Search

We developed a deep ensemble learning model with a radiomics spatial encoding execution for improved glioma segmentation accuracy using multi-parametric MRI (mp-MRI). This model was developed using 369 glioma patients with a 4-modality mp-MRI protocol: T1, contrast-enhanced T1 (T1-Ce), T2, and FLAIR. In each modality volume, a 3D sliding kernel was implemented across the brain to capture image heterogeneity: fifty-six radiomic features were extracted within the kernel, resulting in a 4th order tensor. Each radiomic feature can then be encoded as a 3D image volume, namely a radiomic feature map (RFM). PCA was employed for data dimension reduction and the first 4 PCs were selected. Four deep neural networks as sub-models following the U-Net architecture were trained for the segmenting of a region-of-interest (ROI): each sub-model utilizes the mp-MRI and 1 of the 4 PCs as a 5-channel input for a 2D execution. The 4 softmax probability results given by the U-net ensemble were superimposed and binarized by Otsu method as the segmentation result. Three ensemble models were trained to segment enhancing tumor (ET), tumor core (TC), and whole tumor (WT). The adopted radiomics spatial encoding execution enriches the image heterogeneity information that leads to the successful demonstration of the proposed deep ensemble model, which offers a new tool for mp-MRI based medical image segmentation.

The projection of planar MRI data onto a spherical surface is equivalent to a nonlinear image transformation that retains global anatomical information. By incorporating this image transformation process in our proposed spherical projection-based U-Net (SPU-Net) segmentation model design, multiple independent segmentation predictions can be obtained from a single MRI. The final segmentation is the average of all available results, and the variation can be visualized as a pixel-wise uncertainty map. An uncertainty score was introduced to evaluate and compare the performance of uncertainty measurements. The proposed SPU-Net model was implemented on the basis of 369 glioma patients with MP-MRI scans (T1, T1-Ce, T2, and FLAIR). Three SPU-Net models were trained to segment enhancing tumor (ET), tumor core (TC), and whole tumor (WT), respectively. The SPU-Net model was compared with (1) the classic U-Net model with test-time augmentation (TTA) and (2) linear scaling-based U-Net (LSU-Net) segmentation models in terms of both segmentation accuracy (Dice coefficient, sensitivity, specificity, and accuracy) and segmentation uncertainty (uncertainty map and uncertainty score). The developed SPU-Net model successfully achieved low uncertainty for correct segmentation predictions (e.g., tumor interior or healthy tissue interior) and high uncertainty for incorrect results (e.g., tumor boundaries). This model could allow the identification of missed tumor targets or segmentation errors in U-Net. Quantitatively, the SPU-Net model achieved the highest uncertainty scores for three segmentation targets (ET/TC/WT): 0.826/0.848/0.936, compared to 0.784/0.643/0.872 using the U-Net with TTA and 0.743/0.702/0.876 with the LSU-Net (scaling factor = 2). The SPU-Net also achieved statistically significantly higher Dice coefficients, underscoring the improved segmentation accuracy., Comment: 31 pages, 9 figures, 1 table

Purpose: To develop a novel deep-learning model that integrates radiomics analysis in a multi-dimensional feature fusion workflow for glioblastoma (GBM) post-resection survival prediction. Methods: A cohort of 235 GBM patients with complete surgical resection was divided into short-term/long-term survival groups with 1-yr survival time threshold. Each patient received a pre-surgery multi-parametric MRI exam, and three tumor subregions were segmented by neuroradiologists. The developed model comprises three data source branches: in the 1st radiomics branch, 456 radiomics features (RF) were from each patient; in the 2nd deep learning branch, an encoding neural network architecture was trained for survival group prediction using each single MR modality, and high-dimensional parameters of the last two network layers were extracted as deep features (DF). The extracted radiomics features and deep features were processed by a feature selection procedure to reduce dimension size of each feature space. In the 3rd branch, non-image-based patient-specific clinical features (PSCF) were collected. Finally, data sources from all three branches were fused as an integrated input for a supporting vector machine (SVM) execution for survival group prediction. Different strategies of model design, including 1) 2D/3D-based image analysis, and 2) different data source combinations in SVM input design, were investigated in comparison studies. Results: The model achieved 0.638 prediction accuracy when using PSCF only, which was higher than the results using RF or DF only in both 2D and 3D analysis. The joint use of RF/PSCF improved accuracy results to 0.681 in 3D analysis. The most accurate models in 2D/3D analysis reached the highest accuracy 0.745 with different combinations of RF/DF/ PSCF, and the corresponding ROC AUC results were 0.69(2D) and 0.71(3D), respectively.

Purpose: To develop a neural ordinary differential equation (ODE) model for visualizing deep neural network (DNN) behavior during multi-parametric MRI (mp-MRI) based glioma segmentation as a method to enhance deep learning explainability. Methods: By hypothesizing that deep feature extraction can be modeled as a spatiotemporally continuous process, we designed a novel deep learning model, neural ODE, in which deep feature extraction was governed by an ODE without explicit expression. The dynamics of 1) MR images after interactions with DNN and 2) segmentation formation can be visualized after solving ODE. An accumulative contribution curve (ACC) was designed to quantitatively evaluate the utilization of each MRI by DNN towards the final segmentation results. The proposed neural ODE model was demonstrated using 369 glioma patients with a 4-modality mp-MRI protocol: T1, contrast-enhanced T1 (T1-Ce), T2, and FLAIR. Three neural ODE models were trained to segment enhancing tumor (ET), tumor core (TC), and whole tumor (WT). The key MR modalities with significant utilization by DNN were identified based on ACC analysis. Segmentation results by DNN using only the key MR modalities were compared to the ones using all 4 MR modalities. Results: All neural ODE models successfully illustrated image dynamics as expected. ACC analysis identified T1-Ce as the only key modality in ET and TC segmentations, while both FLAIR and T2 were key modalities in WT segmentation. Compared to the U-Net results using all 4 MR modalities, Dice coefficient of ET (0.784->0.775), TC (0.760->0.758), and WT (0.841->0.837) using the key modalities only had minimal differences without significance. Conclusion: The neural ODE model offers a new tool for optimizing the deep learning model inputs with enhanced explainability. The presented methodology can be generalized to other medical image-related deep learning applications., Comment: 30 pages, 7 figures, 2 tables

To develop a deep-learning model that integrates radiomics analysis for enhanced performance of COVID-19 and Non-COVID-19 pneumonia detection using chest X-ray image, two deep-learning models were trained based on a pre-trained VGG-16 architecture: in the 1st model, X-ray image was the sole input; in the 2nd model, X-ray image and 2 radiomic feature maps (RFM) selected by the saliency map analysis of the 1st model were stacked as the input. Both models were developed using 812 chest X-ray images with 262/288/262 COVID-19/Non-COVID-19 pneumonia/healthy cases, and 649/163 cases were assigned as training-validation/independent test sets. In 1st model using X-ray as the sole input, the 1) sensitivity, 2) specificity, 3) accuracy, and 4) ROC Area-Under-the-Curve of COVID-19 vs Non-COVID-19 pneumonia detection were 1) 0.90$\pm$0.07 vs 0.78$\pm$0.09, 2) 0.94$\pm$0.04 vs 0.94$\pm$0.04, 3) 0.93$\pm$0.03 vs 0.89$\pm$0.03, and 4) 0.96$\pm$0.02 vs 0.92$\pm$0.04. In the 2nd model, two RFMs, Entropy and Short-Run-Emphasize, were selected with their highest cross-correlations with the saliency maps of the 1st model. The corresponding results demonstrated significant improvements (p<0.05) of COVID-19 vs Non-COVID-19 pneumonia detection: 1) 0.95$\pm$0.04 vs 0.85$\pm$0.04, 2) 0.97$\pm$0.02 vs 0.96$\pm$0.02, 3) 0.97$\pm$0.02 vs 0.93$\pm$0.02, and 4) 0.99$\pm$0.01 vs 0.97$\pm$0.02. The reduced variations suggested a superior robustness of 2nd model design.

Purpose: To develop a radiomics filtering technique for characterizing spatial-encoded regional pulmonary ventilation information on lung CT. Methods: The lung volume was segmented on 46 CT images, and a 3D sliding window kernel was implemented across the lung volume to capture the spatial-encoded image information. Fifty-three radiomic features were extracted within the kernel, resulting in a 4th order tensor object. As such, each voxel coordinate of the original lung was represented as a 53-dimensional feature vector, such that radiomic features could be viewed as feature maps within the lungs. To test the technique as a potential pulmonary ventilation biomarker, the radiomic feature maps were compared to paired functional images (Galligas PET or DTPA-SPECT) based on Spearman correlation (r) analysis. Results: The radiomic feature map GLRLM-based Run-Length Non-Uniformity and GLCOM-based Sum Average are found to be highly correlated with the functional imaging. The achieved r (median [range]) for the two features are 0.46 [0.05, 0.67] and 0.45 [0.21, 0.65] across 46 patients and 2 functional imaging modalities, respectively. Conclusions: The results provide evidence that local regions of sparsely encoded heterogeneous lung parenchyma on CT are associated with diminished radiotracer uptake and measured lung ventilation defects on PET/SPECT imaging. These findings demonstrate the potential of radiomics to serve as a complementary tool to the current lung quantification techniques and provide hypothesis-generating data for future studies., Comment: 21 pages, 3 figures, 3 tables

This paper develops a method of biologically guided deep learning for post-radiation FDG-PET image outcome prediction based on pre-radiation images and radiotherapy dose information. Based on the classic reaction-diffusion mechanism, a novel biological model was proposed using a partial differential equation that incorporates spatial radiation dose distribution as a patient-specific treatment information variable. A 7-layer encoder-decoder-based convolutional neural network (CNN) was designed and trained to learn the proposed biological model. As such, the model could generate post-radiation FDG-PET image outcome predictions with possible time-series transition from pre-radiotherapy image states to post-radiotherapy states. The proposed method was developed using 64 oropharyngeal patients with paired FDG-PET studies before and after 20Gy delivery (2Gy/daily fraction) by IMRT. In a two-branch deep learning execution, the proposed CNN learns specific terms in the biological model from paired FDG-PET images and spatial dose distribution as in one branch, and the biological model generates post-20Gy FDG-PET image prediction in the other branch. The proposed method successfully generated post-20Gy FDG-PET image outcome prediction with breakdown illustrations of biological model components. Time-series FDG-PET image predictions were generated to demonstrate the feasibility of disease response rendering. The developed biologically guided deep learning method achieved post-20Gy FDG-PET image outcome predictions in good agreement with ground-truth results. With break-down biological modeling components, the outcome image predictions could be used in adaptive radiotherapy decision-making to optimize personalized plans for the best outcome in the future., Comment: 26 pages, 5 figures

Purpose: To develop an Artificial Intelligence (AI) agent for fully-automated rapid head and neck (H&N) IMRT plan generation without time-consuming inverse planning.$$$$ Methods: This AI agent was trained using a conditional Generative Adversarial Network architecture. The generator, PyraNet, is a novel Deep Learning network that implements 28 classic ResNet blocks in pyramid-like concatenations. The discriminator is a customized 4-layer DenseNet. The AI agent first generates customized 2D projections at 9 template beam angles from 3D CT volume and structures of a patient. These projections are then stacked as 4D inputs of PyraNet, from which 9 radiation fluence maps are generated simultaneously. Finally, the predicted fluence maps are imported into a commercial treatment planning system (TPS) for plan integrity checks. The AI agent was built and tested upon 231 oropharyngeal plans from a TPS plan library. Only the primary plans in the sequential boost regime were studied. A customized Harr wavelet loss was adopted for fluence map comparison. Isodose distributions in test AI plans and TPS plans were qualitatively evaluated. Key dosimetric metrics were statistically compared.$$$$ Results: All test AI plans were successfully generated. Isodose gradients outside of PTV in AI plans were comparable with TPS plans. After PTV coverage normalization, $D_{mean}$ of parotids and oral cavity in AI plans and TPS plans were comparable without statistical significance. AI plans achieved comparable $D_{max}$ at 0.01cc of brainstem and cord+5mm without clinically relevant differences, but body $D_{max}$ was higher than the TPS plan results. The AI agent needs ~3s per case to predict fluence maps.$$$$ Conclusions: The developed AI agent can generate H&N IMRT plans with satisfying dosimetry quality. With rapid and fully automated implementation, it holds great potential for clinical applications., Comment: Withdraw for internal review

Pancreas stereotactic body radiotherapy treatment planning requires planners to make sequential, time consuming interactions with the treatment planning system (TPS) to reach the optimal dose distribution. We seek to develop a reinforcement learning (RL)-based planning bot to systematically address complex tradeoffs and achieve high plan quality consistently and efficiently. The focus of pancreas SBRT planning is finding a balance between organs-at-risk sparing and planning target volume (PTV) coverage. Planners evaluate dose distributions and make planning adjustments to optimize PTV coverage while adhering to OAR dose constraints. We have formulated such interactions between the planner and the TPS into a finite-horizon RL model. First, planning status features are evaluated based on human planner experience and defined as planning states. Second, planning actions are defined to represent steps that planners would commonly implement to address different planning needs. Finally, we have derived a reward system based on an objective function guided by physician-assigned constraints. The planning bot trained itself with 48 plans augmented from 16 previously treated patients and generated plans for 24 cases in a separate validation set. All 24 bot-generated plans achieve similar PTV coverages compared to clinical plans while satisfying all clinical planning constraints. Moreover, the knowledge learned by the bot can be visualized and interpreted as consistent with human planning knowledge, and the knowledge maps learned in separate training sessions are consistent, indicating reproducibility of the learning process., Comment: 13 pages, 5 figures, and a table. Submitted to International Journal of Radiation Oncology Biology Physics

This paper introduces a novel data clustering algorithm based on Langevin dynamics, where the associated potential is constructed directly from the data. To introduce a self-consistent potential, we adopt the potential model from the established Quantum Clustering method. The first step is to use a radial basis function to construct a density distribution from the data. A potential function is then constructed such that this density distribution is the ground state solution to the time-independent Schrodinger equation. The second step is to use this potential function with the Langevin dynamics at sub-critical temperature to avoid ergodicity. The Langevin equations take a classical Gibbs distribution as the invariant measure, where the peaks of the distribution coincide with minima of the potential surface. The time dynamics of individual data points lead to different metastable states, which are interpreted as cluster centers. Clustering is therefore achieved when subsets of the data aggregate - as a result of the Langevin dynamics for a moderate period of time - in the neighborhood of a particular potential minimum. While the data points are pushed towards potential minima by the potential gradient, Brownian motion allows them to effectively tunnel through local potential barriers and escape saddle points into locations of the potential surface otherwise forbidden. The algorithm's feasibility is first established based on several illustrating examples and theoretical analyses, followed by a stricter evaluation using a standard benchmark dataset.

Purpose: Often, the inverse deformation vector field (DVF) is needed together with the corresponding forward DVF in 4D reconstruction and dose calculation, adaptive radiation therapy, and simultaneous deformable registration. This study aims at improving both accuracy and efficiency of iterative algorithms for DVF inversion, and advancing our understanding of divergence and latency conditions. Method: We introduce a framework of fixed-point iteration algorithms with active feedback control for DVF inversion. Based on rigorous convergence analysis, we design control mechanisms for modulating the inverse consistency (IC) residual of the current iterate, to be used as feedback into the next iterate. The control is designed adaptively to the input DVF with the objective to enlarge the convergence area and expedite convergence. Three particular settings of feedback control are introduced: constant value over the domain throughout the iteration; alternating values between iteration steps; and spatially variant values. We also introduce three spectral measures of the displacement Jacobian for characterizing a DVF. These measures reveal the critical role of what we term the non-translational displacement component (NTDC) of the DVF. We carry out inversion experiments with an analytical DVF pair, and with DVFs associated with thoracic CT images of 6 patients at end of expiration and end of inspiration. Results: NTDC-adaptive iterations are shown to attain a larger convergence region at a faster pace compared to previous non-adaptive DVF inversion iteration algorithms. By our numerical experiments, alternating control yields smaller IC residuals and inversion errors than constant control. Spatially variant control renders smaller residuals and errors by at least an order of magnitude, compared to other schemes, in no more than 10 steps. Inversion results also show remarkable quantitative agreement with analysis-based predictions. Conclusion: Our analysis captures properties of DVF data associated with clinical CT images, and provides new understanding of iterative DVF inversion algorithms with a simple residual feedback control. Adaptive control is necessary and highly effective in the presence of non-small NTDCs. The adaptive iterations or the spectral measures, or both, may potentially be incorporated into deformable image registration methods.