Automatic calculation of myocardial perfusion reserve using deep learning with uncertainty quantification.
Ontology highlight
ABSTRACT: Background
Myocardial perfusion reserve index (MPRI) in magnetic resonance imaging (MRI) is an important indicator of ischemia, and its measurement typically involves manual procedures. The purposes of this study were to develop a fully automatic method for estimating the MPRI and to evaluate its performance.Methods
The method consisted of segmenting the myocardium in dynamic contrast-enhanced (DCE) myocardial perfusion MRI data using Monte Carlo dropout U-Net, dividing the myocardium into segments based on landmark localization with machine learning, and estimating the MPRI after the calculation of the left ventricular and myocardial contrast upslopes. The proposed method was compared with a reference method, which involved manual adjustments of the myocardial contours and upslope ranges.Results
In test subjects, MPRIs measured by the proposed technique correlated with those by the manual reference in segmental assessment [intraclass correlation coefficient (ICC) =0.75, 95% CI: 0.70-0.79, P<0.001]. The automatic and reference MPRI values showed a mean difference of -0.02 and 95% limits of agreement of (-0.86, 0.82).Conclusions
The proposed automatic method is based on deep learning segmentation and machine learning landmark detection for MPRI measurements in DCE perfusion MRI. It holds the potential to efficiently and quantitatively assess myocardial ischemia without any user's interaction.
SUBMITTER: Kim YC
PROVIDER: S-EPMC10722070 | biostudies-literature | 2023 Dec
REPOSITORIES: biostudies-literature
Publications
Automatic calculation of myocardial perfusion reserve using deep learning with uncertainty quantification.
Kim Yoon-Chul YC Kim Kyurae K Choe Yeon Hyeon YH
Quantitative imaging in medicine and surgery 20231010 12
<h4>Background</h4>Myocardial perfusion reserve index (MPRI) in magnetic resonance imaging (MRI) is an important indicator of ischemia, and its measurement typically involves manual procedures. The purposes of this study were to develop a fully automatic method for estimating the MPRI and to evaluate its performance.<h4>Methods</h4>The method consisted of segmenting the myocardium in dynamic contrast-enhanced (DCE) myocardial perfusion MRI data using Monte Carlo dropout U-Net, dividing the myoca ...[more]
Similar Datasets
Project description:PurposeTo apply deep convolution neural network to the segmentation task in myocardial arterial spin labeled perfusion imaging and to develop methods that measure uncertainty and that adapt the convolution neural network model to a specific false-positive versus false-negative tradeoff.MethodsThe Monte Carlo dropout U-Net was trained on data from 22 subjects and tested on data from 6 heart transplant recipients. Manual segmentation and regional myocardial blood flow were available for comparison. We consider 2 global uncertainty measures, named "Dice uncertainty" and "Monte Carlo dropout uncertainty," which were calculated with and without the use of manual segmentation, respectively. Tversky loss function with a hyperparameter β was used to adapt the model to a specific false-positive versus false-negative tradeoff.ResultsThe Monte Carlo dropout U-Net achieved a Dice coefficient of 0.91 ± 0.04 on the test set. Myocardial blood flow measured using automatic segmentations was highly correlated to that measured using the manual segmentation (R2 = 0.96). Dice uncertainty and Monte Carlo dropout uncertainty were in good agreement (R2 = 0.64). As β increased, the false-positive rate systematically decreased and false-negative rate systematically increased.ConclusionWe demonstrate the feasibility of deep convolution neural network for automatic segmentation of myocardial arterial spin labeling, with good accuracy. We also introduce 2 simple methods for assessing model uncertainty. Finally, we demonstrate the ability to adapt the convolution neural network model to a specific false-positive versus false-negative tradeoff. These findings are directly relevant to automatic segmentation in quantitative cardiac MRI and are broadly applicable to automatic segmentation problems in diagnostic imaging.
| S-EPMC6989045 | biostudies-literature
Project description:PurposeClinical applicability of renal arterial spin labeling (ASL) MRI is hampered because of time consuming and observer dependent post-processing, including manual segmentation of the cortex to obtain cortical renal blood flow (RBF). Machine learning has proven its value in medical image segmentation, including the kidneys. This study presents a fully automatic workflow for renal cortex perfusion quantification by including machine learning-based segmentation.MethodsFully automatic workflow was achieved by construction of a cascade of 3 U-nets to replace manual segmentation in ASL quantification. All 1.5T ASL-MRI data, including M0 , T1 , and ASL label-control images, from 10 healthy volunteers was used for training (dataset 1). Trained cascade performance was validated on 4 additional volunteers (dataset 2). Manual segmentations were generated by 2 observers, yielding reference and second observer segmentations. To validate the intended use of the automatic segmentations, manual and automatic RBF values in mL/min/100 g were compared.ResultsGood agreement was found between automatic and manual segmentations on dataset 1 (dice score = 0.78 ± 0.04), which was in line with inter-observer variability (dice score = 0.77 ± 0.02). Good agreement was confirmed on dataset 2 (dice score = 0.75 ± 0.03). Moreover, similar cortical RBF was obtained with automatic or manual segmentations, on average and at subject level; with 211 ± 31 mL/min/100 g and 208 ± 31 mL/min/100 g (P < .05), respectively, with narrow limits of agreement at -11 and 4.6 mL/min/100 g. RBF accuracy with automated segmentations was confirmed on dataset 2.ConclusionOur proposed method automates ASL quantification without compromising RBF accuracy. With quick processing and without observer dependence, renal ASL-MRI is more attractive for clinical application as well as for longitudinal and multi-center studies.
| S-EPMC9297892 | biostudies-literature
Project description:IntroductionMachine learning algorithms are expected to work side-by-side with humans in decision-making pipelines. Thus, the ability of classifiers to make reliable decisions is of paramount importance. Deep neural networks (DNNs) represent the state-of-the-art models to address real-world classification. Although the strength of activation in DNNs is often correlated with the network's confidence, in-depth analyses are needed to establish whether they are well calibrated.MethodIn this paper, we demonstrate the use of DNN-based classification tools to benefit cancer registries by automating information extraction of disease at diagnosis and at surgery from electronic text pathology reports from the US National Cancer Institute (NCI) Surveillance, Epidemiology, and End Results (SEER) population-based cancer registries. In particular, we introduce multiple methods for selective classification to achieve a target level of accuracy on multiple classification tasks while minimizing the rejection amount-that is, the number of electronic pathology reports for which the model's predictions are unreliable. We evaluate the proposed methods by comparing our approach with the current in-house deep learning-based abstaining classifier.ResultsOverall, all the proposed selective classification methods effectively allow for achieving the targeted level of accuracy or higher in a trade-off analysis aimed to minimize the rejection rate. On in-distribution validation and holdout test data, with all the proposed methods, we achieve on all tasks the required target level of accuracy with a lower rejection rate than the deep abstaining classifier (DAC). Interpreting the results for the out-of-distribution test data is more complex; nevertheless, in this case as well, the rejection rate from the best among the proposed methods achieving 97% accuracy or higher is lower than the rejection rate based on the DAC.ConclusionsWe show that although both approaches can flag those samples that should be manually reviewed and labeled by human annotators, the newly proposed methods retain a larger fraction and do so without retraining-thus offering a reduced computational cost compared with the in-house deep learning-based abstaining classifier.
| S-EPMC11467893 | biostudies-literature
Project description:BackgroundFully automatic analysis of myocardial perfusion MRI datasets enables rapid and objective reporting of stress/rest studies in patients with suspected ischemic heart disease. Developing deep learning techniques that can analyze multi-center datasets despite limited training data and variations in software (pulse sequence) and hardware (scanner vendor) is an ongoing challenge.MethodsDatasets from 3 medical centers acquired at 3T (n = 150 subjects; 21,150 first-pass images) were included: an internal dataset (inD; n = 95) and two external datasets (exDs; n = 55) used for evaluating the robustness of the trained deep neural network (DNN) models against differences in pulse sequence (exD-1) and scanner vendor (exD-2). A subset of inD (n = 85) was used for training/validation of a pool of DNNs for segmentation, all using the same spatiotemporal U-Net architecture and hyperparameters but with different parameter initializations. We employed a space-time sliding-patch analysis approach that automatically yields a pixel-wise "uncertainty map" as a byproduct of the segmentation process. In our approach, dubbed Data Adaptive Uncertainty-Guided Space-time (DAUGS) analysis, a given test case is segmented by all members of the DNN pool and the resulting uncertainty maps are leveraged to automatically select the "best" one among the pool of solutions. For comparison, we also trained a DNN using the established approach with the same settings (hyperparameters, data augmentation, etc.).ResultsThe proposed DAUGS analysis approach performed similarly to the established approach on the internal dataset (Dice score for the testing subset of inD: 0.896 ± 0.050 vs. 0.890 ± 0.049; p = n.s.) whereas it significantly outperformed on the external datasets (Dice for exD-1: 0.885 ± 0.040 vs. 0.849 ± 0.065, p < 0.005; Dice for exD-2: 0.811 ± 0.070 vs. 0.728 ± 0.149, p < 0.005). Moreover, the number of image series with "failed" segmentation (defined as having myocardial contours that include bloodpool or are noncontiguous in ≥1 segment) was significantly lower for the proposed vs. the established approach (4.3% vs. 17.1%, p < 0.0005).ConclusionsThe proposed DAUGS analysis approach has the potential to improve the robustness of deep learning methods for segmentation of multi-center stress perfusion datasets with variations in the choice of pulse sequence, site location or scanner vendor.
| S-EPMC11326424 | biostudies-literature
Project description:Emerging deep-learning (DL)-based techniques have significant potential to revolutionize biomedical imaging. However, one outstanding challenge is the lack of reliability assessment in the DL predictions, whose errors are commonly revealed only in hindsight. Here, we propose a new Bayesian convolutional neural network (BNN)-based framework that overcomes this issue by quantifying the uncertainty of DL predictions. Foremost, we show that BNN-predicted uncertainty maps provide surrogate estimates of the true error from the network model and measurement itself. The uncertainty maps characterize imperfections often unknown in real-world applications, such as noise, model error, incomplete training data, and out-of-distribution testing data. Quantifying this uncertainty provides a per-pixel estimate of the confidence level of the DL prediction as well as the quality of the model and data set. We demonstrate this framework in the application of large space-bandwidth product phase imaging using a physics-guided coded illumination scheme. From only five multiplexed illumination measurements, our BNN predicts gigapixel phase images in both static and dynamic biological samples with quantitative credibility assessment. Furthermore, we show that low-certainty regions can identify spatially and temporally rare biological phenomena. We believe our uncertainty learning framework is widely applicable to many DL-based biomedical imaging techniques for assessing the reliability of DL predictions.
| S-EPMC8329751 | biostudies-literature
Project description:BackgroundQuantitative myocardial perfusion cardiac MRI can provide a fast and robust assessment of myocardial perfusion status for the noninvasive diagnosis of myocardial ischemia while being more objective than visual assessment. However, it currently has limited use in clinical practice due to the challenging postprocessing required, particularly the segmentation.PurposeTo evaluate the efficacy of an automated deep learning (DL) pipeline for image processing prior to quantitative analysis.Study typeRetrospective.PopulationIn all, 175 (350 MRI scans; 1050 image series) clinical patients under both rest and stress conditions (135/10/30 training/validation/test).Field strength/sequence3.0T/2D multislice saturation recovery T1 -weighted gradient echo sequence.AssessmentAccuracy was assessed, as compared to the manual operator, through the mean square error of the distance between landmarks and the Dice similarity coefficient of the segmentation and bounding box detection. Quantitative perfusion maps obtained using the automated DL-based processing were compared to the results obtained with the manually processed images.Statistical testsBland-Altman plots and intraclass correlation coefficient (ICC) were used to assess the myocardial blood flow (MBF) obtained using the automated DL pipeline, as compared to values obtained by a manual operator.ResultsThe mean (SD) error in the detection of the time of peak signal enhancement in the left ventricle was 1.49 (1.4) timeframes. The mean (SD) Dice similarity coefficients for the bounding box and myocardial segmentation were 0.93 (0.03) and 0.80 (0.06), respectively. The mean (SD) error in the RV insertion point was 2.8 (1.8) mm. The Bland-Altman plots showed a bias of 2.6% of the mean MBF between the automated and manually processed MBF values on a per-myocardial segment basis. The ICC was 0.89, 95% confidence interval = [0.87, 0.90].Data conclusionWe showed high accuracy, compared to manual processing, for the DL-based processing of myocardial perfusion data leading to quantitative values that are similar to those achieved with manual processing.Level of evidence3 Technical Efficacy Stage: 1 J. Magn. Reson. Imaging 2020;51:1689-1696.
| S-EPMC7317373 | biostudies-literature
Project description:BackgroundAdenosine cardiovascular magnetic resonance (CMR) can accurately quantify myocardial perfusion reserve. While regadenoson is increasingly employed due to ease of use, imaging protocols have not been standardized. We sought to determine the optimal regadenoson CMR protocol for quantifying myocardial perfusion reserve index (MPRi) - more specifically, whether regadenoson stress imaging should be performed before or after rest imaging.MethodsTwenty healthy subjects underwent CMR perfusion imaging during resting conditions, during regadenoson-induced hyperemia (0.4 mg), and after 15 min of recovery. In 10/20 subjects, recovery was facilitated with aminophylline (125 mg). Myocardial time-intensity curves were used to obtain left ventricular cavity-normalized myocardial up-slopes. MPRi was calculated in two different ways: as the up-slope ratio of stress to rest (MPRi-rest), and the up-slope ratio of stress to recovery (MPRi-recov).ResultsIn all 20 subjects, MPRi-rest was 1.78 ± 0.60. Recovery up-slope did not return to resting levels, regardless of aminophylline use. Among patients not receiving aminophylline, MPRi-recov was 36 ± 16% lower than MPRi-rest (1.13 ± 0.38 vs. 1.82 ± 0.73, P = 0.001). In the 10 patients whose recovery was facilitated with aminophylline, MPRi-recov was 20 ± 24% lower than MPRi-rest (1.40 ± 0.35 vs. 1.73 ± 0.43, P = 0.04), indicating incomplete reversal. In 3 subjects not receiving aminophylline and 4 subjects receiving aminophylline, up-slope at recovery was greater than at stress, suggesting delayed maximal hyperemia.ConclusionsMPRi measurements from regadenoson CMR are underestimated if recovery perfusion is used as a substitute for resting perfusion, even when recovery is facilitated with aminophylline. True resting images should be used to allow accurate MPRi quantification. The delayed maximal hyperemia observed in some subjects deserves further study.Trial registrationClinicalTrials.gov NCT00871260.
| S-EPMC3552720 | biostudies-literature
Project description:BackgroundDeep learning (DL)-based denoising has been proven to improve image quality and quantitation accuracy of low dose (LD) SPECT. However, conventional DL-based methods used SPECT images with mixed frequency components. This work aims to develop an integrated multi-frequency denoising network to further enhance LD myocardial perfusion (MP) SPECT denoising.MethodsFifty anonymized patients who underwent routine 99mTc-sestamibi stress SPECT/CT scans were retrospectively recruited. Three LD datasets were obtained by reducing the 10 s acquisition time of full dose (FD) SPECT to be 5, 2 and 1 s per projection based on the list mode data for a total of 60 projections. FD and LD projections were Fourier transformed to magnitude and phase images, which were then separated into two or three frequency bands. Each frequency band was then inversed Fourier transformed back to the image domain. We proposed a 3D integrated attention-guided multi-frequency conditional generative adversarial network (AttMFGAN) and compared with AttGAN, and separate AttGAN for multi-frequency bands denoising (AttGAN-MF).The multi-frequency FD and LD projections of 35, 5 and 10 patients were paired for training, validation and testing. The LD projections to be tested were separated to multi-frequency components and input to corresponding networks to get the denoised components, which were summed to get the final denoised projections. Voxel-based error indices were measured on the cardiac region on the reconstructed images. The perfusion defect size (PDS) was also analyzed.ResultsAttGAN-MF and AttMFGAN have superior performance on all physical and clinical indices as compared to conventional AttGAN. The integrated AttMFGAN is better than AttGAN-MF. Multi-frequency denoising with two frequency bands have generally better results than corresponding three-frequency bands methods.ConclusionsAttGAN-MF and AttMFGAN are promising to further improve LD MP SPECT denoising.
| S-EPMC11447183 | biostudies-literature
Project description:PurposeTo develop a deep neural network-based computational workflow for inline myocardial perfusion analysis that automatically delineates the myocardium, which improves the clinical workflow and offers a "one-click" solution.Materials and methodsIn this retrospective study, consecutive adenosine stress and rest perfusion scans were acquired from three hospitals between October 1, 2018 and February 27, 2019. The training and validation set included 1825 perfusion series from 1034 patients (mean age, 60.6 years ± 14.2 [standard deviation]). The independent test set included 200 scans from 105 patients (mean age, 59.1 years ± 12.5). A convolutional neural network (CNN) model was trained to segment the left ventricular cavity, myocardium, and right ventricle by processing an incoming time series of perfusion images. Model outputs were compared with manual ground truth for accuracy of segmentation and flow measures derived on a global and per-sector basis with t test performed for statistical significance. The trained models were integrated onto MR scanners for effective inference.ResultsThe mean Dice ratio of automatic and manual segmentation was 0.93 ± 0.04. The CNN performed similarly to manual segmentation and flow measures for mean stress myocardial blood flow (MBF; 2.25 mL/min/g ± 0.59 vs 2.24 mL/min/g ± 0.59, P = .94) and mean rest MBF (1.08 mL/min/g ± 0.23 vs 1.07 mL/min/g ± 0.23, P = .83). The per-sector MBF values showed no difference between the CNN and manual assessment (P = .92). A central processing unit-based model inference on the MR scanner took less than 1 second for a typical perfusion scan of three slices.ConclusionThe described CNN was capable of cardiac perfusion mapping and integrated an automated inline implementation on the MR scanner, enabling one-click analysis and reporting in a manner comparable to manual assessment. Supplemental material is available for this article. © RSNA, 2020.
| S-EPMC7706884 | biostudies-literature
Project description:PurposeTo develop an end-to-end deep learning solution for quickly reconstructing radial simultaneous multi-slice (SMS) myocardial perfusion datasets with comparable quality to the pixel tracking spatiotemporal constrained reconstruction (PT-STCR) method.MethodsDynamic contrast enhanced (DCE) radial SMS myocardial perfusion data were obtained from 20 subjects who were scanned at rest and/or stress with or without ECG gating using a saturation recovery radial CAIPI turboFLASH sequence. Input to the networks consisted of complex coil combined images reconstructed using the inverse Fourier transform of undersampled radial SMS k-space data. Ground truth images were reconstructed using the PT-STCR pipeline. The performance of the residual booster 3D U-Net was tested by comparing it to state-of-the-art network architectures including MoDL, CRNN-MRI, and other U-Net variants.ResultsResults demonstrate significant improvements in speed requiring approximately 8 seconds to reconstruct one radial SMS dataset which is approximately 200 times faster than the PT-STCR method. Images reconstructed with the residual booster 3D U-Net retain quality of ground truth PT-STCR images (0.963 SSIM/40.238 PSNR/0.147 NRMSE). The residual booster 3D U-Net has superior performance compared to existing network architectures in terms of image quality, temporal dynamics, and reconstruction time.ConclusionResidual and booster learning combined with the 3D U-Net architecture was shown to be an effective network for reconstructing high-quality images from undersampled radial SMS datasets while bypassing the reconstruction time of the PT-STCR method.
| S-EPMC8493758 | biostudies-literature