Musculoskeletal spine modeling in large patient cohorts: how morphological individualization affects lumbar load estimation.

Introduction: Achieving an adequate level of detail is a crucial part of any modeling process. Thus, oversimplification of complex systems can lead to overestimation, underestimation, and general bias of effects, while elaborate models run the risk of losing validity due to the uncontrolled interaction of multiple influencing factors and error propagation. Methods: We used a validated pipeline for the automated generation of multi-body models of the trunk to create 279 models based on CT data from 93 patients to investigate how different degrees of individualization affect the observed effects of different morphological characteristics on lumbar loads. Specifically, individual parameters related to spinal morphology (thoracic kyphosis (TK), lumbar lordosis (LL), and torso height (TH)), as well as torso weight (TW) and distribution, were fully or partly considered in the respective models according to their degree of individualization, and the effect strengths of these parameters on spinal loading were compared between semi- and highly individualized models. T-distributed stochastic neighbor embedding (T-SNE) analysis was performed for overarching pattern recognition and multiple regression analyses to evaluate changes in occurring effects and significance. Results: We were able to identify significant effects (p < 0.05) of various morphological parameters on lumbar loads in models with different degrees of individualization. Torso weight and lumbar lordosis showed the strongest effects on compression (ß ≈ 0.9) and anterior-posterior shear forces (ß ≈ 0.7), respectively. We could further show that the effect strength of individual parameters tended to decrease if more individual characteristics were included in the models. Discussion: The induced variability due to model individualization could only partly be explained by simple morphological parameters. Our study shows that model simplification can lead to an emphasis on individual effects, which needs to be critically assessed with regard to in vivo complexity. At the same time, we demonstrated that individualized models representing a population-based cohort are still able to identify relevant influences on spinal loading while considering a variety of influencing factors and their interactions.

Deep Learning to Differentiate Benign and Malignant Vertebral Fractures at Multidetector CT.

Background Differentiating between benign and malignant vertebral fractures poses diagnostic challenges. Purpose To investigate the reliability of CT-based deep learning models to differentiate between benign and malignant vertebral fractures. Materials and Methods CT scans acquired in patients with benign or malignant vertebral fractures from June 2005 to December 2022 at two university hospitals were retrospectively identified based on a composite reference standard that included histopathologic and radiologic information. An internal test set was randomly selected, and an external test set was obtained from an additional hospital. Models used a three-dimensional U-Net encoder-classifier architecture and applied data augmentation during training. Performance was evaluated using the area under the receiver operating characteristic curve (AUC) and compared with that of two residents and one fellowship-trained radiologist using the DeLong test. Results The training set included 381 patients (mean age, 69.9 years ± 11.4 [SD]; 193 male) with 1307 vertebrae (378 benign fractures, 447 malignant fractures, 482 malignant lesions). Internal and external test sets included 86 (mean age, 66.9 years ± 12; 45 male) and 65 (mean age, 68.8 years ± 12.5; 39 female) patients, respectively. The better-performing model of two training approaches achieved AUCs of 0.85 (95% CI: 0.77, 0.92) in the internal and 0.75 (95% CI: 0.64, 0.85) in the external test sets. Including an uncertainty category further improved performance to AUCs of 0.91 (95% CI: 0.83, 0.97) in the internal test set and 0.76 (95% CI: 0.64, 0.88) in the external test set. The AUC values of residents were lower than that of the best-performing model in the internal test set (AUC, 0.69 [95% CI: 0.59, 0.78] and 0.71 [95% CI: 0.61, 0.80]) and external test set (AUC, 0.70 [95% CI: 0.58, 0.80] and 0.71 [95% CI: 0.60, 0.82]), with significant differences only for the internal test set (P < .001). The AUCs of the fellowship-trained radiologist were similar to those of the best-performing model (internal test set, 0.86 [95% CI: 0.78, 0.93; P = .39]; external test set, 0.71 [95% CI: 0.60, 0.82; P = .46]). Conclusion Developed models showed a high discriminatory power to differentiate between benign and malignant vertebral fractures, surpassing or matching the performance of radiology residents and matching that of a fellowship-trained radiologist. © RSNA, 2024 See also the editorial by Booz and D'Angelo in this issue.

Incidental vertebral fracture prediction using neuronal network-based automatic spine segmentation and volumetric bone mineral density extraction from routine clinical CT scans.

Objectives: To investigate vertebral osteoporotic fracture (VF) prediction by automatically extracted trabecular volumetric bone mineral density (vBMD) from routine CT, and to compare the model with fracture prevalence-based prediction models. Methods: This single-center retrospective study included patients who underwent two thoraco-abdominal CT scans during clinical routine with an average inter-scan interval of 21.7 ± 13.1 months (range 5-52 months). Automatic spine segmentation and vBMD extraction was performed by a convolutional neural network framework (anduin.bonescreen.de). Mean vBMD was calculated for levels T5-8, T9-12, and L1-5. VFs were identified by an expert in spine imaging. Odds ratios (ORs) for prevalent and incident VFs were calculated for vBMD (per standard deviation decrease) at each level, for baseline VF prevalence (yes/no), and for baseline VF count (n) using logistic regression models, adjusted for age and sex. Models were compared using Akaike's and Bayesian information criteria (AIC & BIC). Results: 420 patients (mean age, 63 years ± 9, 276 males) were included in this study. 40 (25 female) had prevalent and 24 (13 female) had incident VFs. Individuals with lower vBMD at any spine level had higher odds for VFs (L1-5, prevalent VF: OR,95%-CI,p: 2.2, 1.4-3.5,p=0.001; incident VF: 3.5, 1.8-6.9,p<0.001). In contrast, VF status (2.15, 0.72-6.43,p=0.170) and count (1.38, 0.89-2.12,p=0.147) performed worse in incident VF prediction. Information criteria revealed best fit for vBMD-based models (AIC vBMD=165.2; VF status=181.0; count=180.7). Conclusions: VF prediction based on automatically extracted vBMD from routine clinical MDCT outperforms prediction models based on VF status and count. These findings underline the importance of opportunistic quantitative osteoporosis screening in clinical routine MDCT data.

Prognostic value of spinal cord lesion measures in early relapsing-remitting multiple sclerosis.

BACKGROUND: Spinal cord (SC) lesions have been associated with unfavourable clinical outcomes in multiple sclerosis (MS). However, the relation of whole SC lesion number (SCLN) and volume (SCLV) to the future occurrence and type of confirmed disability accumulation (CDA) remains largely unexplored. METHODS: In this monocentric retrospective study, SC lesions were manually delineated. Inclusion criteria were: age between 18 and 60 years, relapsing-remitting MS, disease duration under 2 years and clinical follow-up of 5 years. The first CDA event after baseline, determined by a sustained increase in the Expanded Disability Status Scale over 6 months, was classified as either progression independent of relapse activity (PIRA) or relapse-associated worsening (RAW). SCLN and SCLV were compared between different (sub)groups to assess their prospective value. RESULTS: 204 patients were included, 148 of which had at least one SC lesion and 59 experienced CDA. Patients without any SC lesions experienced significantly less CDA (OR 5.8, 95% CI 2.1 to 19.8). SCLN and SCLV were closely correlated (rs=0.91, p<0.001) and were both significantly associated with CDA on follow-up (p<0.001). Subgroup analyses confirmed this association for patients with PIRA on CDA (34 events, p<0.001 for both SC lesion measures) but not for RAW (25 events, p=0.077 and p=0.22). CONCLUSION: Patients without any SC lesions are notably less likely to experience CDA. Both the number and volume of SC lesions on MRI are associated with future accumulation of disability largely independent of relapses.

Synthetic T2-weighted fat sat based on a generative adversarial network shows potential for scan time reduction in spine imaging in a multicenter test dataset.

OBJECTIVES: T2-weighted (w) fat sat (fs) sequences, which are important in spine MRI, require a significant amount of scan time. Generative adversarial networks (GANs) can generate synthetic T2-w fs images. We evaluated the potential of synthetic T2-w fs images by comparing them to their true counterpart regarding image and fat saturation quality, and diagnostic agreement in a heterogenous, multicenter dataset. METHODS: A GAN was used to synthesize T2-w fs from T1- and non-fs T2-w. The training dataset comprised scans of 73 patients from two scanners, and the test dataset, scans of 101 patients from 38 multicenter scanners. Apparent signal- and contrast-to-noise ratios (aSNR/aCNR) were measured in true and synthetic T2-w fs. Two neuroradiologists graded image (5-point scale) and fat saturation quality (3-point scale). To evaluate whether the T2-w fs images are indistinguishable, a Turing test was performed by eleven neuroradiologists. Six pathologies were graded on the synthetic protocol (with synthetic T2-w fs) and the original protocol (with true T2-w fs) by the two neuroradiologists. RESULTS: aSNR and aCNR were not significantly different between the synthetic and true T2-w fs images. Subjective image quality was graded higher for synthetic T2-w fs (p = 0.023). In the Turing test, synthetic and true T2-w fs could not be distinguished from each other. The intermethod agreement between synthetic and original protocol ranged from substantial to almost perfect agreement for the evaluated pathologies. DISCUSSION: The synthetic T2-w fs might replace a physical T2-w fs. Our approach validated on a challenging, multicenter dataset is highly generalizable and allows for shorter scan protocols. KEY POINTS: • Generative adversarial networks can be used to generate synthetic T2-weighted fat sat images from T1- and non-fat sat T2-weighted images of the spine. • The synthetic T2-weighted fat sat images might replace a physically acquired T2-weighted fat sat showing a better image quality and excellent diagnostic agreement with the true T2-weighted fat images. • The present approach validated on a challenging, multicenter dataset is highly generalizable and allows for significantly shorter scan protocols.

Lesion location across diagnostic regions in multiple sclerosis.

BACKGROUND: Lesions in the periventricular, (juxta)cortical, and infratentorial region, as visible on brain MRI, are part of the diagnostic criteria for Multiple sclerosis (MS) whereas lesions in the subcortical region are currently only a marker of disease activity. It is unknown whether MS lesions follow individual spatial patterns or whether they occur in a random manner across diagnostic regions. AIM: First, to describe cross-sectionally the spatial lesion patterns in patients with MS. Second, to investigate the spatial association of new lesions and lesions at baseline across diagnostic regions. METHODS: Experienced neuroradiologists analyzed brain MRI (3D, 3T) in a cohort of 330 early MS patients. Lesions at baseline and new solitary lesions after two years were segmented (manually and by consensus) and classified as periventricular, (juxta)cortical, or infratentorial (diagnostic regions) or subcortical-with or without Gadolinium-enhancement. Gadolinium enhancement of lesions in the different regions was compared by Chi square test. New lesions in the four regions served as dependent variable in four zero-inflated Poisson models each with the six independent variables of lesions in the four regions at baseline, age and gender. RESULTS: At baseline, lesions were most often observed in the subcortical region (mean 13.0 lesions/patient), while lesion volume was highest in the periventricular region (mean 2287 µl/patient). Subcortical lesions were less likely to show gadolinium enhancement (3.1 %) than juxtacortical (4.3 %), periventricular (5.3 %) or infratentorial lesions (7.2 %). Age was inversely correlated with new periventricular, juxtacortical and subcortical lesions. New lesions in the periventricular, juxtacortical and infratentorial region showed a significant autocorrelative behavior being positively related to the number of lesions in the respective regions at baseline. New lesions in the subcortical region showed a different behavior with a positive association with baseline periventricular lesions and a negative association with baseline infratentorial lesions. CONCLUSION: Across regions, new lesions do not occur randomly; instead, new lesions in the periventricular, juxtacortical and infratentorial diagnostic region are associated with that at baseline. Lesions in the subcortical regions are more closely related to periventricular lesions. Moreover, subcortical lesions substantially contribute to lesion burden in MS but are less likely to show gadolinium enhancement (than lesions in the diagnostic regions).

Somatosensory evoked potentials and magnetic resonance imaging of the central nervous system in early multiple sclerosis.

BACKGROUND: Somatosensory evoked potentials (SSEP) are still broadly used, although not explicitly recommended, for the diagnostic work-up of suspected multiple sclerosis (MS). OBJECTIVE: To relate disability, SSEP, and lesions on T2-weighted magnetic resonance imaging (MRI) in patients with early MS. METHODS: In this monocentric retrospective study, we analyzed a cohort of patients with relapsing-remitting MS or clinically isolated syndrome, with a maximum disease duration of two years, as well as with available data on the score at the expanded disability status scale (EDSS), on SSEP, on whole spinal cord (SC) MRI, and on brain MRI. RESULTS: Complete data of 161 patients were available. Tibial nerve SSEP (tSSEP) were less frequently abnormal than SC MRI (22% vs. 68%, p < 0.001). However, higher EDSS scores were significantly associated with abnormal tSSEP (median, 2.0 vs. 1.0; p = 0.001) but not with abnormal SC MRI (i.e., at least one lesion; median, 1.5 vs. 1.5; p = 0.7). Of the 35 patients with abnormal tSSEP, 32 had lesions on SC MRI, and 2 had corresponding lesions on brain MRI. CONCLUSION: Compared to tSSEP, SC MRI is the more sensitive diagnostic biomarker regarding SC involvement. In early MS, lesions as detectable by T2-weighted MRI are the main driver of abnormal tSSEP. However, tSSEP were more closely associated with disability, which is compatible with a potential role of tSSEP as prognostic biomarker in complementation of MRI.

Validation of a Patient-Specific Musculoskeletal Model for Lumbar Load Estimation Generated by an Automated Pipeline From Whole Body CT.

Background: Chronic back pain is a major health problem worldwide. Although its causes can be diverse, biomechanical factors leading to spinal degeneration are considered a central issue. Numerical biomechanical models can identify critical factors and, thus, help predict impending spinal degeneration. However, spinal biomechanics are subject to significant interindividual variations. Therefore, in order to achieve meaningful findings on potential pathologies, predictive models have to take into account individual characteristics. To make these highly individualized models suitable for systematic studies on spinal biomechanics and clinical practice, the automation of data processing and modeling itself is inevitable. The purpose of this study was to validate an automatically generated patient-specific musculoskeletal model of the spine simulating static loading tasks. Methods: CT imaging data from two patients with non-degenerative spines were processed using an automated deep learning-based segmentation pipeline. In a semi-automated process with minimal user interaction, we generated patient-specific musculoskeletal models and simulated various static loading tasks. To validate the model, calculated vertebral loadings of the lumbar spine and muscle forces were compared with in vivo data from the literature. Finally, results from both models were compared to assess the potential of our process for interindividual analysis. Results: Calculated vertebral loads and muscle activation overall stood in close correlation with data from the literature. Compression forces normalized to upright standing deviated by a maximum of 16% for flexion and 33% for lifting tasks. Interindividual comparison of compression, as well as lateral and anterior-posterior shear forces, could be linked plausibly to individual spinal alignment and bodyweight. Conclusion: We developed a method to generate patient-specific musculoskeletal models of the lumbar spine. The models were able to calculate loads of the lumbar spine for static activities with respect to individual biomechanical properties, such as spinal alignment, bodyweight distribution, and ligament and muscle insertion points. The process is automated to a large extent, which makes it suitable for systematic investigation of spinal biomechanics in large datasets.

Level-Specific Volumetric BMD Threshold Values for the Prediction of Incident Vertebral Fractures Using Opportunistic QCT: A Case-Control Study.

Purpose: To establish and evaluate the diagnostic accuracy of volumetric bone mineral density (vBMD) threshold values at different spinal levels, derived from opportunistic quantitative computed tomography (QCT), for the prediction of incident vertebral fractures (VF). Materials and Methods: In this case-control study, 35 incident VF cases (23 women, 12 men; mean age: 67 years) and 70 sex- and age-matched controls were included, based on routine multi detector CT (MDCT) scans of the thoracolumbar spine. Trabecular vBMD was measured from routine baseline CT scans of the thoracolumbar spine using an automated pipeline including vertebral segmentation, asynchronous calibration for HU-to-vBMD conversion, and correction of intravenous contrast medium (https://anduin.bonescreen.de). Threshold values at T1-L5 were calculated for the optimal operating point according to the Youden index and for fixed sensitivities (60 - 85%) in receiver operating characteristic (ROC) curves. Results: vBMD at each single level of the thoracolumbar spine was significantly associated with incident VFs (odds ratio per SD decrease [OR], 95% confidence interval [CI] at T1-T4: 3.28, 1.66-6.49; at T5-T8: 3.28, 1.72-6.26; at T9-T12: 3.37, 1.78-6.36; and at L1-L4: 3.98, 1.97-8.06), independent of adjustment for age, sex, and prevalent VF. AUC showed no significant difference between vertebral levels and was highest at the thoracolumbar junction (AUC = 0.75, 95%-CI = 0.63 - 0.85 for T11-L2). Optimal threshold values increased from lumbar (L1-L4: 52.0 mg/cm³) to upper thoracic spine (T1-T4: 69.3 mg/cm³). At T11-L2, T12-L3 and L1-L4, a threshold of 80.0 mg/cm³ showed sensitivities of 85 - 88%, and specificities of 41 - 49%. To achieve comparable sensitivity (85%) at more superior spinal levels, resulting thresholds were higher: 114.1 mg/cm³ (T1-T4), 92.0 mg/cm³ (T5-T8), 88.2 mg/cm³ (T9-T12). Conclusions: At all levels of the thoracolumbar spine, lower vBMD was associated with incident VFs in an elderly, predominantly oncologic patient population. Automated opportunistic osteoporosis screening of vBMD along the entire thoracolumbar spine allows for risk assessment of imminent VFs. We propose level-specific vBMD threshold at the thoracolumbar spine to identify individuals at high fracture risk.

Automated Opportunistic Osteoporosis Screening in Routine Computed Tomography of the Spine: Comparison With Dedicated Quantitative CT.

Opportunistic osteoporosis screening in nondedicated routine computed tomography (CT) is of increasing importance. The purpose of this study was to compare lumbar volumetric bone mineral density (vBMD) assessed by a convolutional neural network (CNN)-based framework in routine CT to vBMD from dedicated quantitative CT (QCT), and to evaluate the ability of vBMD and surrogate measurements of Hounsfield units (HU) to distinguish between patients with and without osteoporotic vertebral fractures (VFs). A total of 144 patients (median age: 70.7 years, 93 females) with clinical routine CT (eight different CT scanners, 120 kVp or 140 kVp, with and without intravenous contrast medium) and dedicated QCT acquired within ≤30 days were included. Vertebral measurements included (i) vBMD from the CNN-based approach including automated vertebral body labeling, segmentation, and correction of the contrast media phase for routine CT data (vBMD_OPP), (ii) vBMD from dedicated QCT (vBMD_QCT), and (iii) noncalibrated HU from vertebral bodies of routine CT data as previously proposed for immanent opportunistic osteoporosis screening based on CT attenuation. The intraclass correlation coefficient (ICC) for vBMD_QCT versus vBMD_OPP indicated better agreement (ICC = 0.913) than the ICC for vBMD_QCT versus noncalibrated HU (ICC = 0.704). Bland-Altman analysis showed data points from 137 patients (95.1%) within the limits of agreement (LOA) of -23.2 to 25.0 mg/cm3 for vBMD_QCT versus vBMD_OPP. Osteoporosis (vBMD <80 mg/cm3 ) was detected in 89 patients (vBMD_QCT) and 88 patients (vBMD_OPP), whereas no patient crossed the diagnostic thresholds from normal vBMD to osteoporosis or vice versa. In a subcohort of 88 patients (thoracolumbar spine covered by imaging for VF reading), 69 patients showed one or more prevalent VFs, and the performance for discrimination between patients with and without VFs was best for vBMD_OPP (area under the curve [AUC] = 0.862; 95% confidence interval [CI], 0.771-0.953). In conclusion, automated opportunistic osteoporosis screening in routine CT of various scanner setups is feasible and may demonstrate high diagnostic accuracy for prevalent VFs. © 2022 The Authors. Journal of Bone and Mineral Research published by Wiley Periodicals LLC on behalf of American Society for Bone and Mineral Research (ASBMR).

Multiple sclerosis lesions and atrophy in the spinal cord: Distribution across vertebral levels and correlation with disability.

BACKGROUND: The vast majority of magnetic resonance imaging (MRI) studies on multiple sclerosis (MS) covered the spinal cord (SC), if at all, incompletely. OBJECTIVE: To assess SC involvement in MS, as detectable by whole SC MRI, with regard to distribution across vertebral levels and relation to clinical phenotypes and disability. METHODS: We investigated SC MRI with sagittal and axial coverage. Analyzed were brain and SC MRI scans of 17 healthy controls (HC) and of 370 patients with either clinically isolated syndrome (CIS, 27), relapsing remitting MS (RRMS, 303) or progressive MS (PMS, 40). Across vertebral levels, cross-sectional areas were semiautomatically segmented, and lesions manually delineated. RESULTS: The frequency of SC lesions was highest at the level C3-4. The volume of SC lesions increased from CIS to RRMS, and from RRMS to PMS whereas lesion distribution across SC levels did not differ. SC atrophy was demonstrated in RRMS and, to a higher degree, in PMS; apart from an accentuation at the level C3-4, it was evenly distributed across SC levels. SC lesions and atrophy volume were not correlated with each other and were independently associated with disability. CONCLUSION: SC lesions and atrophy already exist at the stage of RRMS in the whole SC with an accentuation in the cervical enlargement; SC lesions and atrophy are more pronounced in the stage of PMS. Both contribute to the clinical picture but are largely independent.

Proposed diagnostic volumetric bone mineral density thresholds for osteoporosis and osteopenia at the cervicothoracic spine in correlation to the lumbar spine.

OBJECTIVES: To determine the correlation between cervicothoracic and lumbar volumetric bone mineral density (vBMD) in an average cohort of adults and to identify specific diagnostic thresholds for the cervicothoracic spine on the individual subject level. METHODS: In this HIPPA-compliant study, we retrospectively included 260 patients (59.7 ± 18.3 years, 105 women), who received a contrast-enhanced or non-contrast-enhanced CT scan. vBMD was extracted using an automated pipeline ( https://anduin.bonescreen.de ). The association of vBMD between each vertebra spanning C2-T12 and the averaged values at the lumbar spine (L1-L3) was analyzed before and after semiquantitative assessment of fracture status and degeneration, and respective vertebra-specific cut-off values for osteoporosis were calculated using linear regression. RESULTS: In both women and men, trabecular vBMD decreased with age in the cervical, thoracic, and lumbar regions. vBMD values of cervicothoracic vertebrae showed strong correlations with lumbar vertebrae (L1-L3), with a median Pearson value of r = 0.87 (range: rC2 = 0.76 to rT12 = 0.96). The correlation coefficients were significantly lower (p < 0.0001) without excluding fractured and degenerated vertebrae, median r = 0.82 (range: rC2 = 0.69 to rT12 = 0.93). Respective cut-off values for osteoporosis peaked at C4 (209.2 mg/ml) and decreased to 83.8 mg/ml at T12. CONCLUSION: Our data show a high correlation between clinically used mean L1-L3 values and vBMD values elsewhere in the spine, independent of age. The proposed cut-off values for the cervicothoracic spine therefore may allow the determination of low bone mass even in clinical cases where only parts of the spine are imaged. KEY POINTS: vBMD of all cervicothoracic vertebrae showed strong correlation with lumbar vertebrae (L1-L3), with a median Pearson's correlation coefficient of r = 0.87 (range: rC2 = 0.76 to rT12 = 0.96). The correlation coefficients were significantly lower (p < 0.0001) without excluding fractured and moderate to severely degenerated vertebrae, median r = 0.82 (range: rC2 = 0.69 to rT12 = 0.93). We postulate that trabecular vBMD < 200 mg/ml for the cervical spine and < 100 mg/ml for the thoracic spine are strong indicators of osteoporosis, similar to < 80 mg/ml at the lumbar spine.

Automated detection of the contrast phase in MDCT by an artificial neural network improves the accuracy of opportunistic bone mineral density measurements.

OBJECTIVES: To determine the accuracy of an artificial neural network (ANN) for fully automated detection of the presence and phase of iodinated contrast agent in routine abdominal multidetector computed tomography (MDCT) scans and evaluate the effect of contrast correction for osteoporosis screening. METHODS: This HIPPA-compliant study retrospectively included 579 MDCT scans in 193 patients (62.4 ± 14.6 years, 48 women). Three different ANN models (2D DenseNet with random slice selection, 2D DenseNet with anatomy-guided slice selection, 3D DenseNet) were trained in 462 MDCT scans of 154 patients (threefold cross-validation), who underwent triphasic CT. All ANN models were tested in 117 unseen triphasic scans of 39 patients, as well as in a public MDCT dataset containing 311 patients. In the triphasic test scans, trabecular volumetric bone mineral density (BMD) was calculated using a fully automated pipeline. Root-mean-square errors (RMSE) of BMD measurements with and without correction for contrast application were calculated in comparison to nonenhanced (NE) scans. RESULTS: The 2D DenseNet with anatomy-guided slice selection outperformed the competing models and achieved an F1 score of 0.98 and an accuracy of 98.3% in the test set (public dataset: F1 score 0.93; accuracy 94.2%). Application of contrast agent resulted in significant BMD biases (all p < .001; portal-venous (PV): RMSE 18.7 mg/ml, mean difference 17.5 mg/ml; arterial (AR): RMSE 6.92 mg/ml, mean difference 5.68 mg/ml). After the fully automated correction, this bias was no longer significant (p > .05; PV: RMSE 9.45 mg/ml, mean difference 1.28 mg/ml; AR: RMSE 3.98 mg/ml, mean difference 0.94 mg/ml). CONCLUSION: Automatic detection of the contrast phase in multicenter CT data was achieved with high accuracy, minimizing the contrast-induced error in BMD measurements. KEY POINTS: • A 2D DenseNet with anatomy-guided slice selection achieved an F1 score of 0.98 and an accuracy of 98.3% in the test set. In a public dataset, an F1 score of 0.93 and an accuracy of 94.2% were obtained. • Automated adjustment for contrast injection improved the accuracy of lumbar bone mineral density measurements (RMSE 18.7 mg/ml vs. 9.45 mg/ml respectively, in the portal-venous phase). • An artificial neural network can reliably reveal the presence and phase of iodinated contrast agent in multidetector CT scans ( https://github.com/ferchonavarro/anatomy_guided_contrast_c ). This allows minimizing the contrast-induced error in opportunistic bone mineral density measurements.

A computed tomography vertebral segmentation dataset with anatomical variations and multi-vendor scanner data.

With the advent of deep learning algorithms, fully automated radiological image analysis is within reach. In spine imaging, several atlas- and shape-based as well as deep learning segmentation algorithms have been proposed, allowing for subsequent automated analysis of morphology and pathology. The first "Large Scale Vertebrae Segmentation Challenge" (VerSe 2019) showed that these perform well on normal anatomy, but fail in variants not frequently present in the training dataset. Building on that experience, we report on the largely increased VerSe 2020 dataset and results from the second iteration of the VerSe challenge (MICCAI 2020, Lima, Peru). VerSe 2020 comprises annotated spine computed tomography (CT) images from 300 subjects with 4142 fully visualized and annotated vertebrae, collected across multiple centres from four different scanner manufacturers, enriched with cases that exhibit anatomical variants such as enumeration abnormalities (n = 77) and transitional vertebrae (n = 161). Metadata includes vertebral labelling information, voxel-level segmentation masks obtained with a human-machine hybrid algorithm and anatomical ratings, to enable the development and benchmarking of robust and accurate segmentation algorithms.

Automatic opportunistic osteoporosis screening in routine CT: improved prediction of patients with prevalent vertebral fractures compared to DXA.

OBJECTIVES: To compare spinal bone measures derived from automatic and manual assessment in routine CT with dual energy X-ray absorptiometry (DXA) in their association with prevalent osteoporotic vertebral fractures using our fully automated framework ( https://anduin.bonescreen.de ) to assess various bone measures in clinical CT. METHODS: We included 192 patients (141 women, 51 men; age 70.2 ± 9.7 years) who had lumbar DXA and CT available (within 1 year). Automatic assessment of spinal bone measures in CT included segmentation of vertebrae using a convolutional neural network (CNN), reduction to the vertebral body, and extraction of bone mineral content (BMC), trabecular and integral volumetric bone mineral density (vBMD), and CT-based areal BMD (aBMD) using asynchronous calibration. Moreover, trabecular bone was manually sampled (manual vBMD). RESULTS: A total of 148 patients (77%) had vertebral fractures and significantly lower values in all bone measures compared to patients without fractures (p ≤ 0.001). Except for BMC, all CT-based measures performed significantly better as predictors for vertebral fractures compared to DXA (e.g., AUC = 0.885 for trabecular vBMD and AUC = 0.86 for integral vBMD vs. AUC = 0.668 for DXA aBMD, respectively; both p < 0.001). Age- and sex-adjusted associations with fracture status were strongest for manual vBMD (OR = 7.3, [95%] CI 3.8-14.3) followed by automatically assessed trabecular vBMD (OR = 6.9, CI 3.5-13.4) and integral vBMD (OR = 4.3, CI 2.5-7.6). Diagnostic cutoffs of integral vBMD for osteoporosis (< 160 mg/cm3) or low bone mass (160 ≤ BMD < 190 mg/cm3) had sensitivity (84%/41%) and specificity (78%/95%) similar to trabecular vBMD. CONCLUSIONS: Fully automatic osteoporosis screening in routine CT of the spine is feasible. CT-based measures can better identify individuals with reduced bone mass who suffered from vertebral fractures than DXA. KEY POINTS: • Opportunistic osteoporosis screening of spinal bone measures derived from clinical routine CT is feasible in a fully automatic fashion using a deep learning-driven framework ( https://anduin.bonescreen.de ). • Manually sampled volumetric BMD (vBMD) and automatically assessed trabecular and integral vBMD were the best predictors for prevalent vertebral fractures. • Except for bone mineral content, all CT-based bone measures performed significantly better than DXA-based measures. • We introduce diagnostic thresholds of integral vBMD for osteoporosis (< 160 mg/cm3) and low bone mass (160 ≤ BMD < 190 mg/cm3) with almost equal sensitivity and specificity compared to conventional thresholds of quantitative CT as proposed by the American College of Radiology (osteoporosis < 80 mg/cm3).

Robust, Primitive, and Unsupervised Quality Estimation for Segmentation Ensembles.

A multitude of image-based machine learning segmentation and classification algorithms has recently been proposed, offering diagnostic decision support for the identification and characterization of glioma, Covid-19 and many other diseases. Even though these algorithms often outperform human experts in segmentation tasks, their limited reliability, and in particular the inability to detect failure cases, has hindered translation into clinical practice. To address this major shortcoming, we propose an unsupervised quality estimation method for segmentation ensembles. Our primitive solution examines discord in binary segmentation maps to automatically flag segmentation results that are particularly error-prone and therefore require special assessment by human readers. We validate our method both on segmentation of brain glioma in multi-modal magnetic resonance - and of lung lesions in computer tomography images. Additionally, our method provides an adaptive prioritization mechanism to maximize efficacy in use of human expert time by enabling radiologists to focus on the most difficult, yet important cases while maintaining full diagnostic autonomy. Our method offers an intuitive and reliable uncertainty estimation from segmentation ensembles and thereby closes an important gap toward successful translation of automatic segmentation into clinical routine.

Gender-, Age- and Region-Specific Characterization of Vertebral Bone Microstructure Through Automated Segmentation and 3D Texture Analysis of Routine Abdominal CT.

Purpose: To identify long-term reproducible texture features (TFs) of spinal computed tomography (CT), and characterize variations with regard to gender, age and vertebral level using our automated quantification framework. Methods: We performed texture analysis (TA) on baseline and follow-up CT (follow-up duration: 30-90 days) of 21 subjects (8 females, 13 males, age at baseline 61.2 ± 9.2 years) to determine long-term reproducibility. TFs with a long-term reproducibility error Δrel<5% were further analyzed for an association with age and vertebral level in a cohort of 376 patients (129 females, 247 males, age 62.5 ± 9.2 years). Automated analysis comprised labeling and segmentation of vertebrae into subregions using a convolutional neural network, calculation of volumetric bone mineral density (vBMD) with asynchronous calibration and TF extraction. Varianceglobal measures the spread of the gray-level distribution in an image while Entropy reflects the uniformity of gray-levels. Short-run emphasis (SRE), Long-run emphasis (LRE), Run-length non-uniformity (RLN) and Run percentage (RP) contain information on consecutive voxels of a particular grey-level, or grey-level range, in a particular direction. Long runs (LRE) represent coarse texture while short runs (SRE) represent fine texture. RLN reflects similarities in the length of runs while RP reflects distribution and homogeneity of runs with a specific direction. Results: Six of the 24 extracted TFs had Δrel<5% (Varianceglobal, Entropy, SRE, LRE, RLN, RP), and were analyzed further in 4716 thoracolumbar vertebrae. Five TFs (Varianceglobal,SRE,LRE, RLN,RP) showed a significant difference between genders (p<0.001), potentially being caused by a finer and more directional vertebral trabecular microstructure in females compared to males. Varianceglobal and Entropy showed a significant increase from the thoracic to the lumbar spine (p<0.001), indicating a higher degree and earlier initiation of trabecular microstructure deterioration at lower spinal levels. The four higher-order TFs showed significant variations between spine regions without a clear directional gradient (p ≤ 0.001-0.012). No TF showed a clear age dependence. vBMD differed significantly between genders, age groups and spine regions (p ≤ 0.001-0.002). Conclusion: Long-term reproducible CT-based TFs of the thoracolumbar spine were established and characterized in a predominantly older adult study population. The gender-, age- and vertebral-level-specific values may serve as foundation for osteoporosis diagnostics and facilitate future studies investigating vertebral microstructure.

A Vertebral Segmentation Dataset with Fracture Grading.

Published under a CC BY 4.0 license. Supplemental material is available for this article.