Dataset Information

Automated Contouring of Contrast and Noncontrast Computed Tomography Liver Images With Fully Convolutional Networks.

ABSTRACT:

Purpose

The deformable nature of the liver can make focal treatment challenging and is not adequately addressed with simple rigid registration techniques. More advanced registration techniques can take deformations into account (eg, biomechanical modeling) but require segmentations of the whole liver for each scan, which is a time-intensive process. We hypothesize that fully convolutional networks can be used to rapidly and accurately autosegment the liver, removing the temporal bottleneck for biomechanical modeling.

Methods and materials

Manual liver segmentations on computed tomography scans from 183 patients treated at our institution and 30 scans from the Medical Image Computing & Computer Assisted Intervention challenges were collected for this study. Three architectures were investigated for rapid automated segmentation of the liver (VGG-16, DeepLabv3 +, and a 3-dimensional UNet). Fifty-six cases were set aside as a final test set for quantitative model evaluation. Accuracy of the autosegmentations was assessed using Dice similarity coefficient and mean surface distance. Qualitative evaluation was also performed by 3 radiation oncologists on 50 independent cases with previously clinically treated liver contours.

Results

The mean (minimum-maximum) mean surface distance for the test groups with the final model, DeepLabv3 +, were as follows: ?_{Contrast(N = 17)}: 0.99 mm (0.47-2.2), ?_{Non_Contrast(N = 19)l}: 1.12 mm (0.41-2.87), and ?_{Miccai(N = 30)t}: 1.48 mm (0.82-3.96). The qualitative evaluation showed that 30 of 50 autosegmentations (60%) were preferred to manual contours (majority voting) in a blinded comparison, and 48 of 50 autosegmentations (96%) were deemed clinically acceptable by at least 1 reviewing physician.

Conclusions

The autosegmentations were preferred compared with manually defined contours in the majority of cases. The ability to rapidly segment the liver with high accuracy achieved in this investigation has the potential to enable the efficient integration of biomechanical model-based registration into a clinical workflow.

SUBMITTER: Anderson BM

PROVIDER: S-EPMC7807136 | biostudies-literature | 2021 Jan-Feb

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Automated Contouring of Contrast and Noncontrast Computed Tomography Liver Images With Fully Convolutional Networks.

Anderson Brian M BM Lin Ethan Y EY Cardenas Carlos E CE Gress Dustin A DA Erwin William D WD Odisio Bruno C BC Koay Eugene J EJ Brock Kristy K KK

Advances in radiation oncology 20200516 1

<h4>Purpose</h4>The deformable nature of the liver can make focal treatment challenging and is not adequately addressed with simple rigid registration techniques. More advanced registration techniques can take deformations into account (eg, biomechanical modeling) but require segmentations of the whole liver for each scan, which is a time-intensive process. We hypothesize that fully convolutional networks can be used to rapidly and accurately autosegment the liver, removing the temporal bottlene ...[more]

PMID: 33490720

Similar Datasets

Project description:Cardiovascular resonance (CMR) imaging is a standard imaging modality for assessing cardiovascular diseases (CVDs), the leading cause of death globally. CMR enables accurate quantification of the cardiac chamber volume, ejection fraction and myocardial mass, providing information for diagnosis and monitoring of CVDs. However, for years, clinicians have been relying on manual approaches for CMR image analysis, which is time consuming and prone to subjective errors. It is a major clinical challenge to automatically derive quantitative and clinically relevant information from CMR images.Deep neural networks have shown a great potential in image pattern recognition and segmentation for a variety of tasks. Here we demonstrate an automated analysis method for CMR images, which is based on a fully convolutional network (FCN). The network is trained and evaluated on a large-scale dataset from the UK Biobank, consisting of 4,875 subjects with 93,500 pixelwise annotated images. The performance of the method has been evaluated using a number of technical metrics, including the Dice metric, mean contour distance and Hausdorff distance, as well as clinically relevant measures, including left ventricle (LV) end-diastolic volume (LVEDV) and end-systolic volume (LVESV), LV mass (LVM); right ventricle (RV) end-diastolic volume (RVEDV) and end-systolic volume (RVESV).By combining FCN with a large-scale annotated dataset, the proposed automated method achieves a high performance in segmenting the LV and RV on short-axis CMR images and the left atrium (LA) and right atrium (RA) on long-axis CMR images. On a short-axis image test set of 600 subjects, it achieves an average Dice metric of 0.94 for the LV cavity, 0.88 for the LV myocardium and 0.90 for the RV cavity. The mean absolute difference between automated measurement and manual measurement is 6.1 mL for LVEDV, 5.3 mL for LVESV, 6.9 gram for LVM, 8.5 mL for RVEDV and 7.2 mL for RVESV. On long-axis image test sets, the average Dice metric is 0.93 for the LA cavity (2-chamber view), 0.95 for the LA cavity (4-chamber view) and 0.96 for the RA cavity (4-chamber view). The performance is comparable to human inter-observer variability.We show that an automated method achieves a performance on par with human experts in analysing CMR images and deriving clinically relevant measures.

Project description:PurposeComputed tomography (CT)-derived ventilation methods compute respiratory induced volume changes as a surrogate for pulmonary ventilation. Currently, there are no known methods to derive perfusion information from noncontrast CT. We introduce a novel CT-Perfusion (CT-P) method for computing the magnitude mass changes apparent on dynamic noncontrast CT as a surrogate for pulmonary perfusion.MethodsCT-Perfusion is based on a mass conservation model which describes the unknown mass change as a linear combination of spatially corresponding inhale and exhale HU estimated voxel densities. CT-P requires a deformable image registration (DIR) between the inhale/exhale lung CT pair, a preprocessing lung volume segmentation, and an estimate for the Jacobian of the DIR transformation. Given this information, the CT-P image, which provides the magnitude mass change for each voxel within the lung volume, is formulated as the solution to a constrained linear least squares problem defined by a series of subregional mean magnitude mass change measurements. Similar to previous robust CT-ventilation methods, the amount of uncertainty in a subregional sample mean measurement is related to measurement resolution and can be characterized with respect to a tolerance parameter τ . Spatial Spearman correlation between single photon emission CT perfusion (SPECT-P) and the proposed CT-P method was assessed in two patient cohorts via a parameter sweep of τ . The first cohort was comprised of 15 patients diagnosed with pulmonary embolism (PE) who had SPECT-P and 4DCT imaging acquired within 24 h of PE diagnosis. The second cohort was comprised of 15 nonsmall cell lung cancer patients who had SPECT-P and 4DCT images acquired prior to radiotherapy. For each test case, CT-P images were computed for 30 different uncertainty parameter values, uniformly sampled from the range [0.01, 0.125], and the Spearman correlation between the SPECT-P and the resulting CT-P images were computed.ResultsThe median correlations between CT-P and SPECT-P taken over all 30 test cases ranged between 0.49 and 0.57 across the parameter sweep. For the optimal tolerance τ = 0.0385, the CT-P and SPECT-P correlations across all 30 test cases ranged between 0.02 and 0.82. A one-sample sign test was applied separately to the PE and lung cancer cohorts. A low Spearmen correlation of 15% was set as the null median value and two-sided alternative was tested. The PE patients showed a median correlation of 0.57 (IQR = 0.305). One-sample sign test was statistically significant with 96.5 % confidence interval: 0.20-0.63, P < 0.00001. Lung cancer patients had a median correlation of 0.57(IQR = 0.230). Again, a one-sample sign test for median was statistically significant with 96.5 percent confidence interval: 0.45-0.71, P < 0.00001.ConclusionCT-Perfusion is the first mechanistic model designed to quantify magnitude blood mass changes on noncontrast dynamic CT as a surrogate for pulmonary perfusion. While the reported correlations with SPECT-P are promising, further investigation is required to determine the optimal CT acquisition protocol and numerical method implementation for CT-P imaging.

Project description:BackgroundCardiovascular magnetic resonance (CMR) myocardial native T1 mapping allows assessment of interstitial diffuse fibrosis. In this technique, the global and regional T1 are measured manually by drawing region of interest in motion-corrected T1 maps. The manual analysis contributes to an already lengthy CMR analysis workflow and impacts measurements reproducibility. In this study, we propose an automated method for combined myocardium segmentation, alignment, and T1 calculation for myocardial T1 mapping.MethodsA deep fully convolutional neural network (FCN) was used for myocardium segmentation in T1 weighted images. The segmented myocardium was then resampled on a polar grid, whose origin is located at the center-of-mass of the segmented myocardium. Myocardium T1 maps were reconstructed from the resampled T1 weighted images using curve fitting. The FCN was trained and tested using manually segmented images for 210 patients (5 slices, 11 inversion times per patient). An additional image dataset for 455 patients (5 slices and 11 inversion times per patient), analyzed by an expert reader using a semi-automatic tool, was used to validate the automatically calculated global and regional T1 values. Bland-Altman analysis, Pearson correlation coefficient, r, and the Dice similarity coefficient (DSC) were used to evaluate the performance of the FCN-based analysis on per-patient and per-slice basis. Inter-observer variability was assessed using intraclass correlation coefficient (ICC) of the T1 values calculated by the FCN-based automatic method and two readers.ResultsThe FCN achieved fast segmentation (<?0.3?s/image) with high DSC (0.85?±?0.07). The automatically and manually calculated T1 values (1091?±?59?ms and 1089?±?59?ms, respectively) were highly correlated in per-patient (r?=?0.82; slope?=?1.01; p?<?0.0001) and per-slice (r?=?0.72; slope?=?1.01; p?<?0.0001) analyses. Bland-Altman analysis showed good agreement between the automated and manual measurements with 95% of measurements within the limits-of-agreement in both per-patient and per-slice analyses. The intraclass correllation of the T1 calculations by the automatic method vs reader 1 and reader 2 was respectively 0.86/0.56 and 0.74/0.49 in the per-patient/per-slice analyses, which were comparable to that between two expert readers (=0.72/0.58 in per-patient/per-slice analyses).ConclusionThe proposed FCN-based image processing platform allows fast and automatic analysis of myocardial native T1 mapping images mitigating the burden and observer-related variability of manual analysis.

Dataset Information

Automated Contouring of Contrast and Noncontrast Computed Tomography Liver Images With Fully Convolutional Networks.

Purpose

Methods and materials

Results

Conclusions

Publications

Automated Contouring of Contrast and Noncontrast Computed Tomography Liver Images With Fully Convolutional Networks.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure