Purpose
To evaluate diagnostic accuracy of fully automated analysis of multimodal imaging data using
18
F-FET-PET and MRI (including amide proton transfer-weighted (APTw) imaging and ...dynamic-susceptibility-contrast (DSC) perfusion) in differentiation of tumor progression from treatment-related changes in patients with glioma.
Material and methods
At suspected tumor progression, MRI and
18
F-FET-PET data as part of a retrospective analysis of an observational cohort of 66 patients/74 scans (51 glioblastoma and 23 lower-grade-glioma, 8 patients included at two different time points) were automatically segmented into necrosis, FLAIR-hyperintense, and contrast-enhancing areas using an ensemble of deep learning algorithms. In parallel, previous MR exam was processed in a similar way to subtract preexisting tumor areas and focus on progressive tumor only. Within these progressive areas, intensity statistics were automatically extracted from
18
F-FET-PET, APTw, and DSC-derived cerebral-blood-volume (CBV) maps and used to train a Random Forest classifier with threefold cross-validation. To evaluate contribution of the imaging modalities to the classifier’s performance, impurity-based importance measures were collected. Classifier performance was compared with radiology reports and interdisciplinary tumor board assessments.
Results
In 57/74 cases (77%), tumor progression was confirmed histopathologically (39 cases) or via follow-up imaging (18 cases), while remaining 17 cases were diagnosed as treatment-related changes. The classification accuracy of the Random Forest classifier was 0.86, 95% CI 0.77–0.93 (sensitivity 0.91, 95% CI 0.81–0.97; specificity 0.71, 95% CI 0.44–0.9), significantly above the no-information rate of 0.77 (p = 0.03), and higher compared to an accuracy of 0.82 for MRI (95% CI 0.72–0.9), 0.81 for
18
F-FET-PET (95% CI 0.7–0.89), and 0.81 for expert consensus (95% CI 0.7–0.89), although these differences were not statistically significant (p > 0.1 for all comparisons, McNemar test).
18
F-FET-PET hot-spot volume was single-most important variable, with relevant contribution from all imaging modalities.
Conclusion
Automated, joint image analysis of
18
F-FET-PET and advanced MR imaging techniques APTw and DSC perfusion is a promising tool for objective response assessment in gliomas.
Intratumoral heterogeneity at the cellular and molecular level is a hallmark of glioblastoma (GB) that contributes to treatment resistance and poor clinical outcome. Little is known regarding ...epigenetic heterogeneity and intratumoral phylogeny and their implication for molecular classification and targeted therapies.
Multiple tissue biopsies (238 in total) were sampled from 56 newly-diagnosed, treatment-naive GB patients from a prospective in-house cohort and publicly available data and profiled for DNA methylation using the Illumina MethylationEPIC array. Methylation-based classification using the glioma classifier developed by Ceccarelli et al. and estimation of the MGMT promoter methylation status via the MGMT-STP27 model were carried out. In addition, copy number variations (CNVs) and phylogeny were analyzed.
Almost half of the patients (22/56, 39%) harbored tumors composed of heterogeneous methylation subtypes. We found two predominant subtype combinations: classic-/mesenchymal-like, and mesenchymal-/pilocytic astrocytoma-like. Nine patients (16%) had tumors composed of subvolumes with and without MGMT promoter methylation, whereas 20 patients (36%) were homogeneously methylated, and 27 patients (48%) were homogeneously unmethylated. CNV analysis revealed high variations in many genes, including CDKN2A/B, EGFR, and PTEN. Phylogenetic analysis correspondingly showed a general pattern of CDKN2A/B loss and gain of EGFR, PDGFRA, and CDK4 during early stages of tumor development.
(Epi)genetic intratumoral heterogeneity is a hallmark of GB, both at DNA methylation and CNV level. This intratumoral heterogeneity is of utmost importance for molecular classification as well as for defining therapeutic targets in this disease, as single biopsies might underestimate the true molecular diversity in a tumor.
•Glioblastoma exhibits significant heterogeneity, from epigenome-wide methylation phenotypes to single molecular targets.•Phylogeny showed CDKN2A/B loss and gain of EGFR, PDGFRA, and CDK4 early in tumor development.•Intratumoral heterogeneity is of utmost importance for molecular classification as well as for defining therapeutic targets.•Assessing single biopsies underestimates the true molecular diversity in a tumor.
Reaction-diffusion based tumor growth models have been widely used in the literature for modeling the growth of brain gliomas. Lately, recent models have started integrating medical images in their ...formulation. Including different tissue types, geometry of the brain and the directions of white matter fiber tracts improved the spatial accuracy of reaction-diffusion models. The adaptation of the general model to the specific patient cases on the other hand has not been studied thoroughly yet. In this paper, we address this adaptation. We propose a parameter estimation method for reaction-diffusion tumor growth models using time series of medical images. This method estimates the patient specific parameters of the model using the images of the patient taken at successive time instances. The proposed method formulates the evolution of the tumor delineation visible in the images based on the reaction-diffusion dynamics; therefore, it remains consistent with the information available. We perform thorough analysis of the method using synthetic tumors and show important couplings between parameters of the reaction-diffusion model. We show that several parameters can be uniquely identified in the case of fixing one parameter, namely the proliferation rate of tumor cells. Moreover, regardless of the value the proliferation rate is fixed to, the speed of growth of the tumor can be estimated in terms of the model parameters with accuracy. We also show that using the model-based speed, we can simulate the evolution of the tumor for the specific patient case. Finally, we apply our method to two real cases and show promising preliminary results.
Dynamic contrast-enhanced magnetic resonance (DCE-MR) imaging can be used to study microvascular structure in vivo by monitoring the abundance of an injected diffusible contrast agent over time. The ...resulting spatially resolved intensity-time curves are usually interpreted in terms of kinetic parameters obtained by fitting a pharmacokinetic model to the observed data. Least squares estimates of the highly nonlinear model parameters, however, can exhibit high variance and can be severely biased. As a remedy, we bring to bear spatial prior knowledge by means of a generalized Gaussian Markov random field (GGMRF). By using information from neighboring voxels and computing the maximum a posteriori solution for entire parameter maps at once, both bias and variance of the parameter estimates can be reduced thus leading to smaller root mean square error (RMSE). Since the number of variables gets very big for common image resolutions, sparse solvers have to be employed. To this end, we propose a generalized iterated conditional modes (ICM) algorithm operating on blocks instead of sites which is shown to converge considerably faster than the conventional ICM algorithm. Results on simulated DCE-MR images show a clear reduction of RMSE and variance as well as, in some cases, reduced estimation bias. The mean residual bias (MRB) is reduced on the simulated data as well as for all 37 patients of a prostate DCE-MRI dataset. Using the proposed algorithm, average computation times only increase by a factor of 1.18 (871 ms per voxel) for a Gaussian prior and 1.51 (1.12 s per voxel) for an edge-preserving prior compared to the single voxel approach (740 ms per voxel).
Regularized regression methods such as principal component or partial least squares regression perform well in learning tasks on high dimensional spectral data, but cannot explicitly eliminate ...irrelevant features. The random forest classifier with its associated Gini feature importance, on the other hand, allows for an explicit feature elimination, but may not be optimally adapted to spectral data due to the topology of its constituent classification trees which are based on orthogonal splits in feature space.
We propose to combine the best of both approaches, and evaluated the joint use of a feature selection based on a recursive feature elimination using the Gini importance of random forests' together with regularized classification methods on spectral data sets from medical diagnostics, chemotaxonomy, biomedical analytics, food science, and synthetically modified spectral data. Here, a feature selection using the Gini feature importance with a regularized classification by discriminant partial least squares regression performed as well as or better than a filtering according to different univariate statistical tests, or using regression coefficients in a backward feature elimination. It outperformed the direct application of the random forest classifier, or the direct application of the regularized classifiers on the full set of features.
The Gini importance of the random forest provided superior means for measuring feature relevance on spectral data, but - on an optimal subset of features - the regularized classifiers might be preferable over the random forest classifier, in spite of their limitation to model linear dependencies only. A feature selection based on Gini importance, however, may precede a regularized linear classification to identify this optimal subset of features, and to earn a double benefit of both dimensionality reduction and the elimination of noise from the classification task.
Variations in the shape and appearance of anatomical structures in medical images are often relevant radiological signs of disease. Automatic tools can help automate parts of this manual process. A ...cloud-based evaluation framework is presented in this paper including results of benchmarking current state-of-the-art medical imaging algorithms for anatomical structure segmentation and landmark detection: the VISCERAL Anatomy benchmarks. The algorithms are implemented in virtual machines in the cloud where participants can only access the training data and can be run privately by the benchmark administrators to objectively compare their performance in an unseen common test set. Overall, 120 computed tomography and magnetic resonance patient volumes were manually annotated to create a standard Gold Corpus containing a total of 1295 structures and 1760 landmarks. Ten participants contributed with automatic algorithms for the organ segmentation task, and three for the landmark localization task. Different algorithms obtained the best scores in the four available imaging modalities and for subsets of anatomical structures. The annotation framework, resulting data set, evaluation setup, results and performance analysis from the three VISCERAL Anatomy benchmarks are presented in this article. Both the VISCERAL data set and Silver Corpus generated with the fusion of the participant algorithms on a larger set of non-manually-annotated medical images are available to the research community.
•Evaluation framework for automatic stroke lesion segmentation from MRI•Public multi-center, multi-vendor, multi-protocol databases released•Ongoing fair and automated benchmark with expert created ...ground truth sets•Comparison of 14+7 groups who responded to an open challenge in MICCAI•Segmentation feasible in acute and unsolved in sub-acute cases
Display omitted
Ischemic stroke is the most common cerebrovascular disease, and its diagnosis, treatment, and study relies on non-invasive imaging. Algorithms for stroke lesion segmentation from magnetic resonance imaging (MRI) volumes are intensely researched, but the reported results are largely incomparable due to different datasets and evaluation schemes. We approached this urgent problem of comparability with the Ischemic Stroke Lesion Segmentation (ISLES) challenge organized in conjunction with the MICCAI 2015 conference. In this paper we propose a common evaluation framework, describe the publicly available datasets, and present the results of the two sub-challenges: Sub-Acute Stroke Lesion Segmentation (SISS) and Stroke Perfusion Estimation (SPES). A total of 16 research groups participated with a wide range of state-of-the-art automatic segmentation algorithms. A thorough analysis of the obtained data enables a critical evaluation of the current state-of-the-art, recommendations for further developments, and the identification of remaining challenges. The segmentation of acute perfusion lesions addressed in SPES was found to be feasible. However, algorithms applied to sub-acute lesion segmentation in SISS still lack accuracy. Overall, no algorithmic characteristic of any method was found to perform superior to the others. Instead, the characteristics of stroke lesion appearances, their evolution, and the observed challenges should be studied in detail. The annotated ISLES image datasets continue to be publicly available through an online evaluation system to serve as an ongoing benchmarking resource (www.isles-challenge.org).
In patients with soft-tissue sarcomas, tumor grading constitutes a decisive factor to determine the best treatment decision. Tumor grading is obtained by pathological work-up after focal biopsies. ...Deep learning (DL)-based imaging analysis may pose an alternative way to characterize STS tissue. In this work, we sought to non-invasively differentiate tumor grading into low-grade (G1) and high-grade (G2/G3) STS using DL techniques based on MR-imaging.
Contrast-enhanced T1-weighted fat-saturated (T1FSGd) MRI sequences and fat-saturated T2-weighted (T2FS) sequences were collected from two independent retrospective cohorts (training: 148 patients, testing: 158 patients). Tumor grading was determined following the French Federation of Cancer Centers Sarcoma Group in pre-therapeutic biopsies. DL models were developed using transfer learning based on the DenseNet 161 architecture.
The T1FSGd and T2FS-based DL models achieved area under the receiver operator characteristic curve (AUC) values of 0.75 and 0.76 on the test cohort, respectively. T1FSGd achieved the best F1-score of all models (0.90). The T2FS-based DL model was able to significantly risk-stratify for overall survival. Attention maps revealed relevant features within the tumor volume and in border regions.
MRI-based DL models are capable of predicting tumor grading with good reproducibility in external validation.