Optimized flip angle schemes for the split acquisition of fast spin-echo signals (SPLICE) sequence and application to diffusion-weighted imaging.

PURPOSE: The diffusion-weighted SPLICE (split acquisition of fast spin-echo signals) sequence employs split-echo rapid acquisition with relaxation enhancement (RARE) readout to provide images almost free of geometric distortions. However, due to the varying T 2 $$ {}_2 $$ -weighting during k-space traversal, SPLICE suffers from blurring. This work extends a method for controlling the spatial point spread function (PSF) while optimizing the signal-to-noise ratio (SNR) achieved by adjusting the flip angles in the refocusing pulse train of SPLICE. METHODS: An algorithm based on extended phase graph (EPG) simulations optimizes the flip angles by maximizing SNR for a flexibly chosen predefined target PSF that describes the desired k-space density weighting and spatial resolution. An optimized flip angle scheme and a corresponding post-processing correction filter which together achieve the target PSF was tested by healthy subject brain imaging using a clinical 1.5 T scanner. RESULTS: Brain images showed a clear and consistent improvement over those obtained with a standard constant flip angle scheme. SNR was increased and apparent diffusion coefficient estimates were more accurate. For a modified Hann k-space weighting example, considerable benefits resulted from acquisition weighting by flip angle control. CONCLUSION: The presented flexible method for optimizing SPLICE flip angle schemes offers improved MR image quality of geometrically accurate diffusion-weighted images that makes the sequence a strong candidate for radiotherapy planning or stereotactic surgery.

Decomposition-based framework for tumor classification and prediction of treatment response from longitudinal MRI.

Objective.In the field of radiation oncology, the benefit of MRI goes beyond that of providing high soft-tissue contrast images for staging and treatment planning. With the recent clinical introduction of hybrid MRI linear accelerators it has become feasible to map physiological parameters describing diffusion, perfusion, and relaxation during the entire course of radiotherapy, for example. However, advanced data analysis tools are required for extracting qualified prognostic and predictive imaging biomarkers from longitudinal MRI data. In this study, we propose a new prediction framework tailored to exploit temporal dynamics of tissue features from repeated measurements. We demonstrate the framework using a newly developed decomposition method for tumor characterization.Approach.Two previously published MRI datasets with multiple measurements during and after radiotherapy, were used for development and testing:T2-weighted multi-echo images obtained for two mouse models of pancreatic cancer, and diffusion-weighted images for patients with brain metastases. Initially, the data was decomposed using the novel monotonous slope non-negative matrix factorization (msNMF) tailored for MR data. The following processing consisted of a tumor heterogeneity assessment using descriptive statistical measures, robust linear modelling to capture temporal changes of these, and finally logistic regression analysis for stratification of tumors and volumetric outcome.Main Results.The framework was able to classify the two pancreatic tumor types with an area under curve (AUC) of 0.999,P< 0.001 and predict the tumor volume change with a correlation coefficient of 0.513,P= 0.034. A classification of the human brain metastases into responders and non-responders resulted in an AUC of 0.74,P= 0.065.Significance.A general data processing framework for analyses of longitudinal MRI data has been developed and applications were demonstrated by classification of tumor type and prediction of radiotherapy response. Further, as part of the assessment, the merits of msNMF for tumor tissue decomposition were demonstrated.

Data-driven separation of MRI signal components for tissue characterization.

PURPOSE: MRI can be utilized for quantitative characterization of tissue. To assess e.g. water fractions or diffusion coefficients for compartments in the brain, a decomposition of the signal is necessary. Imposing standard models carries the risk of estimating biased parameters if model assumptions are violated. This work introduces a data-driven multicomponent analysis, the monotonous slope non-negative matrix factorization (msNMF), tailored to extract data features expected in MR signals. METHODS: The msNMF was implemented by extending the standard NMF with monotonicity constraints on the signal profiles and their first derivatives. The method was validated using simulated data, and subsequently applied to both ex vivo DWI data and in vivo relaxometry data. Reproducibility of the method was tested using the latter. RESULTS: The msNMF recovered the multi-exponential signals in the simulated data and showed superiority to standard NMF (based on the explained variance, area under the ROC curve, and coefficient of variation). Diffusion components extracted from the DWI data reflected the cell density of the underlying tissue. The relaxometry analysis resulted in estimates of edema water fractions (EWF) highly correlated with published results, and demonstrated acceptable reproducibility. CONCLUSION: The msNMF can robustly separate MR signals into components with relation to the underlying tissue composition, and may potentially be useful for e.g. tumor tissue characterization.

Combined hyperpolarized 13C-pyruvate MRS and 18F-FDG PET (hyperPET) estimates of glycolysis in canine cancer patients.

13C Magnetic Resonance Spectroscopy (MRS) using hyperpolarized 13C-labeled pyruvate as a substrate offers a measure of pyruvate-lactate interconversion and is thereby a marker of the elevated aerobic glycolysis (Warburg effect) generally exhibited by cancer cells. Here, we aim to compare hyperpolarized [1-13C]pyruvate MRS with simultaneous 18F-2-fluoro-2-deoxy-d-glucose (FDG) PET in a cross-sectional study of canine cancer patients. METHODS: Canine cancer patients underwent integrated PET/MRI using a clinical whole-body system. Hyperpolarized [1-13C]pyruvate was obtained using dissolution-DNP. 18F-FDG PET, dynamic 13C MRS, 13C MRS Imaging (MRSI) and anatomical 1H MRI was acquired from 17 patients. Apparent pyruvate-to-lactate rate constants were estimated from dynamic 13C MRS. 18F-FDG Standard Uptake Values and maximum [1-13C]lactate-to-total-13C ratios were obtained from tumor regions of interest. Following inspection of data, patients were grouped according to main cancer type and linear regression between measures of lactate generation and 18F-FDG uptake were tested within groups. Between groups, the same measures were tested for group differences. RESULTS: The main cancer types of the 17 patients were sarcoma (n = 11), carcinoma (n = 5) and mastocytoma (n = 1). Significant correlations between pyruvate-to-lactate rate constants and 18F-FDG uptake were found for sarcoma patients, whereas no significant correlations appeared for carcinoma patients. The sarcoma patients showed a non-significant trend towards lower 18F-FDG uptake and higher lactate generation than carcinoma patients. However, the ratio of lactate generation to 18F-FDG uptake was found to be significantly higher in sarcoma as compared to carcinoma. The results were found both when lactate generation was estimated as an apparent pyruvate-to-lactate rate constant from dynamic 13C MRS and as an [1-13C]lactate to total 13C ratio from 13C MRSI. CONCLUSIONS: A comparison of hyperpolarized [1-13C]pyruvate MRS with simultaneous 18F-FDG PET indicate that lactate generation and 18F-FDG uptake in cancers can be related and that their relation depend on cancer type. This finding could be important for the interpretation and eventual clinical implementation of hyperpolarized 13C. In addition, the differences between the two modalities may allow for better metabolic phenotyping performing hybrid imaging in the form of hyperPET.

In Vivo Phenotyping of Tumor Metabolism in a Canine Cancer Patient with Simultaneous (18)F-FDG-PET and Hyperpolarized (13)C-Pyruvate Magnetic Resonance Spectroscopic Imaging (hyperPET): Mismatch Demonstrates that FDG may not Always Reflect the Warburg Effect.

In this communication the mismatch between simultaneous (18)F-FDG-PET and a (13)C-lactate imaging (hyperPET) in a biopsy verified squamous cell carcinoma in the right tonsil of a canine cancer patient is shown. The results demonstrate that (18)F-FDG-PET may not always reflect the Warburg effect in all tumors.

Simultaneous Hyperpolarized 13C-Pyruvate MRI and 18F-FDG PET (HyperPET) in 10 Dogs with Cancer.

UNLABELLED: With the introduction of combined PET/MR spectroscopic (MRS) imaging, it is now possible to directly and indirectly image the Warburg effect with hyperpolarized (13)C-pyruvate and (18)F-FDG PET imaging, respectively, via a technique we have named hyperPET. The main purpose of this present study was to establish a practical workflow for performing (18)F-FDG PET and hyperpolarized (13)C-pyruvate MRS imaging simultaneously for tumor tissue characterization and on a larger scale test its feasibility. In addition, we evaluated the correlation between (18)F-FDG uptake and (13)C-lactate production. METHODS: Ten dogs with biopsy-verified spontaneous malignant tumors were included for imaging. All dogs underwent a protocol of simultaneous (18)F-FDG PET, anatomic MR, and hyperpolarized dynamic nuclear polarization with (13)C-pyruvate imaging. The data were acquired using a combined clinical PET/MR imaging scanner. RESULTS: We found that combined (18)F-FDG PET and (13)C-pyruvate MRS imaging was possible in a single session of approximately 2 h. A continuous workflow was obtained with the injection of (18)F-FDG when the dogs was placed in the PET/MR scanner. (13)C-MRS dynamic acquisition demonstrated in an axial slab increased (13)C-lactate production in 9 of 10 dogs. For the 9 dogs, the (13)C-lactate was detected after a mean of 25 s (range, 17-33 s), with a mean to peak of (13)C-lactate at 49 s (range, 40-62 s). (13)C-pyruvate could be detected on average after 13 s (range, 5-26 s) and peaked on average after 25 s (range, 13-42 s). We noticed concordance of (18)F-FDG uptake and production of (13)C-lactate in most, but not all, axial slices. CONCLUSION: In this study, we have shown in a series of dogs with cancer that hyperPET can easily be performed within 2 h. We showed mostly correspondence between (13)C-lactate production and (18)F-FDG uptake and expect the combined modalities to reveal additional metabolic information to improve prognostic value and improve response monitoring.