RESUMO
BACKGROUND: The extended acquisition times required for MRI limit its availability in resource-constrained settings. Consequently, accelerating MRI by undersampling k-space data, which is necessary to reconstruct an image, has been a long-standing but important challenge. We aimed to develop a deep convolutional neural network (dCNN) optimisation method for MRI reconstruction and to reduce scan times and evaluate its effect on image quality and accuracy of oncological imaging biomarkers. METHODS: In this multicentre, retrospective, cohort study, MRI data from patients with glioblastoma treated at Heidelberg University Hospital (775 patients and 775 examinations) and from the phase 2 CORE trial (260 patients, 1083 examinations, and 58 institutions) and the phase 3 CENTRIC trial (505 patients, 3147 examinations, and 139 institutions) were used to develop, train, and test dCNN for reconstructing MRI from highly undersampled single-coil k-space data with various acceleration rates (R=2, 4, 6, 8, 10, and 15). Independent testing was performed with MRIs from the phase 2/3 EORTC-26101 trial (528 patients with glioblastoma, 1974 examinations, and 32 institutions). The similarity between undersampled dCNN-reconstructed and original MRIs was quantified with various image quality metrics, including structural similarity index measure (SSIM) and the accuracy of undersampled dCNN-reconstructed MRI on downstream radiological assessment of imaging biomarkers in oncology (automated artificial intelligence-based quantification of tumour burden and treatment response) was performed in the EORTC-26101 test dataset. The public NYU Langone Health fastMRI brain test dataset (558 patients and 558 examinations) was used to validate the generalisability and robustness of the dCNN for reconstructing MRIs from available multi-coil (parallel imaging) k-space data. FINDINGS: In the EORTC-26101 test dataset, the median SSIM of undersampled dCNN-reconstructed MRI ranged from 0·88 to 0·99 across different acceleration rates, with 0·92 (95% CI 0·92-0·93) for 10-times acceleration (R=10). The 10-times undersampled dCNN-reconstructed MRI yielded excellent agreement with original MRI when assessing volumes of contrast-enhancing tumour (median DICE for spatial agreement of 0·89 [95% CI 0·88 to 0·89]; median volume difference of 0·01 cm3 [95% CI 0·00 to 0·03] equalling 0·21%; p=0·0036 for equivalence) or non-enhancing tumour or oedema (median DICE of 0·94 [95% CI 0·94 to 0·95]; median volume difference of -0·79 cm3 [95% CI -0·87 to -0·72] equalling -1·77%; p=0·023 for equivalence) in the EORTC-26101 test dataset. Automated volumetric tumour response assessment in the EORTC-26101 test dataset yielded an identical median time to progression of 4·27 months (95% CI 4·14 to 4·57) when using 10-times-undersampled dCNN-reconstructed or original MRI (log-rank p=0·80) and agreement in the time to progression in 374 (95·2%) of 393 patients with data. The dCNN generalised well to the fastMRI brain dataset, with significant improvements in the median SSIM when using multi-coil compared with single-coil k-space data (p<0·0001). INTERPRETATION: Deep-learning-based reconstruction of undersampled MRI allows for a substantial reduction of scan times, with a 10-times acceleration demonstrating excellent image quality while preserving the accuracy of derived imaging biomarkers for the assessment of oncological treatment response. Our developments are available as open source software and hold considerable promise for increasing the accessibility to MRI, pending further prospective validation. FUNDING: Deutsche Forschungsgemeinschaft (German Research Foundation) and an Else Kröner Clinician Scientist Endowed Professorship by the Else Kröner Fresenius Foundation.
Assuntos
Aprendizado Profundo , Glioblastoma , Humanos , Inteligência Artificial , Biomarcadores , Estudos de Coortes , Glioblastoma/diagnóstico por imagem , Imageamento por Ressonância Magnética , Estudos RetrospectivosRESUMO
OBJECTIVES: Radiomic features have demonstrated encouraging results for non-invasive detection of molecular biomarkers, but the lack of guidelines for pre-processing MRI-data has led to poor generalizability. Here, we assessed the influence of different MRI-intensity normalization techniques on the performance of radiomics-based models for predicting molecular glioma subtypes. METHODS: Preoperative MRI-data from n = 615 patients with newly diagnosed glioma and known isocitrate dehydrogenase (IDH) and 1p/19q status were pre-processed using four different methods: no normalization (naive), N4 bias field correction (N4), N4 followed by either WhiteStripe (N4/WS), or z-score normalization (N4/z-score). A total of 377 Image-Biomarker-Standardisation-Initiative-compliant radiomic features were extracted from each normalized data, and 9 different machine-learning algorithms were trained for multiclass prediction of molecular glioma subtypes (IDH-mutant 1p/19q codeleted vs. IDH-mutant 1p/19q non-codeleted vs. IDH wild type). External testing was performed in public glioma datasets from UCSF (n = 410) and TCGA (n = 160). RESULTS: Support vector machine yielded the best performance with macro-average AUCs of 0.84 (naive), 0.84 (N4), 0.87 (N4/WS), and 0.87 (N4/z-score) in the internal test set. Both N4/WS and z-score outperformed the other approaches in the external UCSF and TCGA test sets with macro-average AUCs ranging from 0.85 to 0.87, replicating the performance of the internal test set, in contrast to macro-average AUCs ranging from 0.19 to 0.45 for naive and 0.26 to 0.52 for N4 alone. CONCLUSION: Intensity normalization of MRI data is essential for the generalizability of radiomic-based machine-learning models. Specifically, both N4/WS and N4/z-score approaches allow to preserve the high model performance, yielding generalizable performance when applying the developed radiomic-based machine-learning model in an external heterogeneous, multi-institutional setting. CLINICAL RELEVANCE STATEMENT: Intensity normalization such as N4/WS or N4/z-score can be used to develop reliable radiomics-based machine learning models from heterogeneous multicentre MRI datasets and provide non-invasive prediction of glioma subtypes. KEY POINTS: ⢠MRI-intensity normalization increases the stability of radiomics-based models and leads to better generalizability. ⢠Intensity normalization did not appear relevant when the developed model was applied to homogeneous data from the same institution. ⢠Radiomic-based machine learning algorithms are a promising approach for simultaneous classification of IDH and 1p/19q status of glioma.
Assuntos
Neoplasias Encefálicas , Glioma , Humanos , Neoplasias Encefálicas/diagnóstico por imagem , Neoplasias Encefálicas/genética , Radiômica , Glioma/diagnóstico por imagem , Glioma/genética , Imageamento por Ressonância Magnética/métodos , Biomarcadores , Isocitrato Desidrogenase/genética , Mutação , Estudos RetrospectivosRESUMO
BACKGROUND: A variety of deep learning-based and iterative approaches are available to predict Tracer Kinetic (TK) parameters from fully sampled or undersampled dynamic contrast-enhanced (DCE) MRI data. However, both the methods offer distinct benefits and drawbacks. PURPOSE: To propose a hybrid algorithm (named as 'Greybox'), using both model- as well as DL-based, for solving a multi-parametric non-linear inverse problem of directly estimating TK parameters from undersampled DCE MRI data, which is invariant to undersampling rate. METHODS: The proposed algorithm was inspired by plug-and-play algorithms used for solving linear inverse imaging problems. This technique was tested for its effectiveness in solving the nonlinear ill-posed inverse problem of generating 3D TK parameter maps from four-dimensional (4D; Spatial + Temporal) retrospectively undersampled k-space data. The algorithm learns a deep learning-based prior using UNET to estimate the K trans $\mathbf {K_{trans}}$ and V p $\mathbf {V_{p}}$ parameters based on the Patlak pharmacokinetic model, and this trained prior was utilized to estimate the TK parameter maps using an iterative gradient-based optimization scheme. Unlike the existing DL models, this network is invariant to the undersampling rate of the input data. The proposed method was compared with the total variation-based direct reconstruction technique on brain, breast, and prostate DCE-MRI datasets for various undersampling rates using the Radial Golden Angle (RGA) scheme. For the breast dataset, an indirect estimation using the Fast Composite Splitting algorithm was utilized for comparison. Undersampling rates of 8 × $\times$ , 12 × $\times$ and 20 × $\times$ were used for the experiments, and the results were compared using the PSNR and SSIM as metrics. For the breast dataset of 10 patients, data from four patients were utilized for training (1032 samples), two for validation (752 samples), and the entire volume of four patients for testing. Similarly, for the prostate dataset of 18 patients, 10 patients were utilized for training (720 samples), five for validation (216 samples), and the whole volume of three patients for testing. For the brain dataset of nineteen patients, ten patients were used for training (3152 samples), five for validation (1168 samples), and the whole volume of four patients for testing. Statistical tests were also conducted to assess the significance of the improvement in performance. RESULTS: The experiments showed that the proposed Greybox performs significantly better than other direct reconstruction methods. The proposed algorithm improved the estimated K trans $\mathbf {K_{trans}}$ and V p $\mathbf {V_{p}}$ in terms of the peak signal-to-noise ratio by up to 3 dB compared to other standard reconstruction methods. CONCLUSION: The proposed hybrid reconstruction algorithm, Greybox, can provide state-of-the-art performance in solving the nonlinear inverse problem of DCE-MRI. This is also the first of its kind to utilize convolutional neural network-based encodings as part of the plug-and-play priors to improve the performance of the reconstruction algorithm.
Assuntos
Algoritmos , Meios de Contraste , Processamento de Imagem Assistida por Computador , Imageamento por Ressonância Magnética , Imageamento por Ressonância Magnética/métodos , Processamento de Imagem Assistida por Computador/métodos , Humanos , Cinética , Masculino , Neoplasias da Próstata/diagnóstico por imagemRESUMO
Background: This study investigates the influence of diffusion-weighted Magnetic Resonance Imaging (DWI-MRI) on radiomic-based prediction of glioma types according to molecular status and assesses the impact of DWI intensity normalization on model generalizability. Methods: Radiomic features, compliant with image biomarker standardization initiative standards, were extracted from preoperative MRI of 549 patients with diffuse glioma, known IDH, and 1p19q-status. Anatomical sequences (T1, T1c, T2, FLAIR) underwent N4-Bias Field Correction (N4) and WhiteStripe normalization (N4/WS). Apparent diffusion coefficient (ADC) maps were normalized using N4 or N4/z-score. Nine machine-learning algorithms were trained for multiclass prediction of glioma types (IDH-mutant 1p/19q codeleted, IDH-mutant 1p/19q non-codeleted, IDH-wild type). Four approaches were compared: Anatomical, anatomicalâ +â ADC naive, anatomicalâ +â ADC N4, and anatomicalâ +â ADC N4/z-score. The University of California San Francisco (UCSF)-glioma dataset (nâ =â 409) was used for external validation. Results: Naïve-Bayes algorithms yielded overall the best performance on the internal test set. Adding ADC radiomics significantly improved AUC from 0.79 to 0.86 (Pâ =â .011) for the IDH-wild-type subgroup, but not for the other 2 glioma subgroups (Pâ >â .05). In the external UCSF dataset, the addition of ADC radiomics yielded a significantly higher AUC for the IDH-wild-type subgroup (Pâ ≤â .001): 0.80 (N4/WS anatomical alone), 0.81 (anatomicalâ +â ADC naive), 0.81 (anatomicalâ +â ADC N4), and 0.88 (anatomicalâ +â ADC N4/z-score) as well as for the IDH-mutant 1p/19q non-codeleted subgroup (Pâ <â .012 each). Conclusions: ADC radiomics can enhance the performance of conventional MRI-based radiomic models, particularly for IDH-wild-type glioma. The benefit of intensity normalization of ADC maps depends on the type and context of the used data.
RESUMO
Background: The purpose of this study was to elucidate the relationship between distinct brain regions and molecular subtypes in glioblastoma (GB), focusing on integrating modern statistical tools and molecular profiling to better understand the heterogeneity of Isocitrate Dehydrogenase wild-type (IDH-wt) gliomas. Methods: This retrospective study comprised 441 patients diagnosed with new IDH-wt glioma between 2009 and 2020 at Heidelberg University Hospital. The diagnostic process included preoperative magnetic resonance imaging and molecular characterization, encompassing IDH-status determination and subclassification, through DNA-methylation profiling. To discern and map distinct brain regions associated with specific methylation subtypes, a support-vector regression-based lesion-symptom mapping (SVR-LSM) was employed. Lesion maps were adjusted to 2 mm³ resolution. Significance was assessed with beta maps, using a threshold of Pâ <â .005, with 10 000 permutations and a cluster size minimum of 100 voxels. Results: Of 441 initially screened glioma patients, 423 (95.9%) met the inclusion criteria. Following DNA-methylation profiling, patients were classified into RTK II (40.7%), MES (33.8%), RTK I (18%), and other methylation subclasses (7.6%). Between molecular subtypes, there was no difference in tumor volume. Using SVR-LSM, distinct brain regions correlated with each subclass were identified: MES subtypes were associated with left-hemispheric regions involving the superior temporal gyrus and insula cortex, RTK I with right frontal regions, and RTK II with 3 clusters in the left hemisphere. Conclusions: This study linked molecular diversity and spatial features in glioblastomas using SVR-LSM. Future studies should validate these findings in larger, independent cohorts to confirm the observed patterns.
RESUMO
PURPOSE: To propose a robust time and space invariant deep learning (DL) method to directly estimate the pharmacokinetic/tracer kinetic (PK/TK) parameters from undersampled dynamic contrast-enhanced (DCE) magnetic resonance imaging (MRI) data. METHODS: DCE-MRI consists of 4D (3D-spatial + temporal) data and has been utilized to estimate 3D (spatial) tracer kinetic maps. Existing DL architecture for this task needs retraining for variation in temporal and/or spatial dimensions. This work proposes a DL algorithm that is invariant to training and testing in both temporal and spatial dimensions. The proposed network was based on a 2.5-dimensional Unet architecture, where the encoder consists of a 3D convolutional layer and the decoder consists of a 2D convolutional layer. The proposed VTDCE-Net was evaluated for solving the ill-posed inverse problem of directly estimating TK parameters from undersampled k - t $k-t$ space data of breast cancer patients, and the results were systematically compared with a total variation (TV) regularization based direct parameter estimation scheme. In the breast dataset, the training was performed on patients with 32 time samples, and testing was carried out on patients with 26 and 32 time samples. Translation of the proposed VTDCE-Net for brain dataset to show the generalizability was also carried out. Undersampling rates (R) of 8× , 12× , and 20× were utilized with PSNR and SSIM as the figures of merit in this evaluation. TK parameter maps estimated from fully sampled data were utilized as ground truth. RESULTS: Experiments carried out in this work demonstrate that the proposed VTDCE-Net outperforms the TV scheme on both breast and brain datasets across all undersampling rates. For K trans $\mathbf {K_{trans}}$ and V p $\mathbf {V_{p}}$ maps, the improvement over TV is as high as 2 and 5 dB, respectively, using the proposed VTDCE-Net. CONCLUSION: Temporal points invariant DL network that was proposed in this work to estimate the TK-parameters using DCE-MRI data has provided state-of-the-art performance compared to standard image reconstruction methods and is shown to work across all undersampling rates.
Assuntos
Neoplasias Encefálicas , Imageamento por Ressonância Magnética , Humanos , Imageamento por Ressonância Magnética/métodos , Redes Neurais de Computação , Encéfalo/diagnóstico por imagem , Processamento de Imagem Assistida por Computador/métodos , AlgoritmosRESUMO
Swift diagnosis and treatment play a decisive role in the clinical outcome of patients with acute ischemic stroke (AIS), and computer-aided diagnosis (CAD) systems can accelerate the underlying diagnostic processes. Here, we developed an artificial neural network (ANN) which allows automated detection of abnormal vessel findings without any a-priori restrictions and in <2 minutes. Pseudo-prospective external validation was performed in consecutive patients with suspected AIS from 4 different hospitals during a 6-month timeframe and demonstrated high sensitivity (≥87%) and negative predictive value (≥93%). Benchmarking against two CE- and FDA-approved software solutions showed significantly higher performance for our ANN with improvements of 25-45% for sensitivity and 4-11% for NPV (p ≤ 0.003 each). We provide an imaging platform ( https://stroke.neuroAI-HD.org ) for online processing of medical imaging data with the developed ANN, including provisions for data crowdsourcing, which will allow continuous refinements and serve as a blueprint to build robust and generalizable AI algorithms.
Assuntos
Aprendizado Profundo , AVC Isquêmico , Acidente Vascular Cerebral , Humanos , AVC Isquêmico/diagnóstico por imagem , Estudos Prospectivos , Angiografia por Tomografia Computadorizada/métodos , Acidente Vascular Cerebral/diagnóstico por imagem , Angiografia , Estudos RetrospectivosRESUMO
PURPOSE: To propose a generic deep learning based medical image reconstruction model (named as SpiNet) that can enforce any Schatten p-norm regularization with 0 < p ≤ 2, where the p can be learnt (or fixed) based on the problem at hand. METHODS: Model-based deep learning architecture for solving inverse problems consists of two parts, a deep learning based denoiser and an iterative data consistency solver. The former has either L2 norm or L1 norm enforced on it, which are convex and can be easily minimized. This work proposes a method to enforce any p norm on the noise prior where 0 < p ≤ 2. This is achieved by using Majorization-Minimization algorithm, which upper bounds the cost function with a convex function, thus can be easily minimized. The proposed SpiNet has the capability to work for a fixed p or it can learn p based on the data. The network was tested for solving the inverse problem of reconstructing magnetic resonance (MR) images from undersampled k space data and the results were compared with a popular model-based deep learning architecture MoDL which enforces L2 norm along with other compressive sensing-based algorithms. This comparison between MoDL and proposed SpiNet was performed for undersampling rates (R) of 2×, 4×, 6×, 8×, 12×, 16×, and 20×. Multiple figures of merit such as PSNR, SSIM, and NRMSE were utilized in this comparison. A two-tailed t test was performed for all undersampling rates and for all metrices for proving the superior performance of proposed SpiNet compared to MoDL. For training and testing, the same dataset that was utilized in MoDL implementation was deployed. RESULTS: The results indicate that for all undersampling rates, the proposed SpiNet shows higher PSNR and SSIM and lower NRMSE than MoDL. However, for low undersampling rates of 2× and 4×, there is no significant difference in performance of proposed SpiNet and MoDL in terms of PSNR and NRMSE. This can be expected as the learnt p value is close to 2 (norm enforced by MoDL). For higher undersampling rates ≥6×, SpiNet significantly outperforms MoDL in all metrices with improvement as high as 4 dB in PSNR and 0.5 points in SSIM. CONCLUSION: As deep learning based medical image reconstruction methods are gaining popularity, the proposed SpiNet provides a generic framework to incorporate Schatten p-norm regularization with 0
Assuntos
Compressão de Dados , Processamento de Imagem Assistida por Computador , Algoritmos , Imageamento por Ressonância Magnética , Redes Neurais de ComputaçãoRESUMO
PURPOSE: To compare the performance of iterative direct and indirect parametric reconstruction methods with indirect deep learning-based reconstruction methods in estimating tracer-kinetic parameters from highly undersampled DCE-MR Imaging breast data and provide a systematic comparison of the same. METHODS: Estimation of tracer-kinetic parameters using indirect methods from undersampled data requires to reconstruct the anatomical images initially by solving an inverse problem. This reconstructed images gets utilized in turn to estimate the tracer-kinetic parameters. In direct estimation, the parameters are estimated without reconstructing the anatomical images. Both problems are ill-posed and are typically solved using prior-based regularization or using deep learning. In this study, for indirect estimation, two deep learning-based reconstruction frameworks namely, ISTA-Net+ and MODL, were utilized. For direct and indirect parametric estimation, sparsity inducing priors (L1 and Total-Variation) and limited memory Broyden-Fletcher-Goldfarb-Shanno algorithm as solver was deployed. The performance of these techniques were compared systematically in estimation of vascular permeability ( K trans ) from undersampled DCE-MRI breast data using Patlak as pharmaco-kinetic model. The experiments involved retrospective undersampling of the data 20×, 50×, and 100× and compared the results using PSNR, nRMSE, SSIM, and Xydeas metrics. The K trans maps estimated from fully sampled data were utilized as ground truth. The developed code was made available as https://github.com/Medical-Imaging-Group/DCE-MRI-Compare open-source for enthusiastic users. RESULTS: The reconstruction methods performance was evaluated using ten patients breast data (five patients each for training and testing). Consistent with other studies, the results indicate that direct parametric reconstruction methods provide improved performance compared to the indirect parameteric reconstruction methods. The results also indicate that for 20× undersampling, deep learning-based methods performs better or at par with direct estimation in terms of PSNR, SSIM, and nRMSE. However, for higher undersampling rates (50× and 100×) direct estimation performs better in all metrics. For all undersampling rates, direct reconstruction performed better in terms of Xydeas metric, which indicated fidelity in magnitude and orientation of edges. CONCLUSION: Deep learning-based indirect techniques perform at par with direct estimation techniques for lower undersampling rates in the breast DCE-MR imaging. At higher undersampling rates, they are not able to provide much needed generalization. Direct estimation techniques are able to provide more accurate results than both deep learning- and parametric-based indirect methods in these high undersampling scenarios.
Assuntos
Aprendizado Profundo , Algoritmos , Humanos , Processamento de Imagem Assistida por Computador , Cinética , Imageamento por Ressonância Magnética , Estudos RetrospectivosRESUMO
This study examines the feasibility and application of Advanced Oxidation Technologies (AOTs) for the treatment of chlorophenols that are included in US EPA priority pollutant list. A novel class of sulfate/hydroxyl radical-based homogeneous AOTs (Fe(II)/PS, Fe(II)/PMS, Fe(II)/H2O2) was successfully tested for the degradation of series of chlorophenols (4-CP, 2,4-CP, 2,4,6-CP, 2,3,4,5-CP). The major objective of the present study was to evaluate the effectiveness of three representative chelating agents (citrate, ethylenediaminedisuccinate (EDDS), and pyrophosphate) on Fe(II)-mediated activation of three common peroxide (peroxymonosulfate (PMS), persulfate (PS), and hydrogen peroxide (H2O2)) at neutral pH conditions. Short term (4 h) and long term (7 days) experiments were conducted to evaluate the kinetics and longevity of different oxidative systems for 4-chlorophenol degradation. Results showed that each of the iron-chelating agent couple was superior in activating a particular oxidant and consequently for 4-CP degradation. In case of Fe(II)/PMS system, the inorganic chelating agent pyrophosphate showed effective activation of PMS whereas very fast dissociation of PMS was recorded in the case of EDDS without any apparent 4-CP degradation. In Fe(II)/H2O2 system, EDDS was proven to be the most effective whereas pyrophosphate showed negligible activation of H2O2. Fe(II)/Citrate system showed moderate activation of all three oxidants. PMS was found to be the most universal oxidant, which was activated by all three iron-chelating agent systems and Fe(II)/Citrate was the most universal chelating agent system, which was able to activate all three oxidants to a certain extent.