Whole-body Composition Profiling Using a Deep Learning Algorithm: Influence of Different Acquisition Parameters on Algorithm Performance and Robustness

Body composition as a biomarker for assessing future lung cancer risk

Purpose

To investigate if body composition is a biomarker for assessing the risk of developing lung cancer.

Materials and Methods

Low-dose computed tomography (LDCT) scans from the Pittsburgh Lung Screening Study (PLuSS) (n=3,635, 22 follow-up years) and NLST-ACRIN (n=16,435, 8 follow-up years) cohorts were used in the study. Artificial intelligence (AI) algorithms were developed to automatically segment and quantify subcutaneous adipose tissue (SAT), visceral adipose tissue (VAT), intramuscular adipose tissue (IMAT), skeletal muscle (SM), and bone. Cause-specific Cox proportional hazards models were used to evaluate the hazard ratios (HRs). Standard time-dependent receiver operating characteristic (ROC) analysis was used to evaluate the prognostic ability of different models over time.

Results

The final composite models were formed by seven variables: age (HR=1.20), current smoking status (HR=1.59), bone volume (HR=1.79), SM density (HR=0.29), IMAT ratio (HR=0.33), IMAT density (HR=0.56), and SAT volume (HR=0.56). The models trained on the PLuSS cohort achieved a mean AUC of 0.76 (95% CI: 0.74-0.79) over 21 follow-up years and 0.70 (95% CI: 0.66-0.74) over the first 7 follow-up years for predicting lung cancer development within the PLuSS cohort. In contrast, models trained on the PLuSS cohort alone, as well as in combination with the NLST cohorts, achieved an AUC ranging from 0.61 to 0.68 in the NLST cohort over a 7-year follow-up period.

Conclusion

Body composition assessed on LDCT is a significant predictor of lung cancer risk and could improve the effectiveness of LDCT lung cancer screening by optimizing the screening eligibility and frequency.

Summary statement

Body composition assessed on LDCT is a significant predictor of lung cancer risk and could improve the effectiveness of LDCT lung cancer screening by optimizing the screening eligibility and frequency.

Key Points

This study unveils the significant associations between body tissues and lung cancer risk.The prediction models based on body composition alone, as well as the combination of demographics and body composition features can effectively identify patients at higher risk of developing lung cancer.

Reproducibility and across-site transferability of an improved deep learning approach for aneurysm detection and segmentation in time-of-flight MR-angiograms

This study aimed to (1) replicate a deep-learning-based model for cerebral aneurysm segmentation in TOF-MRAs, (2) improve the approach by testing various fully automatic pre-processing pipelines, and (3) rigorously validate the model's transferability on independent, external test-datasets. A convolutional neural network was trained on 235 TOF-MRAs acquired on local scanners from a single vendor to segment intracranial aneurysms. Different pre-processing pipelines including bias field correction, resampling, cropping and intensity-normalization were compared regarding their effect on model performance. The models were tested on independent, external same-vendor and other-vendor test-datasets, each comprised of 70 TOF-MRAs, including patients with and without aneurysms. The best-performing model achieved excellent results on the external same-vendor test-dataset, surpassing the results of the previous publication with an improved sensitivity (0.97 vs. ~ 0.86), a higher Dice score coefficient (DSC, 0.60 ± 0.25 vs. 0.53 ± 0.31), and an improved false-positive rate (0.87 ± 1.35 vs. ~ 2.7 FPs/case). The model further showed excellent performance in the external other-vendor test-datasets (DSC 0.65 ± 0.26; sensitivity 0.92, 0.96 ± 2.38 FPs/case). Specificity was 0.38 and 0.53, respectively. Raising the voxel-size from 0.5 × 0.5×0.5 mm to 1 × 1×1 mm reduced the false-positive rate seven-fold. This study successfully replicated core principles of a previous approach for detecting and segmenting cerebral aneurysms in TOF-MRAs with a robust, fully automatable pre-processing pipeline. The model demonstrated robust transferability on two independent external datasets using TOF-MRAs from the same scanner vendor as the training dataset and from other vendors. These findings are very encouraging regarding the clinical application of such an approach.

Muscle Reference Values From Thoracic and Abdominal CT for Sarcopenia Assessment: The Framingham Heart Study

BACKGROUND

Loss of muscle mass is a known feature of sarcopenia and predicts poor clinical outcomes. Although muscle metrics can be derived from routine computed tomography (CT) images, sex-specific reference values at multiple vertebral levels over a wide age range are lacking.

OBJECTIVE

The aim of this study was to provide reference values for skeletal muscle mass and attenuation on thoracic and abdominal CT scans in the community-based Framingham Heart Study cohort to aid in the identification of sarcopenia.

MATERIALS AND METHODS

This secondary analysis of a prospective trial describes muscle metrics by age and sex for participants from the Framingham Heart Study without prior history of cancer who underwent at least 1 CT scan between 2002 and 2011. Using 2 previously validated machine learning algorithms followed by human quality assurance, skeletal muscle was analyzed on a single axial CT image per level at the 5th, 8th, 10th thoracic, and 3rd lumbar vertebral body (T5, T8, T10, L3). Cross-sectional muscle area (cm 2 ), mean skeletal muscle radioattenuation (SMRA, in Hounsfield units), skeletal muscle index (SMI, in cm 2 /m 2 ), and skeletal muscle gauge (SMRA·SMI) were calculated. Measurements were summarized by age group (<45, 45-54, 55-64, 65-74, ≥75 years), sex, and vertebral level. Models enabling the calculation of age-, sex-, and vertebral-level-specific reference values were created and embedded into an open access online Web application.

RESULTS

The cohort consisted of 3804 participants (1917 [50.4%] males; mean age, 55.6 ± 11.8 years; range, 33-92 years) and 7162 CT scans. Muscle metrics qualitatively decreased with increasing age and female sex.

CONCLUSIONS

This study established age- and sex-specific reference values for CT-based muscle metrics at thoracic and lumbar vertebral levels. These values may be used in future research investigating the role of muscle mass and attenuation in health and disease, and to identify sarcopenia.

Deep learning-based super-resolution gradient echo imaging of the pancreas: Improvement of image quality and reduction of acquisition time

PURPOSE

The purpose of this study was to evaluate the impact of a deep learning-based super-resolution technique on T1-weighted gradient-echo acquisitions (volumetric interpolated breath-hold examination; VIBE) on the assessment of pancreatic MRI at 1.5 T compared to standard VIBE imaging (VIBESTD).

MATERIALS AND METHODS

This retrospective single-center study was conducted between April 2021 and October 2021. Fifty patients with a total of 50 detectable pancreatic lesion entities were included in this study. There were 27 men and 23 women, with a mean age of 69 ± 13 (standard deviation [SD]) years (age range: 33-89 years). VIBESTD (precontrast, dynamic, postcontrast) was retrospectively processed with a deep learning-based super-resolution algorithm including a more aggressive partial Fourier setting leading to a simulated acquisition time reduction (VIBESR). Image analysis was performed by two radiologists regarding lesion detectability, noise levels, sharpness and contrast of pancreatic edges, as well as regarding diagnostic confidence using a 5-point Likert-scale with 5 being the best.

RESULTS

VIBESR was rated better than VIBESTD by both readers regarding lesion detectability (5 [IQR: 5, 5] vs. 5 [IQR: 4, 5], for reader 1; 5 [IQR: 5, 5] vs. 4 [IQR: 4, 5]) for reader 2; both P <0.001), noise levels (5 [IQR: 5, 5] vs. 5 [IQR: 4, 5] for reader 1; 5 [IQR: 5, 5] vs. 4 [IQR: 4, 5] for reader 2; both P <0.001), sharpness and contrast of pancreatic edges (5 [IQR: 5, 5] vs. 5 [IQR: 4, 5] for reader 1; 5 [IQR: 5, 5] vs. 4 [IQR: 4, 5] for reader 2; both P <0.001), as well as regarding diagnostic confidence (5 [IQR: 5, 5] vs. 5 [IQR: 4, 5] for reader 1; 5 [IQR: 5, 5] vs. 4 [IQR: 4, 5] for reader 2; both P <0.001). There were no significant differences between lesion sizes as measured by the two readers on VIBESR and VIBESTD images (P > 0.05). The mean acquisition time for VIBESTD (15 ± 1 [SD] s; range: 11-16 s) was longer than that for VIBESR (13 ± 1 [SD] s; range: 11-14 s) (P < 0.001).

CONCLUSION

Our results indicate that the newly developed deep learning-based super-resolution algorithm adapted to partial Fourier acquisitions has a positive influence not only on shortening the examination time but also on improvement of image quality in pancreatic MRI.

Automated segmentation of five different body tissues on computed tomography using deep learning

PURPOSE

To develop and validate a computer tool for automatic and simultaneous segmentation of five body tissues depicted on computed tomography (CT) scans: visceral adipose tissue (VAT), subcutaneous adipose tissue (SAT), intermuscular adipose tissue (IMAT), skeletal muscle (SM), and bone.

METHODS

A cohort of 100 CT scans acquired on different subjects were collected from The Cancer Imaging Archive-50 whole-body positron emission tomography-CTs, 25 chest, and 25 abdominal. Five different body tissues (i.e., VAT, SAT, IMAT, SM, and bone) were manually annotated. A training-while-annotating strategy was used to improve the annotation efficiency. The 10-fold cross-validation method was used to develop and validate the performance of several convolutional neural networks (CNNs), including UNet, Recurrent Residual UNet (R2Unet), and UNet++. A grid-based three-dimensional patch sampling operation was used to train the CNN models. The CNN models were also trained and tested separately for each body tissue to see if they could achieve a better performance than segmenting them jointly. The paired sample t-test was used to statistically assess the performance differences among the involved CNN models RESULTS: When segmenting the five body tissues simultaneously, the Dice coefficients ranged from 0.826 to 0.840 for VAT, from 0.901 to 0.908 for SAT, from 0.574 to 0.611 for IMAT, from 0.874 to 0.889 for SM, and from 0.870 to 0.884 for bone, which were significantly higher than the Dice coefficients when segmenting the body tissues separately (p < 0.05), namely, from 0.744 to 0.819 for VAT, from 0.856 to 0.896 for SAT, from 0.433 to 0.590 for IMAT, from 0.838 to 0.871 for SM, and from 0.803 to 0.870 for bone.

CONCLUSION

There were no significant differences among the CNN models in segmenting body tissues, but jointly segmenting body tissues achieved a better performance than segmenting them separately.

Combining Deep Learning and Radiomics for Automated, Objective, Comprehensive Bone Marrow Characterization From Whole-Body MRI: A Multicentric Feasibility Study

OBJECTIVES

Disseminated bone marrow (BM) involvement is frequent in multiple myeloma (MM). Whole-body magnetic resonance imaging (wb-MRI) enables to evaluate the whole BM. Reading of such whole-body scans is time-consuming, and yet radiologists can transfer only a small fraction of the information of the imaging data set to the report. This limits the influence that imaging can have on clinical decision-making and in research toward precision oncology. The objective of this feasibility study was to implement a concept for automatic, comprehensive characterization of the BM from wb-MRI, by automatic BM segmentation and subsequent radiomics analysis of 30 different BM spaces (BMS).

MATERIALS AND METHODS

This retrospective multicentric pilot study used a total of 106 wb-MRI from 102 patients with (smoldering) MM from 8 centers. Fifty wb-MRI from center 1 were used for training of segmentation algorithms (nnU-Nets) and radiomics algorithms. Fifty-six wb-MRI from 8 centers, acquired with a variety of different MRI scanners and protocols, were used for independent testing. Manual segmentations of 2700 BMS from 90 wb-MRI were performed for training and testing of the segmentation algorithms. For each BMS, 296 radiomics features were calculated individually. Dice score was used to assess similarity between automatic segmentations and manual reference segmentations.

RESULTS

The "multilabel nnU-Net" segmentation algorithm, which performs segmentation of 30 BMS and labels them individually, reached mean dice scores of 0.88 ± 0.06/0.87 ± 0.06/0.83 ± 0.11 in independent test sets from center 1/center 2/center 3-8 (interrater variability between radiologists, 0.88 ± 0.01). The subset from the multicenter, multivendor test set (center 3-8) that was of high imaging quality was segmented with high precision (mean dice score, 0.87), comparable to the internal test data from center 1. The radiomic BM phenotype consisting of 8880 descriptive parameters per patient, which result from calculation of 296 radiomics features for each of the 30 BMS, was calculated for all patients. Exemplary cases demonstrated connections between typical BM patterns in MM and radiomic signatures of the respective BMS. In plausibility tests, predicted size and weight based on radiomics models of the radiomic BM phenotype significantly correlated with patients' actual size and weight ( P = 0.002 and P = 0.003, respectively).

CONCLUSIONS

This pilot study demonstrates the feasibility of automatic, objective, comprehensive BM characterization from wb-MRI in multicentric data sets. This concept allows the extraction of high-dimensional phenotypes to capture the complexity of disseminated BM disorders from imaging. Further studies need to assess the clinical potential of this method for automatic staging, therapy response assessment, or prediction of biopsy results.

Automated Detection, Segmentation, and Classification of Pleural Effusion From Computed Tomography Scans Using Machine Learning

OBJECTIVE

This study trained and evaluated algorithms to detect, segment, and classify simple and complex pleural effusions on computed tomography (CT) scans.

MATERIALS AND METHODS

For detection and segmentation, we randomly selected 160 chest CT scans out of all consecutive patients (January 2016-January 2021, n = 2659) with reported pleural effusion. Effusions were manually segmented and a negative cohort of chest CTs from 160 patients without effusions was added. A deep convolutional neural network (nnU-Net) was trained and cross-validated (n = 224; 70%) for segmentation and tested on a separate subset (n = 96; 30%) with the same distribution of reported pleural complexity features as in the training cohort (eg, hyperdense fluid, gas, pleural thickening and loculation). On a separate consecutive cohort with a high prevalence of pleural complexity features (n = 335), a random forest model was implemented for classification of segmented effusions with Hounsfield unit thresholds, density distribution, and radiomics-based features as input. As performance measures, sensitivity, specificity, and area under the curves (AUCs) for detection/classifier evaluation (per-case level) and Dice coefficient and volume analysis for the segmentation task were used.

RESULTS

Sensitivity and specificity for detection of effusion were excellent at 0.99 and 0.98, respectively (n = 96; AUC, 0.996, test data). Segmentation was robust (median Dice, 0.89; median absolute volume difference, 13 mL), irrespective of size, complexity, or contrast phase. The sensitivity, specificity, and AUC for classification in simple versus complex effusions were 0.67, 0.75, and 0.77, respectively.

CONCLUSION

Using a dataset with different degrees of complexity, a robust model was developed for the detection, segmentation, and classification of effusion subtypes. The algorithms are openly available at https://github.com/usb-radiology/pleuraleffusion.git.

Magnetic resonance imaging-based body composition is associated with nutritional and inflammatory status: a longitudinal study in patients with Crohn's disease

Objective

To evaluate the changes in magnetic resonance imaging-based body composition parameters during follow-ups in patients with Crohn's disease (CD).

Methods

Between November 1, 2017, and June 30, 2021, patients diagnosed with CD, who underwent two or more magnetic resonance enterography (MRE) scans at our institution were retrospectively reviewed. The baseline and one subsequent follow-up scan for each patient were paired to form longitudinal comparisons. Skeletal muscle, visceral adipose tissue (VAT), and subcutaneous adipose tissue (SAT) indexes were calculated from tissue areas measured at the third lumbar vertebra level per scan, standardized by dividing the height² and lumbar height² (height_L1–L5). We also assessed the correlation between changes in VAT to total adipose tissue ratio (VA/TA index) and CD activity scores (5-point MRE classification) using Spearman’s correlation analysis. A multivariate linear regression model was used to adjust for the follow-up duration and treatment type.

Results

Overall, 49 patients (with 49 paired scans) were enrolled. VA/TA index changes were negatively correlated with changes in skeletal muscle index (SMI; r =  − 0.339, p < 0.05). The VA/TA index (52.69 ± 10.66% vs. 49.18 ± 10.80%, p < 0.001) and the total MRE score (8.0 ± 3.9 vs. 5.7 ± 3.4, p < 0.001) decreased significantly during follow-up, regardless of follow-up duration and treatment type (both p > 0.05). Changes in total MRE score were negatively correlated with SMI changes (r =  − 0.408, p < 0.01) but positively correlated with VA/TA index changes (r = 0.479, p < 0.01).

Conclusion

An increase in SMI and a decrease in VA/TA index could reflect improved nutritional and inflammatory status.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13244-021-01121-3.