Automated body composition analysis of clinically acquired computed tomography scans using neural networks.
Ontology highlight
ABSTRACT: Background & aims
The quantity and quality of skeletal muscle and adipose tissue is an important prognostic factor for clinical outcomes across several illnesses. Clinically acquired computed tomography (CT) scans are commonly used for quantification of body composition, but manual analysis is laborious and costly. The primary aim of this study was to develop an automated body composition analysis framework using CT scans.Methods
CT scans of the 3rd lumbar vertebrae from critically ill, liver cirrhosis, pancreatic cancer, and clear cell renal cell carcinoma patients, as well as renal and liver donors, were manually analyzed for body composition. Ninety percent of scans were used for developing and validating a neural network for the automated segmentation of skeletal muscle and adipose tissues. Network accuracy was evaluated with the remaining 10 percent of scans using the Dice similarity coefficient (DSC), which quantifies the overlap (0 = no overlap, 1 = perfect overlap) between human and automated segmentations.Results
Of the 893 patients, 44% were female, with a mean (±SD) age and body mass index of 52.7 (±15.8) years old and 28.0 (±6.1) kg/m2, respectively. In the testing cohort (n = 89), DSC scores indicated excellent agreement between human and network-predicted segmentations for skeletal muscle (0.983 ± 0.013), and intermuscular (0.900 ± 0.034), visceral (0.979 ± 0.019), and subcutaneous (0.986 ± 0.016) adipose tissue. Network segmentation took ~350 milliseconds/scan using modern computing hardware.Conclusions
Our network displayed excellent ability to analyze diverse body composition phenotypes and clinical cohorts, which will create feasible opportunities to advance our capacity to predict health outcomes in clinical populations.
SUBMITTER: Paris MT
PROVIDER: S-EPMC7374050 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:This study aimed to evaluate acute pancreatitis (AP) severity using convolutional neural network (CNN) models with enhanced computed tomography (CT) scans. Three-dimensional DenseNet CNN models were developed and trained using the enhanced CT scans labeled with two severity assessment methods: the computed tomography severity index (CTSI) and Atlanta classification. Each labeling method was used independently for model training and validation. Model performance was evaluated using confusion matrices, areas under the receiver operating characteristic curve (AUC-ROC), accuracy, precision, recall, F1 score, and respective macro-average metrics. A total of 1,798 enhanced CT scans met the inclusion criteria were included in this study. The dataset was randomly divided into a training dataset (n = 1618) and a test dataset (n = 180) with a ratio of 9:1. The DenseNet model demonstrated promising predictions for both CTSI and Atlanta classification-labeled CT scans, with accuracy greater than 0.7 and AUC-ROC greater than 0.8. Specifically, when trained with CT scans labeled using CTSI, the DenseNet model achieved good performance, with a macro-average F1 score of 0.835 and a macro-average AUC-ROC of 0.980. The findings of this study affirm the feasibility of employing CNN models to predict the severity of AP using enhanced CT scans.
| S-EPMC10579320 | biostudies-literature
Project description:BackgroundBody composition during childhood may predispose to negative health outcomes later in life. Automatic segmentation may assist in quantifying pediatric body composition in children.ObjectiveTo evaluate automatic segmentation for body composition on pediatric computed tomography (CT) scans and to provide normative data on muscle and fat areas throughout childhood using automatic segmentation.Materials and methodsIn this pilot study, 537 children (ages 1-17 years) who underwent abdominal CT after high-energy trauma at a Dutch tertiary center (2002-2019) were retrospectively identified. Of these, the CT images of 493 children (66% boys) were used to establish normative data. Muscle (psoas, paraspinal and abdominal wall) and fat (subcutaneous and visceral) areas were measured at the third lumbar vertebral (L3) level by automatic segmentation. A representative subset of 52 scans was also manually segmented to evaluate the performance of automatic segmentation.ResultsFor manually-segmented versus automatically-segmented areas (52 scans), mean Dice coefficients were high for muscle (0.87-0.90) and subcutaneous fat (0.88), but lower for visceral fat (0.60). In the control group, muscle area was comparable for both sexes until the age of 13 years, whereafter, boys developed relatively more muscle. From a young age, boys were more prone to visceral fat storage than girls. Overall, boys had significantly higher visceral-to-subcutaneous fat ratios (median 1.1 vs. 0.6, P<0.01) and girls higher fat-to-muscle ratios (median 1.0 vs. 0.7, P<0.01).ConclusionAutomatic segmentation of L3-level muscle and fat areas allows for accurate quantification of pediatric body composition. Using automatic segmentation, the development in muscle and fat distribution during childhood (in otherwise healthy) Dutch children was demonstrated.
| S-EPMC10635977 | biostudies-literature
Project description:Background and purposeGlioblastoma (GBM) patients face a strongly unfavorable prognosis despite multimodal therapy regimens. However, individualized mortality prediction remains imprecise. Harnessing routine radiation planning cranial computed tomography (CT) scans, we assessed cervical body composition measures as novel biomarkers for overall survival (OS) in GBM patients.Materials and methodsWe performed threshold-based semi-automated quantification of muscle and subcutaneous fat cross-sectional area (CSA) at the levels of the first and second cervical vertebral body. First, we tested this method's validity by correlating cervical measures to established abdominal body composition in an open-source whole-body CT cohort. We then identified consecutive patients undergoing radiation planning for recent GBM diagnosis at our institution from 2010 to 2020 and quantified cervical body composition on radiation planning CT scans. Finally, we performed univariable and multivariable time-to-event analyses, adjusting for age, sex, body mass index, comorbidities, performance status, extent of surgical resection, extent of tumor at diagnosis, and MGMT methylation.ResultsCervical body composition measurements were well-correlated with established abdominal markers (Spearman's rho greater than 0.68 in all cases). Subsequently, we included 324 GBM patients in our study cohort (median age 63 years, 60.8% male). 293 (90.4%) patients died during follow-up. Median survival time was 13 months. Patients with below-average muscle CSA or above-average fat CSA demonstrated shorter survival. In multivariable analyses, continuous cervical muscle measurements remained independently associated with OS.ConclusionThis exploratory study establishes novel cervical body composition measures routinely available on cranial radiation planning CT scans and confirms their association with OS in patients diagnosed with GBM.
| S-EPMC10063381 | biostudies-literature
Project description:BackgroundConvolutional neural networks (CNNs) are advanced artificial intelligence algorithms well suited to image classification tasks with variable features. These have been used to great effect in various real-world applications including handwriting recognition, face detection, image search, and fraud prevention. We sought to retrain a robust CNN with coronal computed tomography (CT) images to classify osteomeatal complex (OMC) occlusion and assess the performance of this technology with rhinologic data.MethodsThe Google Inception-V3 CNN trained with 1.28 million images was used as the base model. Preoperative coronal sections through the OMC were obtained from 239 patients enrolled in 2 prospective chronic rhinosinusitis (CRS) outcomes studies, labeled according to OMC status, and mirrored to obtain a set of 956 images. Using this data, the classification layer of Inception-V3 was retrained in Python using a transfer learning method to adapt the CNN to the task of interpreting sinonasal CT images.ResultsThe retrained neural network achieved 85% classification accuracy for OMC occlusion, with a 95% confidence interval for algorithm accuracy of 78% to 92%. Receiver operating characteristic (ROC) curve analysis on the test set confirmed good classification ability of the CNN with an area under the ROC curve (AUC) of 0.87, significantly different than both random guessing and a dominant classifier that predicts the most common class (p < 0.0001).ConclusionCurrent state-of-the-art CNNs may be able to learn clinically relevant information from 2-dimensional sinonasal CT images with minimal supervision. Future work will extend this approach to 3-dimensional images in order to further refine this technology for possible clinical applications.
| S-EPMC6318014 | biostudies-literature
Project description:Quick, efficient, fully automated open-source programs to segment muscle and adipose tissues from computed tomography (CT) images would be a great contribution to body composition research. This study examined the concordance of cross-sectional areas (CSA) and densities for muscle, visceral adipose tissue (VAT), subcutaneous adipose tissue (SAT), and intramuscular adipose tissue (IMAT) from CT images at the third lumbar (L3) between an automated neural network (test method) and a semi-automatic human-based program (reference method). Concordance was further evaluated by disease status, sex, race/ethnicity, BMI categories. Agreement statistics applied included Lin's Concordance (CCC), Spearman correlation coefficient (SCC), Sorensen dice-similarity coefficient (DSC), and Bland-Altman plots with limits of agreement (LOA) within 1.96 standard deviation. A total of 420 images from a diverse cohort of patients (60.35 ± 10.92 years; body mass index (BMI) of 28.77 ± 7.04 kg/m2; 55% female; 53% Black) were included in this study. About 30% of patients were healthy (i.e., received a CT scan for acute illness or pre-surgical donor work-up), while another 30% had a diagnosis of colorectal cancer. The CCC, SCC, and DSC estimates for muscle, VAT, SAT were all greater than 0.80 (>0.80 indicates good performance). Agreement analysis by diagnosis showed good performance for the test method except for critical illness (DSC 0.65-0.87). Bland-Altman plots revealed narrow LOA suggestive of good agreement despite minimal proportional bias around the zero-bias line for muscle, SAT, and IMAT CSA. The test method shows good performance and almost perfect concordance for L3 muscle, VAT, SAT, and IMAT per DSC estimates, and Bland-Altman plots even after stratification by sex, race/ethnicity, and BMI categories. Care must be taken to assess the density of the CT images from critically ill patients before applying the automated neural network (test method).
| S-EPMC9101564 | biostudies-literature
Project description:BackgroundBody composition from computed tomography (CT) scans is associated with cancer outcomes including surgical complications, chemotoxicity, and survival. Most studies manually segment CT scans, but Automatic Body composition Analyser using Computed tomography image Segmentation (ABACS) software automatically segments muscle and adipose tissues to speed analysis. Here, we externally evaluate ABACS in an independent dataset.MethodsAmong patients with non-metastatic colorectal (n = 3102) and breast (n = 2888) cancer diagnosed from 2005 to 2013 at Kaiser Permanente, expert raters annotated tissue areas at the third lumbar vertebra (L3). To compare ABACS segmentation results to manual analysis, we quantified the proportion of pixel-level image overlap using Jaccard scores and agreement between methods using intra-class correlation coefficients for continuous tissue areas. We examined performance overall and among subgroups defined by patient and imaging characteristics. To compare the strength of the mortality associations obtained from ABACS's segmentations to manual analysis, we computed Cox proportional hazards ratios (HRs) and 95% confidence intervals (95% CI) by tertile of tissue area.ResultsMean ± SD age was 63 ± 11 years for colorectal cancer patients and 56 ± 12 for breast cancer patients. There was strong agreement between manual and automatic segmentations overall and within subgroups of age, sex, body mass index, and cancer stage: average Jaccard scores and intra-class correlation coefficients exceeded 90% for all tissues. ABACS underestimated muscle and visceral and subcutaneous adipose tissue areas by 1-2% versus manual analysis: mean differences were small at -2.35, -1.97 and -2.38 cm2 , respectively. ABACS's performance was lowest for the <2% of patients who were underweight or had anatomic abnormalities. ABACS and manual analysis produced similar associations with mortality; comparing the lowest to highest tertile of skeletal muscle from ABACS versus manual analysis, the HRs were 1.23 (95% CI: 1.00-1.52) versus 1.38 (95% CI: 1.11-1.70) for colorectal cancer patients and 1.30 (95% CI: 1.01-1.66) versus 1.29 (95% CI: 1.00-1.65) for breast cancer patients.ConclusionsIn the first study to externally evaluate a commercially available software to assess body composition, automated segmentation of muscle and adipose tissues using ABACS was similar to manual analysis and associated with mortality after non-metastatic cancer. Automated methods will accelerate body composition research and, eventually, facilitate integration of body composition measures into clinical care.
| S-EPMC7567141 | biostudies-literature
Project description:BackgroundExcess adiposity in children is strongly correlated with obesity-related metabolic disease in adulthood, including diabetes, cardiovascular disease, and 13 types of cancer. Despite the many long-term health risks of childhood obesity, body mass index (BMI) Z-score is typically the only adiposity marker used in pediatric studies and clinical applications. The effects of regional adiposity are not captured in a single scalar measurement, and their effects on short- and long-term metabolic health are largely unknown. However, clinicians and researchers rarely deploy gold-standard methods for measuring compartmental fat such as magnetic resonance imaging (MRI) and dual X-ray absorptiometry (DXA) on children and adolescents due to cost or radiation concerns. Three-dimensional optical (3DO) scans are relatively inexpensive to obtain and use non-invasive and radiation-free imaging techniques to capture the external surface geometry of a patient's body. This 3D shape contains cues about the body composition that can be learned from a structured correlation between 3D body shape parameters and reference DXA scans obtained on a sample population.Study aimThis study seeks to introduce a radiation-free, automated 3D optical imaging solution for monitoring body shape and composition in children aged 5-17.MethodsWe introduce an automated, linear learning method to predict total and regional body composition of children aged 5-17 from 3DO scans. We collected 145 male and 206 female 3DO scans on children between the ages of 5 and 17 with three scanners from independent manufacturers. We used an automated shape templating method first introduced on an adult population to fit a topologically consistent 60,000 vertex (60 k) mesh to 3DO scans of arbitrary scanning source and mesh topology. We constructed a parameterized body shape space using principal component analysis (PCA) and estimated a regression matrix between the shape parameters and their associated DXA measurements. We automatically fit scans of 30 male and 38 female participants from a held-out test set and predicted 12 body composition measurements.ResultsThe coefficient of determination (R2) between 3DO predicted body composition and DXA measurements was at least 0.85 for all measurements with the exception of visceral fat on 3D scan predictions. Precision error was 1-4 times larger than that of DXA. No predicted variable was significantly different from DXA measurement except for male trunk lean mass.ConclusionOptical imaging can quickly, safely, and inexpensively estimate regional body composition in children aged 5-17. Frequent repeat measurements can be taken to chart changes in body adiposity over time without risk of radiation overexposure.
| S-EPMC10528749 | biostudies-literature
Project description:ObjectivesBody tissue composition is a long-known biomarker with high diagnostic and prognostic value not only in cardiovascular, oncological, and orthopedic diseases but also in rehabilitation medicine or drug dosage. In this study, the aim was to develop a fully automated, reproducible, and quantitative 3D volumetry of body tissue composition from standard CT examinations of the abdomen in order to be able to offer such valuable biomarkers as part of routine clinical imaging.MethodsTherefore, an in-house dataset of 40 CTs for training and 10 CTs for testing were fully annotated on every fifth axial slice with five different semantic body regions: abdominal cavity, bones, muscle, subcutaneous tissue, and thoracic cavity. Multi-resolution U-Net 3D neural networks were employed for segmenting these body regions, followed by subclassifying adipose tissue and muscle using known Hounsfield unit limits.ResultsThe Sørensen Dice scores averaged over all semantic regions was 0.9553 and the intra-class correlation coefficients for subclassified tissues were above 0.99.ConclusionsOur results show that fully automated body composition analysis on routine CT imaging can provide stable biomarkers across the whole abdomen and not just on L3 slices, which is historically the reference location for analyzing body composition in the clinical routine.Key points• Our study enables fully automated body composition analysis on routine abdomen CT scans. • The best segmentation models for semantic body region segmentation achieved an averaged Sørensen Dice score of 0.9553. • Subclassified tissue volumes achieved intra-class correlation coefficients over 0.99.
| S-EPMC7979624 | biostudies-literature
Project description:BackgroundMany predictors of morbidity caused by metabolic disease are associated with body shape. 3D optical (3DO) scanning captures body shape and has been shown to accurately and precisely predict body composition variables associated with mortality risk. 3DO is safer, less expensive, and more accessible than criterion body composition assessment methods such as dual-energy X-ray absorptiometry (DXA). However, 3DO scanning has not been standardized across manufacturers for pose, mesh resolution, and post processing methods.PurposeWe introduce a scanner-agnostic algorithm that automatically fits a topologically consistent human mesh to 3DO scanned point clouds and predicts clinically important body metrics using a standardized body shape model. Our models transform raw scans captured by any 3DO scanner into fixed topology meshes with anatomical consistency, standardizing the outputs of 3DO scans across manufacturers and allowing for the use of common prediction models across scanning devices.MethodsA fixed-topology body mesh template was automatically registered to 848 training scans from three different 3DO systems. Participants were between 18 and 89 years old with body mass index ranging from 14 to 52 kg/m2 . Scans were registered by first performing a coarse nearest neighbor alignment between the template and the input scan with an anatomically constrained principal component analysis (PCA) domain deformation using a device and gender specific bootstrap basis trained on 70 seed scans each. The template mesh was then optimized to fit the target with a smooth per-vertex surface-to-surface deformation. A combined unified PCA model was created from the superset of all automatically fit training scans including all three devices. Body composition predictions to DXA measurements were learned from the training mesh PCA coefficients using linear regression. Using this final unified model, we tested the accuracy of our body composition models on a withheld sample of 562 scans by fitting a PCA parameterized template mesh to each raw scan and predicting the expected body composition metrics from the principal components using the learned regression model.ResultsWe achieved coefficients of determination (R2 ) above 0.8 on all nine fat and lean predictions except female visceral fat (0.77). R2 was as high as 0.94 (total fat and lean, trunk fat), and all root-mean-squared errors were below 3.0 kg. All predicted body composition variables were not significantly different from reference DXA measurements except for visceral fat and female trunk fat. Repeatability precision as measured by the coefficient of variation (%CV) was around 2-3x worse than DXA precision, with visceral fat %CV below 2x DXA %CV and female total fat mass at 5x.ConclusionsOur method provides an accurate, automated, and scanner agnostic framework for standardizing 3DO scans and a low cost, radiation-free alternative to criterion radiology imaging for body composition analysis. We published a web-app version of this work at https://shapeup.shepherdresearchlab.org/3do-bodycomp-analyzer/ that accepts mesh file uploads and returns templated meshes with body composition predictions for demo purposes.
| S-EPMC9990507 | biostudies-literature
Project description:Coronary artery calcium is an accurate predictor of cardiovascular events. While it is visible on all computed tomography (CT) scans of the chest, this information is not routinely quantified as it requires expertise, time, and specialized equipment. Here, we show a robust and time-efficient deep learning system to automatically quantify coronary calcium on routine cardiac-gated and non-gated CT. As we evaluate in 20,084 individuals from distinct asymptomatic (Framingham Heart Study, NLST) and stable and acute chest pain (PROMISE, ROMICAT-II) cohorts, the automated score is a strong predictor of cardiovascular events, independent of risk factors (multivariable-adjusted hazard ratios up to 4.3), shows high correlation with manual quantification, and robust test-retest reliability. Our results demonstrate the clinical value of a deep learning system for the automated prediction of cardiovascular events. Implementation into clinical practice would address the unmet need of automating proven imaging biomarkers to guide management and improve population health.
| S-EPMC7846726 | biostudies-literature