Spectral CT Based Training Dataset Generation and Augmentation for Conventional CT Vascular Segmentation

Lartaud, Pierre-Jean; Rouchaud, A; Rouet, Jean-Michel; Nempont, Olivier; Boussel, Loîc

doi:10.1007/978-3-030-32245-8_85

Cited by 9 publications

(9 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Automatic methods for CCTA segmentation are not readily applicable to NCCT images as the contrast between tissues in these images is very different. 14 Moreover, retraining of these methods using manual reference segmentations in NCCT would be very challenging, as reference segmentations can hardly be obtained in NCCT due to the poor contrast between different cardiac tissues. Therefore, several works using reference segmentations obtained in NCCT have only focused on the aorta or the full heart, and not on cardiac chambers.…”

Section: Introductionmentioning

confidence: 99%

“…For this purpose, it would be valuable to perform automatic whole‐heart segmentation in NCCT images. Automatic methods for CCTA segmentation are not readily applicable to NCCT images as the contrast between tissues in these images is very different 14 . Moreover, retraining of these methods using manual reference segmentations in NCCT would be very challenging, as reference segmentations can hardly be obtained in NCCT due to the poor contrast between different cardiac tissues.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Deep learning from dual‐energy information for whole‐heart segmentation in dual‐energy and single‐energy non‐contrast‐enhanced cardiac CT

et al. 2020

View full text Add to dashboard Cite

Purpose: Deep learning-based whole-heart segmentation in coronary computed tomography angiography (CCTA) allows the extraction of quantitative imaging measures for cardiovascular risk prediction. Automatic extraction of these measures in patients undergoing only non-contrast-enhanced CT (NCCT) scanning would be valuable, but defining a manual reference standard that would allow training a deep learning-based method for whole-heart segmentation in NCCT is challenging, if not impossible. In this work, we leverage dual-energy information provided by a dual-layer detector CT scanner to obtain a reference standard in virtual non-contrast (VNC) CT images mimicking NCCT images, and train a three-dimensional (3D) convolutional neural network (CNN) for the segmentation of VNC as well as NCCT images. Methods: Eighteen patients were scanned with and without contrast enhancement on a dual-layer detector CT scanner. Contrast-enhanced acquisitions were reconstructed into a CCTA and a perfectly aligned VNC image. In each CCTA image, manual reference segmentations of the left ventricular (LV) myocardium, LV cavity, right ventricle, left atrium, right atrium, ascending aorta, and pulmonary artery trunk were obtained and propagated to the corresponding VNC image. These VNC images and reference segmentations were used to train 3D CNNs in a sixfold cross-validation for automatic segmentation in either VNC images or NCCT images reconstructed from the non-contrastenhanced acquisition. Automatic segmentation in VNC images was evaluated using the Dice similarity coefficient (DSC) and average symmetric surface distance (ASSD). Automatically determined volumes of the cardiac chambers and LV myocardium in NCCT were compared to reference volumes of the same patient in CCTA by Bland-Altman analysis. An additional independent multivendor multicenter set of single-energy NCCT images from 290 patients was used for qualitative analysis, in which two observers graded segmentations on a five-point scale. Results: Automatic segmentations in VNC images showed good agreement with reference segmentations, with an average DSC of 0.897 AE 0.034 and an average ASSD of 1.42 AE 0.45 mm. Volume differences [95% confidence interval] between automatic NCCT and reference CCTA segmentations were À19 [À67; 30] mL for LV myocardium, À25 [À78; 29] mL for LV cavity, À29 [À73; 14] mL for right ventricle, À20 [À62; 21] mL for left atrium, and À19 [À73; 34] mL for right atrium, respectively. In 214 (74%) NCCT images from the independent multivendor multicenter set, both observers agreed that the automatic segmentation was mostly accurate (grade 3) or better. Conclusion: Our automatic method produced accurate whole-heart segmentations in NCCT images using a CNN trained with VNC images from a dual-layer detector CT scanner. This method might enable quantification of additional cardiac measures from NCCT images for improved cardiovascular risk prediction.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Deep learning from dual‐energy information for whole‐heart segmentation in dual‐energy and single‐energy non‐contrast‐enhanced cardiac CT

et al. 2020

View full text Add to dashboard Cite

show abstract

“…trained a deep learning model for MRI reconstruction and were able to augment their data by undersampling the k‐space. A similar approach can be applied to images captured using a spectral (dual energy) CT scanner since two separate energy spectra are captured, these can be combined in different ways to produce different augmentations 171,172 . Omigbodun et al 173 .…”

Section: Methodsmentioning

confidence: 99%

“…A similar approach can be applied to images captured using a spectral (dual energy) CT scanner since two separate energy spectra are captured, these can be combined in different ways to produce different augmentations. 171,172 Omigbodun et al 173 also augmented images simulating different parameters available on a CT scanner, such as slice thickness or dose. Fig.…”

Section: Other Augmentation Techniquesmentioning

confidence: 99%

A review of medical image data augmentation techniques for deep learning applications

et al. 2021

View full text Add to dashboard Cite

Summary Research in artificial intelligence for radiology and radiotherapy has recently become increasingly reliant on the use of deep learning‐based algorithms. While the performance of the models which these algorithms produce can significantly outperform more traditional machine learning methods, they do rely on larger datasets being available for training. To address this issue, data augmentation has become a popular method for increasing the size of a training dataset, particularly in fields where large datasets aren’t typically available, which is often the case when working with medical images. Data augmentation aims to generate additional data which is used to train the model and has been shown to improve performance when validated on a separate unseen dataset. This approach has become commonplace so to help understand the types of data augmentation techniques used in state‐of‐the‐art deep learning models, we conducted a systematic review of the literature where data augmentation was utilised on medical images (limited to CT and MRI) to train a deep learning model. Articles were categorised into basic, deformable, deep learning or other data augmentation techniques. As artificial intelligence models trained using augmented data make their way into the clinic, this review aims to give an insight to these techniques and confidence in the validity of the models produced.

show abstract

“…[5][6][7][8][9][10] To overcome similar protocol variability issues, other imaging modalities benefit from dedicated augmentation workflow for segmentation. 11,12 Inspired by these techniques, approaches using spectral CT datasets to simulate various iodine concentrations from a single acquisition were recently proposed, 7,[13][14][15][16] making it possible to train segmentation networks to be efficient on multiple injection protocols and TNC scans. Dual-energy CT scanner is a new clinically available medical imaging technique that allows us-from a sin-gle contrast-enhanced scan-to generate not only conventional Hounsfield unit (HU) images like any other CT scanner, but also virtual-noncontrast (VNC) images.…”

Section: Introductionmentioning

confidence: 99%

A conventional‐to‐spectral CT image translation augmentation workflow for robust contrast injection‐independent organ segmentation

et al. 2021

Self Cite

View full text Add to dashboard Cite

In computed tomography (CT) cardiovascular imaging, the numerous contrast injection protocols used to enhance structures make it difficult to gather training datasets for deep learning applications supporting diverse protocols. Moreover, creating annotations on noncontrast scans is extremely tedious. Recently, spectral CT's virtual-noncontrast images (VNC) have been used as data augmentation to train segmentation networks performing on enhanced and true-noncontrast (TNC) scans alike, while improving results on protocols absent of their training dataset. However, spectral data are not widely available, making it difficult to gather specific datasets for each task. As a solution, we present a data augmentation workflow based on a trained image translation network, to bring spectral-like augmentation to any conventional CT dataset. Method: The conventional CT-to-spectral image translation network (HUSpect-Net) was first trained to generate VNC from conventional housnfied units images (HU), using an unannotated spectral dataset of 1830 patients. It was then tested on a second dataset of 300 spectral CT scans by comparing VNC generated through deep learning (VNC DL ) to their true counterparts. To illustrate and compare our workflow's efficiency with true spectral augmentation,HUSpectNet was applied to a third dataset of 112 spectral scans to generate VNC DL along HU and VNC images. Three different three-dimensional (3D) networks (U-Net, X-Net, and U-Net++) were trained for multilabel heart segmentation, following four augmentation strategies. As baselines, trainings were performed on contrasted images without (HUonly) and with conventional gray-values augmentation (HUaug). Then, the same networks were trained using a proportion of contrasted and VNC/VNC DL images (TrueSpec/GenSpec). Each training strategy applied to each architecture was evaluated using Dice coefficients on a fourth multicentric multivendor single-energy CT dataset of 121 patients, including different contrast injection protocols and unenhanced scans. The U-Net++ results were further explored with distance metrics on every label. Results: Tested on 300 full scans, our HUSpectNet translation network shows a mean absolute error of 6.70 ± 2.83 HU between VNC DL and VNC, while peak signal-to-noise ratio reaches 43.89 dB. GenSpec and TrueSpec show very close results regardless of the protocol and used architecture: mean Dice coefficients (DSC mean ) are equal with a margin of 0.006, ranging from 0.879 to 0.938. Their performances significantly increase on TNC scans (p-values < 0.017 for all architectures) compared to HUonly and HUaug, with DSC mean of 1108

show abstract

Spectral CT Based Training Dataset Generation and Augmentation for Conventional CT Vascular Segmentation

Cited by 9 publications

References 8 publications

Deep learning from dual‐energy information for whole‐heart segmentation in dual‐energy and single‐energy non‐contrast‐enhanced cardiac CT

Deep learning from dual‐energy information for whole‐heart segmentation in dual‐energy and single‐energy non‐contrast‐enhanced cardiac CT

A review of medical image data augmentation techniques for deep learning applications

A conventional‐to‐spectral CT image translation augmentation workflow for robust contrast injection‐independent organ segmentation

Contact Info

Product

Resources

About