Radiomics and stacking regression model for measuring bone mineral density using abdominal computed tomography

2021 ◽  
pp. 028418512110681
Author(s):  
Hong Dai ◽  
Yutao Wang ◽  
Randi Fu ◽  
Sijia Ye ◽  
Xiuchao He ◽  
...  
Keyword(s):  
Computed Tomography ◽  
Bone Mineral Density ◽  
Regression Model ◽  
Linear Model ◽  
Bone Mineral ◽  
Lumbar Vertebrae ◽  
Ct Images ◽  
Mineral Density ◽  
Abdominal Ct ◽  
Abdominal Computed Tomography

Background Measurement of bone mineral density (BMD) is the most important method to diagnose osteoporosis. However, current BMD measurement is always performed after a fracture has occurred. Purpose To explore whether a radiomic model based on abdominal computed tomography (CT) can predict the BMD of lumbar vertebrae. Material and Methods A total of 245 patients who underwent both dual-energy X-ray absorptiometry (DXA) and abdominal CT examination (training cohort, n = 196; validation cohort, n = 49) were included in our retrospective study. In total, 1218 image features were extracted from abdominal CT images for each patient. Combined with clinical information, three steps including least absolute shrinkage and selection operator (LASSO) regression were used to select key features. A two-tier stacking regression model with multi-algorithm fusion was used for BMD prediction, which can integrate the advantages of linear model and non-linear model. The prediction results of this model were compared with those using a single regressor. The degree-of-freedom adjusted coefficient of determination (Adjusted-R2), root mean square error (RMSE), and mean absolute error (MAE) were used to evaluate the regression performance. Results Compared with other regression methods, the two-tier stacking regression model has a higher regression performance, with Adjusted-R2, RMSE, and MAE of 0.830, 0.077, and 0.06, respectively. Pearson correlation analysis and Bland–Altman analysis showed that the BMD predicted by the model had a high correlation with the DXA results (r = 0.932, difference = −0.01 ± 0.1412 mg/cm2). Conclusion Using radiomics, the BMD of lumbar vertebrae could be predicted from abdominal CT images.

scholarly journals AGE AND SEX-RELATED STRUCTURAL AND FUNCTIONAL CHANGES OF BONE REMODELING DURING SIMULTATE ABDOMEN CT-SCANNING

2020 ◽  
Vol 73 (1) ◽  
pp. 91-94
Author(s):  
Nazar M. Kostyshyn ◽  
Liubov P. Kostyshyn ◽  
Mechyslav R. Gzhegotskyi
Keyword(s):  
Computed Tomography ◽  
Bone Mineral Density ◽  
Bone Mass ◽  
Old Age ◽  
Bone Mineral ◽  
Lumbar Vertebrae ◽  
Middle Adulthood ◽  
Mineral Density ◽  
Functional Changes ◽  
Early Old Age

The aim of investigation was to study the structural and functional conditions of cortical and trabecular layers of lumbar vertebrae L1-L5 in different age groups. Materials and methods: In order to assess BMD of the lumbar vertebrae 102 people18 to 75 years old was examined. Study of bone mineral density cortical and trabecular layer of lumbar vertebrae (L1-L5) performed by computed tomography in Hounsfield Units (HU) in terms of standard deviation (SD). Results: The results of computed tomography showed a direct relationship of bone mineral density of lumbar vertebrae with age of examined persons. Osteoporosis and osteopenia was registered in 15% of men and 30% women in middle adulthood, in late adulthood – 35% and 50% respectively. During early old age osteopenia and osteoporosis are observed in 37,5% and 25% of men and 26% and 64% women. Osteoporosis in the middle old age has been reported in 50% of men and 75% of women Conclusions: Results of the study showed a direct link between mineral density, age and gender. The middle adulthood age period was characterized by the highest mineral mass compared with the other age periods. Then there is a loss of bone mass throughout life, and with the onset of aging osteopenia and osteoporosis are recorded. It was investigated that in the early old age, loss of bone mass is mainly observed in female.

scholarly journals Correlation between the Computed Tomography Values of the Screw Path and Pedicle Screw Pullout Strength: An Experimental Study in Porcine Vertebrae

2020 ◽  
Vol 14 (3) ◽  
pp. 265-272
Author(s):  
Atsushi Ikeura ◽  
Taketoshi Kushida ◽  
Kenichi Oe ◽  
Yoshihisa Kotani ◽  
Muneharu Ando ◽  
...  
Keyword(s):  
Computed Tomography ◽  
Bone Mineral Density ◽  
Pedicle Screw ◽  
Bone Mineral ◽  
Lumbar Vertebrae ◽  
Pedicle Screws ◽  
Mineral Density ◽  
Screw Insertion ◽  
Pull Out ◽  
Pull Out Strength

Study Design: Biomechanical study.Purpose: To assess the correlation between the computed tomography (CT) values of the pedicle screw path and screw pull-out strength.Overview of Literature: The correlation between pedicle screw pull-out strength and bone mineral density has been well established. In addition, several reports have demonstrated a correlation between bone mineral density and CT values. However, no previous biomechanical studies investigated the correlation between CT values and pedicle screw pull-out strength.Methods: Sixty fresh-frozen lumbar vertebrae from 6-month-old pigs were used. Before screw insertion, the CT values of the screw path were obtained for each sample. Specimens were then randomly divided into three equal groups. Each group had one of three pedicle screws inserted: 4.0-mm LEGACY (4.0-LEG), 4.5-mm LEGACY (4.5-LEG), or 4.5-mm SOLERA (4.5-SOL) (all from Medtronic Sofamor Danek Inc., Memphis, TN, USA). Each screw had a consistent 30-mm thread length. Axial pull-out testing was performed at a rate of 1.0 mm/min. Correlations between the CT values and pedicle screw pull-out strength were evaluated using Pearson’s correlation coefficient analysis.Results: The correlation coefficients between the CT values of the screw path and pedicle screw pull-out strength for the 4.0-LEG, 4.5-LEG, and 4.5-SOL groups were 0.836 (<i>p</i> <0.001), 0.780 (<i>p</i> <0.001), and 0.873 (<i>p</i> <0.001), respectively. Greater CT values were associated with greater screw pull-out strength.Conclusions: The CT values of the screw path were strongly positively correlated with pedicle screw pull-out strength, regardless of the screw type and diameter, suggesting that the CT values could be clinically useful for predicting pedicle screw pull-out strength.

scholarly journals Phantomless assessment of volumetric bone mineral density using virtual non-contrast images from spectral detector computed tomography

2020 ◽  
Vol 93 (1109) ◽  
pp. 20190992 ◽  
Author(s):  
David Zopfs ◽  
Simon Lennartz ◽  
Charlotte Zaeske ◽  
Martin Merkt ◽  
Kai Roman Laukamp ◽  
...  
Keyword(s):  
Computed Tomography ◽  
Bone Mineral Density ◽  
Contrast Enhancement ◽  
Bone Mineral ◽  
Lumbar Vertebrae ◽  
Volumetric Bone Mineral Density ◽  
Mineral Density ◽  
Phase Images ◽  
Spectral Detector Ct ◽  
Venous Phase

Objective: To evaluate phantomless assessment of volumetric bone mineral density (vBMD) based on virtual non-contrast images of arterial (VNCa) and venous phase (VNCv) derived from spectral detector CT in comparison to true non-contrast (TNC) images and adjusted venous phase conventional images (CIV(adjusted)). Methods: 104 consecutive patients who underwent triphasic spectral detector CT between January 2018 and April 2019 were retrospectively included. TNC, VNCa, VNCv and venous phase images (CIV) were reconstructed. vBMD was obtained by two radiologists using an FDA/CE-cleared software. Average vBMD of the first three lumbar vertebrae was determined in each reconstruction; vBMD of CIV was adjusted for contrast enhancement as suggested earlier. Results: vBMD values obtained from CIV(adjusted) are comparable to vBMD values derived from TNC images (91.79 ± 36.52 vs 90.16 ± 41.71 mg/cm3, p = 1.00); however, vBMD values derived from VNCa and VNCv (42.20 ± 22.50 and 41.98 ± 23.3 mg/cm3 respectively) were significantly lower as compared to vBMD values from TNC and CIV(adjusted) (all p ≤ 0.01). Conclusion: Spectral detector CT-derived virtual non-contrast images systematically underestimate vBMD and therefore should not be used without appropriate adjustments. Adjusted venous phase images provide reliable results and may be utilized for an opportunistic BMD screening in CT examinations. Advances in knowledge: Adjustments of venous phase images facilitate opportunistic assessment of vBMD, while spectral detector CT-derived VNC images systematically underestimate vBMD.

Assessment of the bone mineral density in the lumbar vertebrae of newborns by quantitative computed tomography

1996 ◽  
Vol 25 (8) ◽  
pp. 711-715 ◽  
Author(s):  
P. M. Braillon ◽  
A. Lapillonne ◽  
P. S. Ho ◽  
R. Bouvier ◽  
M. Bochu ◽  
...  
Keyword(s):  
Computed Tomography ◽  
Bone Mineral Density ◽  
Bone Mineral ◽  
Quantitative Computed Tomography ◽  
Lumbar Vertebrae ◽  
Mineral Density

scholarly journals Hounsfield Units on Lumbar Computed Tomography for Predicting Regional Bone Mineral Density

Open Medicine ◽  
2019 ◽  
Vol 14 (1) ◽  
pp. 545-551 ◽  
Author(s):  
Kyung Joon Kim ◽  
Dong Hwan Kim ◽  
Jae Il Lee ◽  
Byung Kwan Choi ◽  
In Ho Han ◽  
...  
Keyword(s):  
Computed Tomography ◽  
Bone Mineral Density ◽  
Bone Mineral ◽  
Quantitative Computed Tomography ◽  
Lumbar Vertebrae ◽  
Radiation Hazard ◽  
Mineral Density ◽  
Hounsfield Units ◽  
Fusion Surgery ◽  
T Score

AbstractObjectiveBone mineral density (BMD) is a very important factor in spinal fusion surgery using instrumentation. Our aim was to investigate the utility of Hounsfield units (HU) obtained from preoperative lumbar computed tomography (CT) to predict osteoporosis coupling with data of quantitative computed tomography (QCT) and dual X-ray absorptiometry (DEXA).MethodsWe reviewed 180 patients that underwent both QCT and lumbar CT for spine surgery. HU was retrospectively calculated on the lumbar CT of 503 lumbar vertebrae from L1 to L3. Femur DEXA was performed in all patients and spine DEXA was tested in 120 patients (331 vertebrae). BMD was grouped as osteoporosis (QCT<80mg/cm3, DEXA T score≤-2.5) and non-osteoporosis (QCT≥80mg/cm3, DEXA T score>-2.5) for comparison of HU value.ResultsHU value and BMD showed significant correlations. The optima cut-off value based on QCT was higher than that of DEXA scans which had the best correlation for predicting osteoporosis. ROC curve analysis demonstrated that HU value with QCT of 146 has a sensitivity of 94.3% and a specificity of 87.5% for osteoporosis.ConclusionsSignificant correlation was found between HU measurement and BMD value. These findings provide evidence that HU measurement can be established as a means for predicting osteoporosis before spine fusion surgery with reduced radiation hazard.

Sign in / Sign up

Export Citation Format

Share Document