Dataset Information

Machine Learning for Automatic Paraspinous Muscle Area and Attenuation Measures on Low-Dose Chest CT Scans.

ABSTRACT: RATIONALE AND OBJECTIVES:To develop and evaluate an automated machine learning (ML) algorithm for segmenting the paraspinous muscles on chest computed tomography (CT) scans to evaluate for presence of sarcopenia. MATERIALS AND METHODS:A convolutional neural network based on the U-Net architecture was trained to perform muscle segmentation on a dataset of 1875 single slice CT images and was tested on 209 CT images of participants in the National Lung Screening Trial. Low-dose, noncontrast CT examinations were obtained at 33 clinical sites, using scanners from four manufacturers. The study participants had a mean age of 71.6 years (range, 70-74 years). Ground truth was obtained by manually segmenting the left paraspinous muscle at the level of the T12 vertebra. Muscle cross-sectional area (CSA) and muscle attenuation (MA) were recorded. Comparison between the ML algorithm and ground truth measures of muscle CSA and MA were obtained using Dice similarity coefficients and Pearson correlations. RESULTS:Compared to ground truth segmentation, the ML algorithm achieved median (standard deviation) Dice scores of 0.94 (0.04) in the test set. Mean (SD) muscle CSA was 14.3 (3.6) cm² for ground truth and 13.7 (3.5) cm² for ML segmentation. Mean (SD) MA was 41.6 (7.6) Hounsfield units (HU) for ground truth and 43.5 (7.9) HU for ML segmentation. There was high correlation between ML algorithm and ground truth for muscle CSA (r²?=?0.86; p < 0.0001) and MA (r²?=?0.95; p < 0.0001). CONCLUSION:The ML algorithm for measurement of paraspinous muscles compared favorably to manual ground truth measurements in the NLST. The algorithm generalized well to a heterogeneous set of low-dose CT images and may be capable of automated quantification of muscle metrics to screen for sarcopenia on routine chest CT examinations.

SUBMITTER: Barnard R

PROVIDER: S-EPMC6878160 | biostudies-literature | 2019 Dec

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Machine Learning for Automatic Paraspinous Muscle Area and Attenuation Measures on Low-Dose Chest CT Scans.

Barnard Ryan R Tan Josh J Roller Brandon B Chiles Caroline C Weaver Ashley A AA Boutin Robert D RD Kritchevsky Stephen B SB Lenchik Leon L

Academic radiology 20190717 12

<h4>Rationale and objectives</h4>To develop and evaluate an automated machine learning (ML) algorithm for segmenting the paraspinous muscles on chest computed tomography (CT) scans to evaluate for presence of sarcopenia.<h4>Materials and methods</h4>A convolutional neural network based on the U-Net architecture was trained to perform muscle segmentation on a dataset of 1875 single slice CT images and was tested on 209 CT images of participants in the National Lung Screening Trial. Low-dose, nonc ...[more]

PMID: 31326311

Similar Datasets

Project description:PurposeTo develop and validate a computed tomography (CT) harmonization technique by combining noise-stabilization and autocalibration methodologies to provide reliable densitometry measurements in heterogeneous acquisition protocols.MethodsWe propose to reduce the effects of spatially variant noise such as nonuniform patterns of noise and biases. The method combines the statistical characterization of the signal-to-noise relationship in the CT image intensities, which allows us to estimate both the signal and spatially variant variance of noise, with an autocalibration technique that reduces the nonuniform biases caused by noise and reconstruction techniques. The method is firstly validated with anthropomorphic synthetic images that simulate CT acquisitions with variable scanning parameters: different dosage, nonhomogeneous variance of noise, and various reconstruction methods. We finally evaluate these effects and the ability of our method to provide consistent densitometric measurements in a cohort of clinical chest CT scans from two vendors (Siemens, n = 54 subjects; and GE, n = 50 subjects) acquired with several reconstruction algorithms (filtered back-projection and iterative reconstructions) with high-dose and low-dose protocols.ResultsThe harmonization reduces the effect of nonhomogeneous noise without compromising the resolution of the images (25% RMSE reduction in both clinical datasets). An analysis through hierarchical linear models showed that the average biases induced by differences in dosage and reconstruction methods are also reduced up to 74.20%, enabling comparable results between high-dose and low-dose reconstructions. We also assessed the statistical similarity between acquisitions obtaining increases of up to 30% points and showing that the low-dose vs high-dose comparisons of harmonized data obtain similar and even higher similarity than the observed for high-dose vs high-dose comparisons of nonharmonized data.ConclusionThe proposed harmonization technique allows to compare measures of low-dose with high-dose acquisitions without using a specific reconstruction as a reference. Since the harmonization does not require a precalibration with a phantom, it can be applied to retrospective studies. This approach might be suitable for multicenter trials for which a reference reconstruction is not feasible or hard to define due to differences in vendors, models, and reconstruction techniques.

Project description:Current methods for organ and effective dose estimations in pediatric CT are largely patient generic. Physical phantoms and computer models have only been developed for standard/limited patient sizes at discrete ages (e.g., 0, 1, 5, 10, 15 years old) and do not reflect the variability of patient anatomy and body habitus within the same size/age group. In this investigation, full-body computer models of seven pediatric patients in the same size/protocol group (weight: 11.9-18.2 kg) were created based on the patients' actual multi-detector array CT (MDCT) data. Organs and structures in the scan coverage were individually segmented. Other organs and structures were created by morphing existing adult models (developed from visible human data) to match the framework defined by the segmented organs, referencing the organ volume and anthropometry data in ICRP Publication 89. Organ and effective dose of these patients from a chest MDCT scan protocol (64 slice LightSpeed VCT scanner, 120 kVp, 70 or 75 mA, 0.4 s gantry rotation period, pitch of 1.375, 20 mm beam collimation, and small body scan field-of-view) was calculated using a Monte Carlo program previously developed and validated to simulate radiation transport in the same CT system. The seven patients had normalized effective dose of 3.7-5.3 mSv/100 mAs (coefficient of variation: 10.8%). Normalized lung dose and heart dose were 10.4-12.6 mGy/100 mAs and 11.2-13.3 mGy/100 mAs, respectively. Organ dose variations across the patients were generally small for large organs in the scan coverage (<7%), but large for small organs in the scan coverage (9%-18%) and for partially or indirectly exposed organs (11%-77%). Normalized effective dose correlated weakly with body weight (correlation coefficient: r=-0.80). Normalized lung dose and heart dose correlated strongly with mid-chest equivalent diameter (lung: r=-0.99, heart: r=-0.93); these strong correlation relationships can be used to estimate patient-specific organ dose for any other patient in the same size/protocol group who undergoes the chest scan. In summary, this work reported the first assessment of dose variations across pediatric CT patients in the same size/protocol group due to the variability of patient anatomy and body habitus and provided a previously unavailable method for patient-specific organ dose estimation, which will help in assessing patient risk and optimizing dose reduction strategies, including the development of scan protocols.

Dataset Information

Machine Learning for Automatic Paraspinous Muscle Area and Attenuation Measures on Low-Dose Chest CT Scans.

Publications

Machine Learning for Automatic Paraspinous Muscle Area and Attenuation Measures on Low-Dose Chest CT Scans.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets