An Overdetermined System of Transform Equations in Support of Robust DCE-MRI Registration With Outlier Rejection

Deep learning prediction of post-SBRT liver function changes and NTCP modeling in hepatocellular carcinoma based on DGAE-MRI

BACKGROUND

Stereotactic body radiation therapy (SBRT) produces excellent local control for patients with hepatocellular carcinoma (HCC). However, the risk of toxicity for normal liver tissue is still a limiting factor. Normal tissue complication probability (NTCP) models have been proposed to estimate the toxicity with the assumption of uniform liver function distribution, which is not optimal. With more accurate regional liver functional imaging available for individual patient, we can improve the estimation and be more patient-specific.

PURPOSE

To develop normal tissue complication probability (NTCP) models using pre-/during-treatment (RT) dynamic Gadoxetic Acid-enhanced (DGAE) MRI for adaptation of RT in a patient-specific manner in hepatocellular cancer (HCC) patients who receive SBRT.

METHODS

24 of 146 HCC patients who received SBRT underwent DGAE MRI. Physical doses were converted into EQD2 for analysis. Voxel-by-voxel quantification of the contrast uptake rate (k1) from DGAE-MRI was used to quantify liver function. A logistic dose-response model was used to estimate the fraction of liver functional loss, and NTCP was estimated using the cumulative functional reserve model for changes in Child-Pugh (C-P) scores. Model parameters were calculated using maximum-likelihood estimations. During-RT liver functional maps were predicted from dose distributions and pre-RT k1 maps with a conditional Wasserstein generative adversarial network (cWGAN). Imaging prediction quality was assessed using root-mean-square error (RMSE) and structural similarity (SSIM) metrics. The dose-response and NTCP were fit on both original and cWGAN predicted images and compared using a Wilcoxon signed-rank test.

RESULTS

Logistic dose response models for changes in k1 yielded D50 of 35.2 (95% CI: 26.7-47.5) Gy and k of 0.62 (0.49-0.75) for the whole population. The high baseline ALBI (poor liver function) subgroup showed a significantly smaller D50 of 11.7 (CI: 9.06-15.4) Gy and larger k of 0.96 (CI: 0.74-1.22) compared to a low baseline ALBI (good liver function) subgroup of 54.8 (CI: 38.3-79.1) Gy and 0.59 (CI: 0.48-0.74), with p-values of < 0.001 and = 0.008, respectively, which indicates higher radiosensitivity for the worse baseline liver function cohort. Subset analyses were also performed for high/low baseline CP subgroups. The corresponding NTCP models showed good agreement for the fit parameters between cWGAN predicted and the ground-truth during-RT images with no statistical differences for low ALBI subgroup.

CONCLUSIONS

NTCP models which incorporate voxel-wise functional information from DGAE-MRI k1 maps were successfully developed and feasibility was demonstrated in a small patient cohort. cWGAN predicted functional maps show promise for estimating localized patient-specific response to RT and warrant further validation in a larger patient cohort.

The Effect of Stereotactic Body Radiation Therapy for Hepatocellular Cancer on Regional Hepatic Liver Function

PURPOSE

To investigate direct radiation dose-related and inflammation-mediated regional hepatic function losses after stereotactic body radiation therapy (SBRT) in patients with hepatocellular carcinoma (HCC) and poor liver function.

METHODS AND MATERIALS

Twenty-four patients with HCC enrolled on an IRB-approved adaptive SBRT trial had liver dynamic gadoxetic acid-enhanced magnetic resonance imaging and blood sample collections before and 1 month after SBRT. Gadoxetic acid uptake rate (k1) maps were quantified for regional hepatic function and coregistered to both 2-Gy equivalent dose and physical dose distributions. Regional k1 loss patterns from before to after SBRT were analyzed for effects of dose and patient using a mixed-effects model and logistic function and were associated with pretherapy liver-function albumin-bilirubin scores. Plasma levels of tumor necrosis factor α receptor 1 (TNFR1), an inflammation marker, were correlated with mean k1 losses in the lowest dose regions by Spearman rank correlation.

RESULTS

The whole group had a k1 loss rate of 0.4%/Gy (2-Gy equivalent dose); however, there was a significant random effect of patient in the mixed-effect model (P < .05). Patients with poor and good liver functions lost 50% of k1 values at 12.5 and 57.2 Gy and 33% and 16% of k1 values at the lowest dose regions (<5 Gy), respectively. The k1 losses at the lowest dose regions of individual patients were significantly correlated with their TNFR1 levels after SBRT (P < .02).

CONCLUSIONS

The findings suggest that regional hepatic function losses after SBRT in patients with HCC include both direct radiation dose-dependent and inflammation-mediated effects, which could influence how to manage these patients to preserve their liver function after SBRT.

GAN and dual-input two-compartment model-based training of a neural network for robust quantification of contrast uptake rate in gadoxetic acid-enhanced MRI

PURPOSE

Gadoxetic acid uptake rate (k₁ ) obtained from dynamic, contrast-enhanced (DCE) magnetic resonance imaging (MRI) is a promising measure of regional liver function. Clinical exams are typically poorly temporally characterized, as seen in a low temporal resolution (LTR) compared to high temporal resolution (HTR) experimental acquisitions. Meanwhile, clinical demands incentivize shortening these exams. This study develops a neural network-based approach to quantitation of k₁ , for increased robustness over current models such as the linearized single-input, two-compartment (LSITC) model.

METHODS

Thirty Liver HTR DCE MRI exams were acquired in 22 patients with at least 16 min of postcontrast data sampled at least every 13 s. A simple neural network (NN) with four hidden layers was trained on voxel-wise LTR data to predict k₁ . Low temporal resolution data were created by subsampling HTR data to contain six time points, replicating the characteristics of clinical LTR data. Both the total length and the placement of points in the training data were varied considerably to encourage robustness to variation. A generative adversarial network (GAN) was used to generate arterial and portal venous inputs for use in data augmentation based on the dual-input, two-compartment, pharmacokinetic model of gadoxetic acid in the liver. The performance of the NN was compared to direct application of LSITC on both LTR and HTR data. The error was assessed when subsampling lengths from 16 to 4 min, enabling assessment of robustness to acquisition length.

RESULTS

For acquisition lengths of 16 min NRMSE (Normalized Root-Mean-Squared Error) in k₁ was 0.60, 1.77, and 1.21, for LSITC applied to HTR data, LSITC applied to LTR data, and GAN-augmented NN applied to LTR data, respectively. As the acquisition length was shortened, errors greatly increased for LSITC approaches by several folds. For acquisitions shorter than 12 min the GAN-augmented NN approach outperformed the LSITC approach to a statistically significant extent, even with HTR data.

CONCLUSIONS

The study indicates that data length is significant for LSITC analysis as applied to DCE data for standard temporal sampling, and that machine learning methods, such as the implemented NN, have potential for much greater resilience to shortened acquisition time than directly fitting to the LSITC model.

Modeling of Normal Tissue Complications Using Imaging and Biomarkers After Radiation Therapy for Hepatocellular Carcinoma

PURPOSE

To develop normal tissue complications (NTCP) models for hepatocellular cancer (HCC) patients who undergo liver radiation therapy (RT) and to evaluate the potential role of functional imaging and measurement of blood-based circulating biological markers before and during RT to improve the performance of these models.

METHODS AND MATERIALS

The data from 192 HCC patients who had undergone RT from 2005 to 2014 were evaluated. Of the 192 patients, 146 had received stereotactic body RT (SBRT) and 46 had received conventional RT to a median physical tumor dose of 49.8 Gy and 50.4 Gy, respectively. The physical doses were converted into 2-Gy equivalents for analysis. Two approaches were investigated for modeling NTCP: (1) a generalized Lyman-Kutcher-Burman model; and (2) a generalization of the parallel architecture model. Three clinical endpoints were considered: the change in albumin-bilirubin (ALBI), change in Child-Pugh (C-P) score, and grade ≥3 liver enzymatic changes. Local dynamic contrast-enhanced magnetic resonance imaging portal venous perfusion information was used as an imaging biomarker for local liver function. Four candidate inflammatory cytokines were considered as biological markers. The imaging findings and cytokine levels were incorporated into NTCP modeling, and their role was evaluated using goodness-of-fit metrics.

RESULTS

Using dosimetric information only, the Lyman-Kutcher-Burman model for the ALBI/C-P change had a steeper response curve compared with grade ≥3 enzymatic changes. Incorporating portal venous perfusion imaging information into the parallel architecture model to represent functional reserve resulted in relatively steeper dose-response curves compared with dose-only models. A larger loss of perfusion function was needed for enzymatic changes compared with ALBI/C-P changes. Increased transforming growth factor-β1 and eotaxin expression increased the trend of expected risk in both NTCP modeling approaches but did not reach statistical significance.

CONCLUSIONS

The incorporation of imaging findings and biological markers into NTCP modeling of liver toxicity improved the estimates of expected NTCP risk compared with using dose-only models. In addition, such generalized NTCP models should contribute to a better understanding of the normal tissue response in HCC SBRT patients and facilitate personalized treatment.

Abdominal DCE-MRI reconstruction with deformable motion correction for liver perfusion quantification

PURPOSE

Abdominal dynamic contrast-enhanced (DCE) MRI suffers from motion-induced artifacts that can blur images and distort contrast-agent uptake curves. For liver perfusion analysis, image reconstruction with rigid-body motion correction (RMC) can restore distorted portal-venous input functions (PVIF) to higher peak amplitudes. However, RMC cannot correct for liver deformation during breathing. We present a reconstruction algorithm with deformable motion correction (DMC) that enables correction of breathing-induced deformation in the whole abdomen.

METHODS

Raw data from a golden-angle stack-of-stars gradient-echo sequence were collected for 54 DCE-MRI examinations of 31 patients. For each examination, a respiratory motion signal was extracted from the data and used to reconstruct 21 breathing states from inhale to exhale. The states were aligned with deformable image registration to the end-exhale state. Resulting deformation fields were used to correct back-projection images before reconstruction with view sharing. Images with DMC were compared to uncorrected images and images with RMC.

RESULTS

DMC significantly increased the PVIF peak amplitude compared to uncorrected images (P << 0.01, mean increase: 8%) but not compared to RMC. The increased PVIF peak amplitude significantly decreased estimated portal-venous perfusion in the liver (P << 0.01, mean decrease: 8 ml/(100 ml·min)). DMC also removed artifacts in perfusion maps at the liver edge and reduced blurring of liver tumors for some patients.

CONCLUSIONS

DCE-MRI reconstruction with DMC can restore motion-distorted uptake curves in the abdomen and remove motion artifacts from reconstructed images and parameter maps but does not significantly improve perfusion quantification in the liver compared to RMC.

Quantification of liver function by linearization of a two-compartment model of gadoxetic acid uptake using dynamic contrast-enhanced magnetic resonance imaging

Dynamic gadoxetic acid-enhanced magnetic resonance imaging (MRI) allows the investigation of liver function through the observation of the perfusion and uptake of contrast agent in the parenchyma. Voxel-by-voxel quantification of the contrast uptake rate (k₁ ) from dynamic gadoxetic acid-enhanced MRI through the standard dual-input, two-compartment model could be susceptible to overfitting of variance in the data. The aim of this study was to develop a linearized, but more robust, model. To evaluate the estimated k₁ values using this linearized analysis, high-temporal-resolution gadoxetic acid-enhanced MRI scans were obtained in 13 examinations, and k₁ maps were created using both models. Comparison of liver k₁ values estimated from the two methods produced a median correlation coefficient of 0.91 across the 12 scans that could be used. Temporally sparse clinical MRI data with gadoxetic acid uptake were also employed to create k₁ maps of 27 examinations using the linearized model. Of 20 scans, the created k₁ maps were compared with overall liver function as measured by indocyanine green (ICG) retention, and yielded a correlation coefficient of 0.72. In the 27 k₁ maps created via the linearized model, the mean liver k₁ value was 3.93 ± 1.79 mL/100 mL/min, consistent with previous studies. The results indicate that the linearized model provides a simple and robust method for the assessment of the rate of contrast uptake that can be applied to both high-temporal-resolution dynamic contrast-enhanced MRI and typical clinical multiphase MRI data, and that correlates well with the results of both two-compartment analysis and independent whole liver function measurements.

Rigid-body motion correction of the liver in image reconstruction for golden-angle stack-of-stars DCE MRI

PURPOSE

Respiratory motion can affect pharmacokinetic perfusion parameters quantified from liver dynamic contrast-enhanced MRI. Image registration can be used to align dynamic images after reconstruction. However, intra-image motion blur remains after alignment and can alter the shape of contrast-agent uptake curves. We introduce a method to correct for inter- and intra-image motion during image reconstruction.

METHODS

Sixteen liver dynamic contrast-enhanced MRI examinations of nine subjects were performed using a golden-angle stack-of-stars sequence. For each examination, an image time series with high temporal resolution but severe streak artifacts was reconstructed. Images were aligned using region-limited rigid image registration within a region of interest covering the liver. The transformations resulting from alignment were used to correct raw data for motion by modulating and rotating acquired lines in k-space. The corrected data were then reconstructed using view sharing.

RESULTS

Portal-venous input functions extracted from motion-corrected images had significantly greater peak signal enhancements (mean increase: 16%, t-test, P <  0.001) than those from images aligned using image registration after reconstruction. In addition, portal-venous perfusion maps estimated from motion-corrected images showed fewer artifacts close to the edge of the liver.

CONCLUSIONS

Motion-corrected image reconstruction restores uptake curves distorted by motion. Motion correction also reduces motion artifacts in estimated perfusion parameter maps. Magn Reson Med 79:1345-1353, 2018. © 2017 International Society for Magnetic Resonance in Medicine.