detection performance was analyzed using per-lesion sensitivity and a free-response receiver operating characteristic curve. The performance of the model was compared with that of five radiologists using the external test set. RESULTS: The 2D U-Net T1 + CE model exhibited superior segmentation performance in the external test compared to the other models, with a Dice coefficient of 0.699 and pixel-wise recall of 0.653. The T1 + CE model achieved per-lesion sensitivities of 0.828 (497/600) and 0.857 (150/175) for metastases in the internal and external tests, respectively. The radiologists demonstrated a mean per-lesion sensitivity of 0.746 and a mean per-lesion positive predictive value of 0.701 in the external test. CONCLUSION: The deep learning models proposed for automated segmentation and detection of bone metastases on spinal MRI demonstrated high diagnostic performance.

Comparison of elasticity changes in the paraspinal muscles of adolescent patients with scoliosis treated with surgery and bracing.

Scoliosis is a three-dimensional spinal deformity, and paraspinal muscles play an important role as stabilizers of the spinal curve. In this prospective study, we compared elasticity changes in the paraspinal muscles of adolescent patients with scoliosis after surgery or bracing. Elasticity was measured on the concave and convex sides of the paraspinal muscles at the apex of the curve at the beginning of treatment and 6 and 12 months after treatment. Twenty-six patients with correction surgery (n = 15) or bracing (n = 11) were included. At initial evaluation, the Cobb angle was larger in the surgery group (72.3 ± 20.2° in surgery vs. 30.6 ± 5.1° in brace, p < 0.001). The estimated mean elasticity value of the paraspinal muscles was lower in the surgery group at baseline on the convex side (15.8 vs. 22.8 kPa, p = 0.037) and 6 months on both the concave (12.1 vs. 22.7 kPa, p = 0.004) and convex (13.4 vs. 23.8 kPa, p = 0.005) sides. There was a significant stiffness decrease from baseline to 6 months on the concave side in the surgery group (5.9 kPa, p = 0.025). However, the elasticity change recovered at 12 months without significant differences between the two groups.

Automatic image registration on intraoperative CBCT compared to Surface Matching registration on preoperative CT for spinal navigation: accuracy and workflow.

INTRODUCTION: Spinal navigation solutions have been slower to develop compared to cranial ones. To facilitate greater adoption and use of spinal navigation, the relatively cumbersome registration processes need to be improved upon. This study aims to validate a new solution for automatic image registration and compare it to a traditional Surface Matching method. METHOD: Adult patients undergoing spinal surgery requiring navigation were enrolled after providing consent. A registration matrix-Universal AIR (= Automatic Image Registration)-was placed in the surgical field and used for automatic registration based on intraoperative 3D imaging. A standard Surface Matching method was used for comparison. Accuracy measurements were obtained by comparing planned and acquired coordinates on the vertebrae. RESULTS: Thirty-nine patients with 42 datasets were included. The mean accuracy of Universal AIR registration was 1.20 ± 0.42 mm, while the mean accuracy of Surface Matching registration was 1.94 ± 0.64 mm. Universal AIR registration was non-inferior to Surface Matching registration. Post hoc analysis showed a significantly greater accuracy for Universal AIR registration. In Surface Matching, but not automatic registration, user-related errors such as incorrect identification of the vertebral level were seen. CONCLUSION: Automatic image registration for spinal navigation using Universal AIR and intraoperative 3D imaging provided improved accuracy compared to Surface Matching registration. In addition, it minimizes user errors and offers a standardized workflow, making it a reliable registration method for navigated spinal procedures.

A rare skeletal dysplasia in the etiology of severe scoliosis: Diaphanospondylodysostosis.

Diaphanospondylodysostosis is a rare genetic skeletal disorder caused by biallelic variants in the BMPER gene. The term, diaphanospondylodysostosis, includes ischiospinal dysotosis, which was previously known as a distinct entity with milder clinical features. The clinical phenotype of diaphanospondylodysostosis is quite variable with mortality in early postnatal life in some patients. Main clinical and radiographic features are narrow thorax, vertebral segmentation defects, rib anomalies, ossification defects of vertebrae, ischium and sacrum, and renal cysts. In this study, we report on a 14-year-old girl patient with diaphanospondylodysostosis harbouring a novel BMPER mutation. The patient presented with severe scoliosis and severely hypoplastic/aplastic distal phalanges of the fingers and toes, findings yet hitherto not described in this syndrome.

Image-Guided Spine Surgery.

The realm of spine surgery is undergoing a transformative shift, thanks to the integration of image-guided navigation technology. This innovative system seamlessly blends real-time imaging data with precise location tracking. While the indispensable expertise of experienced spine surgeons remains irreplaceable, navigation systems bring a host of valuable advantages to the operating room. By offering a comprehensive view of the surgical anatomy, these systems empower surgeons to conduct procedures with accuracy, while minimizing radiation exposure for both patients and medical professionals. Moreover, image-guided navigation paves the way for integration of other state-of-the-art technologies, such as augmented reality and robotics. These innovations promise to further revolutionize the field, providing greater precision and expanding the horizons of what is possible in the world of spinal procedures. This article explores the evolution, classification, and impact of image-guided spine surgery, underscoring its pivotal role in enhancing efficacy and safety while setting the stage for the incorporation of future technological advancements.

Three-dimensional spine reconstruction from biplane radiographs using convolutional neural networks.

PURPOSE: The purpose of this study was to develop and evaluate a deep learning network for three-dimensional reconstruction of the spine from biplanar radiographs. METHODS: The proposed approach focused on extracting similar features and multiscale features of bone tissue in biplanar radiographs. Bone tissue features were reconstructed for feature representation across dimensions to generate three-dimensional volumes. The number of feature mappings was gradually reduced in the reconstruction to transform the high-dimensional features into the three-dimensional image domain. We produced and made eight public datasets to train and test the proposed network. Two evaluation metrics were proposed and combined with four classical evaluation metrics to measure the performance of the method. RESULTS: In comparative experiments, the reconstruction results of this method achieved a Hausdorff distance of 1.85 mm, a surface overlap of 0.2 mm, a volume overlap of 0.9664, and an offset distance of only 0.21 mm from the vertebral body centroid. The results of this study indicate that the proposed method is reliable.

Instrumentation failure following pediatric spine deformity growth-sparing surgery using traditional growing rods or vertical expandable prosthetic titanium ribs.

BACKGROUND: Instrumentation failure (IF) is a major complication associated with growth-sparing surgery for pediatric spinal deformities; however, studies focusing on IF following each surgical procedure are lacking. We aimed to evaluate the incidence, timing, and rates of unplanned return to the operating room (UPROR) associated with IF following each surgical procedure in growth-sparing surgeries using traditional growing rods (TGRs) and vertical expandable prosthetic titanium ribs (VEPTRs). METHODS: We reviewed 1,139 surgical procedures documented in a Japanese multicenter database from 2015 to 2017. Of these, 544 TGR and 455 VEPTR procedures were included for evaluation on a per-surgery basis. IF was defined as the occurrence of an implant-related complication requiring revision surgery. RESULTS: The surgery-based incidences of IF requiring revision surgery in the TGR and VEPTR groups were 4.3% and 4.0%, respectively, with no significant intergroup difference. Remarkably, there was a negative correlation between IF incidence per surgical procedure and the number of lengthening surgeries in both groups. In addition, rod breakage in the TGR group and anchor-related complications in the VEPTR group tended to occur relatively early in the treatment course. The surgery-based rates of UPROR due to IF in the TGR and VEPTR groups were 2.0% and 1.5%, respectively, showing no statistically significant difference. CONCLUSIONS: We found that IF, such as anchor related-complications and rod breakage, occurs more frequently earlier in the course of lengthening surgeries. This finding may help in patient counseling and highlights the importance of close postoperative follow-up to detect IF and improve outcomes.

Top 25 Most Cited Articles on Intraoperative Computer Tomography-Guided Navigation in Spine Surgery.

BACKGROUND: In recent years, the use of intraoperative computer tomography-guided (CT-guided) navigation has gained significant popularity among health care providers who perform minimally invasive spine surgery. This review aims to identify and analyze trends in the literature related to the widespread adoption of CT-guided navigation in spine surgery, emphasizing the shift from conventional fluoroscopy-based techniques to CT-guided navigation. METHODS: Articles pertaining to this study were identified via a database review and were hierarchically organized based on the number of citations. An "advanced document search" was performed on September 28th, 2022, utilizing Boolean search operator terms. The 25 most referenced articles were combined into a primary list after sorting results in descending order based on the total number of citations. RESULTS: The "Top 25" list for intraoperative CT-guided navigation in spine surgery cumulatively received a total of 2742 citations, with an average of 12 new citations annually. The number of citations ranged from 246 for the most cited article to 60 for the 25th most cited article. The most cited article was a paper by Siewerdsen et al., with 246 total citations, averaging 15 new citations per year. CONCLUSIONS: Intraoperative CT-guided navigation is 1 of many technological advances that is used to increase surgical accuracy, and it has become an increasingly popular alternative to conventional fluoroscopy-based techniques. Given the increasing adoption of intraoperative CT-guided navigation in spine surgery, this review provides impactful evidence for its utility in spine surgery.

Spinal axial torque assessment after surgical correction in adolescent idiopathic scoliosis: a new approach to 3D barycentremetry and mass distribution based on biplanar radiographs.

PURPOSE: Barycentremetry in adolescent idiopathic scoliosis (AIS) allows the distribution of masses and their loading of the spine to be studied. In particular, the axial torque on the spine has been studied in AIS, but not after surgical correction. Spinal axial torque was studied in AIS before and after surgery. METHODS: All AIS (Lenke 1 and 3) who underwent posterior spinal fusion surgery at our center in 2019 were included retrospectively. AIS underwent frontal and sagittal biplanar radiographs in the free-standing position before surgery, 4 months after surgery, and at the last follow-up. Their spine and external envelope were reconstructed with validated methods. Spinal axial torque at the apex and the upper and lower end vertebra was calculated. Finally, the preoperative and postoperative values were compared to a previously published reference corridor for asymptomatic subjects. RESULTS: Twenty-nine patients were included (54 ± 11° Cobb angle, 15 ± 2 years old at surgery). The surgical procedure decreased the Cobb angle by 36° ± 11° and decreased the spinal axial torque at the upper end vertebra by 2.5 N/m (95% CI = [1.9; 3]; p < 0.001), at the apex by 0.6 N/m (95% CI = [0.4; 1]; p = 0.004), at the lower end vertebra by 2 N/m (95% CI = [1.5; 2.8]; p < 0.001). Compared to 95th percentile of torque, which was previously evaluated in asymptomatic subjects, more than 90% of patients had higher values at the upper and lower end vertebrae before surgery. Postoperatively, 62% of patients still had higher torque at the upper end vertebra than asymptomatic subjects, while only 38% patients showed abnormal values at the lower junction. CONCLUSION: Results of this study confirm that AIS patients show abnormally high spinal axial torque, especially at the end vertebrae, and that this parameter is normalized postoperatively for only a small number of patients.

Comparison of the effect of treatment with NSAIDs added to anti-TNF therapy versus anti-TNF therapy alone on the progression of structural damage in the spine over 2 years in patients with radiographic axial spondyloarthritis from the randomised-controlled CONSUL trial.

OBJECTIVES: The study aimed to evaluate the effect of adding a non-steroidal anti-inflammatory drug (NSAID), celecoxib (CEL), to a tumour necrosis factor inhibitor (TNFi), golimumab (GOL), compared with TNFi monotherapy on radiographic spinal progression in patients with radiographic axial spondyloarthritis (r-axSpA) over 2 years. METHODS: R-axSpA patients, having risk factors for radiographic progression (high disease activity plus C reactive protein >5 mg/L and/or ≥1 syndesmophyte(s)), underwent a 12-week run-in phase with GOL 50 mg every 4 weeks. In the core phase (96 weeks), only patients with a good clinical response at week 12 were randomised (1:1) to GOL+CEL 200 mg two times per day (combination therapy) or GOL monotherapy. The primary endpoint was radiographic progression assessed by modified Stoke Ankylosing Spondylitis Spinal Score (mSASSS) change at week 108 in the intent-to-treat population. RESULTS: A total of 128 patients were enrolled in the run-in phase; and 109 patients were randomised at week 12 to monotherapy (n=55) or combination therapy (n=54). At week 108, 97 (52 vs 45) patients completed the study. The change in mSASSS at week 108 was 1.7 (95% CI 0.8 to 2.6) in the monotherapy vs 1.1 (95% CI 0.4 to 1.8) in the combination therapy groups (p=0.79). New syndesmophytes occurred in 25% of patients in the monotherapy vs 11% of patients in the combination therapy groups (p=0.12). During the study, no significant differences in adverse events and serious adverse events were observed between the groups. CONCLUSIONS: Combination therapy with GOL+CEL did not demonstrate statistically significant superiority over GOL monotherapy in retarding radiographic spinal progression over 2 years in r-axSpA.

Total spine MRI for the preoperative evaluation of adolescent idiopathic scoliosis: part 1.

Adolescent idiopathic scoliosis is a commonly encountered condition often diagnosed on screening examination. Underlying, asymptomatic neural axis abnormalities may be present at the time of diagnosis. At certain institutions, total spine MRI is obtained preoperatively to identify these abnormalities. We provide a framework for the radiologist to follow while interpreting these studies. In part 1, we discuss Arnold Chiari malformations, syringomyelia, and the tethered cord. In part 2, we focus on spinal cord tumors, dysraphisms, to include diastematomyelia, and vertebral anomalies.

Intraobserver and Interobserver Reliability of Pelvic Obliquity Measurement Methods in Patients With Idiopathic Scoliosis.

BACKGROUND: Despite the importance of the assessment of pelvic obliquity, especially in the planning of surgery, there is no consensus on the pelvic obliquity measurements. The purpose of this study is to assess the intraobserver and interobserver reliability of 4 different pelvic obliquity measuring methods in patients with idiopathic scoliosis (IS): the Osebold, O'Brien, Maloney, and Allen&Ferguson methods. METHODS: A retrospective cohort of 85 posteroanterior full-spine radiographs in the standing position of patients with IS involving the pelvic obliquity was evaluated by a team of 3 raters. The same researcher recorded the curve magnitude, apical vertebral rotation, Risser grade, curve pattern, and femoral head height difference. The pelvic obliquity angle was measured using the Osebold, O'Brien, Maloney, and Allen&Ferguson methods. The same graders were asked to regrade the same radiographs after at least 1 month. RESULTS: The Osebold method showed the highest interobserver reliability with an ICC of 0.994 and 0.983. The Allen&Ferguson method had the lowest reliability with an ICC of 0.911 and 0.934, but all of the methods were considered having excellent reliability. The Osebold method also showed the highest intraobserver reliability, ranging from 0.909 to 0.997. The Allen&Ferguson method had the lowest intraobserver reliability, with a range of 0.741 to 0.960. Also, all observers preferred the Osebold Method. The observers reported that the Allen&Ferguson method was considered the most time-consuming method, while the least time-consuming method was specified as the Osebold method. CONCLUSIONS: To the best of our knowledge, this study is the first to evaluate the intraobserver and interobserver reliability of pelvic obliquity among common measurement methods in patients with idiopathic scoliosis. All methods were evaluated as having excellent to good reliability in this study, but the Osebold method is the most reliable method of measuring pelvic obliquity on a frontal view radiograph in idiopathic scoliosis. The Osebold method is easier to use as it requires only the iliac crests to be visualized. LEVEL OF EVIDENCE: Level III-Retrospective cohort study.

Patient-specific rods in adult spinal deformity: a systematic review.

PURPOSE: The purpose of this review was to evaluate the effectiveness of patient-specific rods for adult spinal deformity. METHODS: A systematic review of the literature was performed through an electronic search of the PubMed, Scopus, and Web of Science databases. Human studies between 2012 and 2023 were included. Sample size, sagittal vertical axis (SVA), pelvic incidence-lumbar lordosis (PI-LL), pelvic tilt (PT), operation time, blood loss, follow-up duration, and complications were recorded for each study when available. RESULTS: Seven studies with a total of 304 adult spinal deformity patients of various etiologies were included. All studies reported SVA, and PT; two studies did not report PI-LL. Four studies reported planned radiographic outcomes. Two found a significant association between preoperative plan and postoperative outcome in all three outcomes. One found a significant association for PI-LL alone. The fourth found no significant associations. SVA improved in six of seven studies, PI-LL improved in all five, and three of seven studies found improved postoperative PT. Significance of these results varied greatly by study. CONCLUSION: Preliminary evidence suggests potential benefits of PSRs in achieving optimal spino-pelvic parameters in ASD surgery. Nevertheless, conclusions regarding the superiority of PSRs over traditional rods must be judiciously drawn, given the heterogeneity of patients and study methodologies, potential confounding variables, and the absence of robust randomized controlled trials. Future investigations should concentrate on enhancing preoperative planning, standardizing surgical methodologies, isolating specific patient subgroups, and head-to-head comparisons with traditional rods to fully elucidate the impact of PSRs in ASD surgery.

Vertebral Venous Congestion That May Mimic Vertebral Metastasis on Contrast-Enhanced Chest Computed Tomography in Chemoport Inserted Patients.

OBJECTIVE: This study aimed to determine the prevalence of vertebral venous congestion (VVC) in patients with chemoport insertion, evaluate the imaging characteristics of nodular VVC, and identify the factors associated with VVC. MATERIALS AND METHODS: This retrospective single-center study was based on follow-up contrast-enhanced chest computed tomography (CT) of 1412 adult patients who underwent chemoport insertion between January 2016 and December 2016. The prevalence of venous stenosis, reflux, and VVC were evaluated. The imaging features of nodular VVC, including specific locations within the vertebral body, were analyzed. To identify the factors associated with VVC, patients with VVC were compared with a subset of patients without VVC who had been followed up for > 3 years without developing VVC after chemoport insertion. Toward this, a multivariable logistic regression analysis was performed. RESULTS: After excluding 333 patients, 1079 were analyzed (mean age ± standard deviation, 62.3 ± 11.6 years; 540 females). The prevalence of VVC was 5.8% (63/1079), with all patients (63/63) demonstrating vertebral venous reflux and 67% (42/63) with innominate vein stenosis. The median interval between chemoport insertion and VVC was 515 days (interquartile range, 204-881 days). The prevalence of nodular VVC was 1.5% (16/1079), with a mean size of 5.9 ± 3.1 mm and attenuation of 784 ± 162 HU. Nodular VVC tended to be located subcortically. Forty-four patients with VVC underwent CT examinations with contrast injections in both arms; the VVC disappeared in 70% (31/44) when the contrast was injected in the arm contralateral to the chemoport site. Bevacizumab use was independently associated with VVC (odds ratio, 3.45; P < 0.001). CONCLUSION: The prevalence of VVC and nodular VVC was low in patients who underwent chemoport insertion. Nodular VVC was always accompanied by vertebral venous reflux and tended to be located subcortically. To avoid VVC, contrast injection in the arm contralateral to the chemoport site is preferred.