Multiparametric MRI in Era of Artificial Intelligence for Bladder Cancer Therapies.

This review focuses on the principles, applications, and performance of mpMRI for bladder imaging. Quantitative imaging biomarkers (QIBs) derived from mpMRI are increasingly used in oncological applications, including tumor staging, prognosis, and assessment of treatment response. To standardize mpMRI acquisition and interpretation, an expert panel developed the Vesical Imaging-Reporting and Data System (VI-RADS). Many studies confirm the standardization and high degree of inter-reader agreement to discriminate muscle invasiveness in bladder cancer, supporting VI-RADS implementation in routine clinical practice. The standard MRI sequences for VI-RADS scoring are anatomical imaging, including T2w images, and physiological imaging with diffusion-weighted MRI (DW-MRI) and dynamic contrast-enhanced MRI (DCE-MRI). Physiological QIBs derived from analysis of DW- and DCE-MRI data and radiomic image features extracted from mpMRI images play an important role in bladder cancer. The current development of AI tools for analyzing mpMRI data and their potential impact on bladder imaging are surveyed. AI architectures are often implemented based on convolutional neural networks (CNNs), focusing on narrow/specific tasks. The application of AI can substantially impact bladder imaging clinical workflows; for example, manual tumor segmentation, which demands high time commitment and has inter-reader variability, can be replaced by an autosegmentation tool. The use of mpMRI and AI is projected to drive the field toward the personalized management of bladder cancer patients.

A Quantitative Multiparametric MRI Analysis Platform for Estimation of Robust Imaging Biomarkers in Clinical Oncology.

There is a need to develop user-friendly imaging tools estimating robust quantitative biomarkers (QIBs) from multiparametric (mp)MRI for clinical applications in oncology. Quantitative metrics derived from (mp)MRI can monitor and predict early responses to treatment, often prior to anatomical changes. We have developed a vendor-agnostic, flexible, and user-friendly MATLAB-based toolkit, MRI-Quantitative Analysis and Multiparametric Evaluation Routines ("MRI-QAMPER", current release v3.0), for the estimation of quantitative metrics from dynamic contrast-enhanced (DCE) and multi-b value diffusion-weighted (DW) MR and MR relaxometry. MRI-QAMPER's functionality includes generating numerical parametric maps from these methods reflecting tumor permeability, cellularity, and tissue morphology. MRI-QAMPER routines were validated using digital reference objects (DROs) for DCE and DW MRI, serving as initial approval stages in the National Cancer Institute Quantitative Imaging Network (NCI/QIN) software benchmark. MRI-QAMPER has participated in DCE and DW MRI Collaborative Challenge Projects (CCPs), which are key technical stages in the NCI/QIN benchmark. In a DCE CCP, QAMPER presented the best repeatability coefficient (RC = 0.56) across test-retest brain metastasis data, out of ten participating DCE software packages. In a DW CCP, QAMPER ranked among the top five (out of fourteen) tools with the highest area under the curve (AUC) for prostate cancer detection. This platform can seamlessly process mpMRI data from brain, head and neck, thyroid, prostate, pancreas, and bladder cancer. MRI-QAMPER prospectively analyzes dose de-escalation trial data for oropharyngeal cancer, which has earned it advanced NCI/QIN approval for expanded usage and applications in wider clinical trials.

Artificial Intelligence in CT and MR Imaging for Oncological Applications.

Cancer care increasingly relies on imaging for patient management. The two most common cross-sectional imaging modalities in oncology are computed tomography (CT) and magnetic resonance imaging (MRI), which provide high-resolution anatomic and physiological imaging. Herewith is a summary of recent applications of rapidly advancing artificial intelligence (AI) in CT and MRI oncological imaging that addresses the benefits and challenges of the resultant opportunities with examples. Major challenges remain, such as how best to integrate AI developments into clinical radiology practice, the vigorous assessment of quantitative CT and MR imaging data accuracy, and reliability for clinical utility and research integrity in oncology. Such challenges necessitate an evaluation of the robustness of imaging biomarkers to be included in AI developments, a culture of data sharing, and the cooperation of knowledgeable academics with vendor scientists and companies operating in radiology and oncology fields. Herein, we will illustrate a few challenges and solutions of these efforts using novel methods for synthesizing different contrast modality images, auto-segmentation, and image reconstruction with examples from lung CT as well as abdome, pelvis, and head and neck MRI. The imaging community must embrace the need for quantitative CT and MRI metrics beyond lesion size measurement. AI methods for the extraction and longitudinal tracking of imaging metrics from registered lesions and understanding the tumor environment will be invaluable for interpreting disease status and treatment efficacy. This is an exciting time to work together to move the imaging field forward with narrow AI-specific tasks. New AI developments using CT and MRI datasets will be used to improve the personalized management of cancer patients.

Quantitative Relaxometry Metrics for Brain Metastases Compared to Normal Tissues: A Pilot MR Fingerprinting Study.

The purpose of the present pilot study was to estimate T1 and T2 metric values derived simultaneously from a new, rapid Magnetic Resonance Fingerprinting (MRF) technique, as well as to assess their ability to characterize-brain metastases (BM) and normal-appearing brain tissues. Fourteen patients with BM underwent MRI, including prototype MRF, on a 3T scanner. In total, 108 measurements were analyzed: 42 from solid parts of BM's (21 each on T1 and T2 maps) and 66 from normal-appearing brain tissue (11 ROIs each on T1 and T2 maps for gray matter [GM], white matter [WM], and cerebrospinal fluid [CSF]). The BM's mean T1 and T2 values differed significantly from normal-appearing WM (p < 0.05). The mean T1 values from normal-appearing GM, WM, and CSF regions were 1205 ms, 840 ms, and 4233 ms, respectively. The mean T2 values were 108 ms, 78 ms, and 442 ms, respectively. The mean T1 and T2 values for untreated BM (n = 4) were 2035 ms and 168 ms, respectively. For treated BM (n = 17) the T1 and T2 values were 2163 ms and 141 ms, respectively. MRF technique appears to be a promising and rapid quantitative method for the characterization of free water content and tumor morphology in BMs.

Qualitative and Quantitative Performance of Magnetic Resonance Image Compilation (MAGiC) Method: An Exploratory Analysis for Head and Neck Imaging.

The present exploratory study investigates the performance of a new, rapid, synthetic MRI method for diagnostic image quality assessment and measurement of relaxometry metric values in head and neck (HN) tumors and normal-appearing masseter muscle. The multi-dynamic multi-echo (MDME) sequence was used for data acquisition, followed by synthetic image reconstruction on a 3T MRI scanner for 14 patients (3 untreated and 11 treated). The MDME enables absolute quantification of physical tissue properties, including T1 and T2, with a shorter scan time than the current state-of-the-art methods used for relaxation measurements. The vendor termed the combined package MAGnetic resonance imaging Compilation (MAGiC). In total, 48 regions of interest (ROIs) were analyzed, drawn on normal-appearing masseter muscle and tumors in the HN region. Mean T1 and T2 values obtained from normal-appearing muscle were 880 ± 52 ms and 46 ± 3 ms, respectively. Mean T1 and T2 values obtained from tumors were 1930 ± 422 ms and 77 ± 13 ms, respectively, for the untreated group, 1745 ± 410 ms and 107 ± 61 ms, for the treated group. A total of 1552 images from both synthetic MRI and conventional clinical imaging were assessed by the radiologists to provide the rating for T1w and T2w image contrasts. The synthetically generated qualitative T2w images were acceptable and comparable to conventional diagnostic images (93% acceptability rating for both). The acceptability ratings for MAGiC-generated T1w, and conventional images were 64% and 100%, respectively. The benefit of MAGiC in HN imaging is twofold, providing relaxometry maps in a clinically feasible time and the ability to generate a different combination of contrast images in a single acquisition.

Quantitative Synthetic Magnetic Resonance Imaging for Brain Metastases: A Feasibility Study.

The present preliminary study aims to characterize brain metastases (BM) using T1 and T2 maps generated from newer, rapid, synthetic MRI (MAGnetic resonance image Compilation; MAGiC) in a clinical setting. We acquired synthetic MRI data from 11 BM patients on a 3T scanner. A multiple-dynamic multiple-echo (MDME) sequence was used for data acquisition and synthetic image reconstruction, including post-processing. MDME is a multi-contrast sequence that enables absolute quantification of physical tissue properties, including T1 and T2, independent of the scanner settings. In total, 82 regions of interest (ROIs) were analyzed, which were obtained from both normal-appearing brain tissue and BM lesions. The mean values obtained from the 48 normal-appearing brain tissue regions and 34 ROIs of BM lesions (T1 and T2) were analyzed using standard statistical methods. The mean T1 and T2 values were 1143 ms and 78 ms, respectively, for normal-appearing gray matter, 701 ms and 64 ms for white matter, and 4206 ms and 390 ms for cerebrospinal fluid. For untreated BMs, the mean T1 and T2 values were 1868 ms and 100 ms, respectively, and 2211 ms and 114 ms for the treated group. The quantitative T1 and T2 values generated from synthetic MRI can characterize BM and normal-appearing brain tissues.

Longitudinal Monitoring of Simulated Interstitial Fluid Pressure for Pancreatic Ductal Adenocarcinoma Patients Treated with Stereotactic Body Radiotherapy.

The present study aims to monitor longitudinal changes in simulated tumor interstitial fluid pressure (IFP) and velocity (IFV) values using dynamic contrast-enhanced (DCE)-MRI-based computational fluid modeling (CFM) in pancreatic ductal adenocarcinoma (PDAC) patients. Nine PDAC patients underwent MRI, including DCE-MRI, on a 3-Tesla MRI scanner at pre-treatment (TX (0)), after the first fraction of stereotactic body radiotherapy (SBRT, (D1-TX)), and six weeks post-TX (D2-TX). The partial differential equation of IFP formulated from the continuity equation, incorporating the Starling Principle of fluid exchange, Darcy velocity, and volume transfer constant (Ktrans), was solved in COMSOL Multiphysics software to generate IFP and IFV maps. Tumor volume (Vt), Ktrans, IFP, and IFV values were compared (Wilcoxon and Spearman) between the time- points. D2-TX Ktrans values were significantly different from pre-TX and D1-TX (p < 0.05). The D1-TX and pre-TX mean IFV values exhibited a borderline significant difference (p = 0.08). The IFP values varying <3.0% between the three time-points were not significantly different (p > 0.05). Vt and IFP values were strongly positively correlated at pre-TX (ρ = 0.90, p = 0.005), while IFV exhibited a strong negative correlation at D1-TX (ρ = -0.74, p = 0.045). Vt, Ktrans, IFP, and IFV hold promise as imaging biomarkers of early response to therapy in PDAC.

Cost-Effective, Integrated In Vitro Phantoms for Cardiac MRI.

To provide lab scale in vitro phantom solutions for cardiac MR (CMR) studies that can be used for imaging structure and function as well as calorimetric measurements. The phantoms were purposed to accept user inputs such as beats per minute (BPM) and flow rate. We developed two generations of phantoms. The first phantom was developed using poly vinyl alcohol driven by a mechanical setup. The second was a 3D-printed phantom controlled through a user interface (UI) and a peristaltic pump. These phantoms were scanned for the characteristics mentioned above, which were qualitatively and quantitatively assessed through postprocessing of CMR images and compared with in vivo data.

Computational Modeling of Interstitial Fluid Pressure and Velocity in Head and Neck Cancer Based on Dynamic Contrast-Enhanced Magnetic Resonance Imaging: Feasibility Analysis.

We developed and tested the feasibility of computational fluid modeling (CFM) based on dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) for quantitative estimation of interstitial fluid pressure (IFP) and velocity (IFV) in patients with head and neck (HN) cancer with locoregional lymph node metastases. Twenty-two patients with HN cancer, with 38 lymph nodes, underwent pretreatment standard MRI, including DCE-MRI, on a 3-Tesla scanner. CFM simulation was performed with the finite element method in COMSOL Multiphysics software. The model consisted of a partial differential equation (PDE) module to generate 3D parametric IFP and IFV maps, using the Darcy equation and Ktrans values (min-1, estimated from the extended Tofts model) to reflect fluid influx into tissue from the capillary microvasculature. The Spearman correlation (ρ) was calculated between total tumor volumes and CFM estimates of mean tumor IFP and IFV. CFM-estimated tumor IFP and IFV mean ± standard deviation for the neck nodal metastases were 1.73 ± 0.39 (kPa) and 1.82 ± 0.9 × (10-7 m/s), respectively. High IFP estimates corresponds to very low IFV throughout the tumor core, but IFV rises rapidly near the tumor boundary where the drop in IFP is precipitous. A significant correlation was found between pretreatment total tumor volume and CFM estimates of mean tumor IFP (ρ = 0.50, P = 0.004). Future studies can validate these initial findings in larger patients with HN cancer cohorts using CFM of the tumor in concert with DCE characterization, which holds promise in radiation oncology and drug-therapy clinical trials.

Diffusion-Weighted Echo Planar Imaging using MUltiplexed Sensitivity Encoding and Reverse Polarity Gradient in Head and Neck Cancer: An Initial Study.

We aimed to compare the geometric distortion (GD) correction performance and apparent diffusion coefficient (ADC) measurements of single-shot diffusion-weighted echo-planar imaging (SS-DWEPI), multiplexed sensitivity encoding (MUSE)-DWEPI, and MUSE-DWEPI with reverse-polarity gradient (RPG) in phantoms and patients. We performed phantom studies at 3T magnetic resonance imaging (MRI) using the American College of Radiology phantom and Quantitative Imaging Biomarker Alliance DW-MRI ice-water phantom to assess GD and effect of distortion in the measurement of ADC, respectively. Institutional review board approved the prospective clinical component of this study. DW-MRI data were obtained from 11 patients with head and neck cancer using these three DW-MRI methods. Wilcoxon signed-rank (WSR) and Kruskal-Wallis (KW) tests were used to compare ADC values, and qualitative rating by radiologist between three DW-MRI methods. In the ACR phantom, GD of 0.17% was observed for the b = 0 s/mm2 image of the MUSE-DWEPI with RPG method compared with that of 1.53% and 2.1% of MUSE-DWEPI and SS-DWEPI, respectively; The corresponding methods root-mean-square errors were 0.58, 3.37, and 5.07 mm. WSR and KW tests showed no significant difference in the ADC measurement between these three DW-MRI methods for both healthy masseter muscles and neoplasms (P > .05). We observed improvement in spatial accuracy for MUSE-DWEPI with RPG in the head and neck region with a higher correlation (R2 = 0.791) compared with that for SS-DWEPI (R2 = 0.707) and MUSE-DWEPI (R2 = 0.745). MUSE-DWEPI with RPG significantly reduces the distortion compared with MUSE-DWEPI or conventional SS-DWEPI techniques, and the ADC values were similar.

Diffusion-Weighted and Dynamic Contrast-Enhanced MRI Derived Imaging Metrics for Stereotactic Body Radiotherapy of Pancreatic Ductal Adenocarcinoma: Preliminary Findings.

We aimed to assess longitudinal changes in quantitative imaging metric values obtained from diffusion-weighted (DW-) and dynamic contrast-enhanced magnetic resonance imaging (DCE)-MRI at pre-treatment (TX[0]), immediately after the first fraction of stereotactic body radiotherapy (D1-TX[1]), and 6 weeks post-TX (Post-TX[2]) in patients with pancreatic ductal adenocarcinoma. Ten enrolled patients (n = 10) underwent DW- and DCE-MRI examinations on a 3.0 T scanner. The apparent diffusion coefficient, ADC (mm2/s), was derived from DW imaging data using a monoexponential model. The tissue relaxation rate, R1t, time-course data were fitted with a shutter-speed model, which provides estimates of the volume transfer constant, Ktrans (min-1), extravascular extracellular volume fraction, ve , and mean lifetime of intracellular water protons, τ i (seconds). Wilcoxon rank-sum test compared the mean values, standard deviation, skewness, kurtosis, and relative percentage (r, %) changes (Δ) in ADC, Ktrans, ve , and τ i values between the magnetic resonance examinations. rADCΔ2-0 values were significantly greater than rADCΔ1-0 values (P = .009). rKtransΔ2-0 values were significantly lower than rKtransΔ1-0 values (P = .048). rveΔ2-1 and rveΔ2-0 values were significantly different (P = .016). rτ iΔ2-1 values were significantly lower than rτ iΔ2-0 values (P = .008). For group comparison, the pre-TX mean and kurtosis of ADC (P = .18 and P = .14), skewness and kurtosis of Ktrans values (P = .14 for both) showed a leaning toward significant difference between patients who experienced local control (n = 2) and failed early (n = 4). DW- and DCE-MRI-derived quantitative metrics could be useful biomarkers to evaluate longitudinal changes to stereotactic body radiotherapy in patients with pancreatic ductal adenocarcinoma.

Accelerated dynamic contrast enhanced MRI based on region of interest compressed sensing.

Magnetic Resonance Imaging (MRI) provides excellent soft tissue contrast with one significant limitation of slow data acquisition. Dynamic Contrast Enhanced MRI (DCE-MRI) is one of the widely employed techniques to estimate tumor tissue physiological parameters using contrast agents. DCE-MRI data acquisition and reconstruction requires high spatiotemporal resolution, especially during the post-contrast phase. The region of Interest Compressed Sensing (ROICS) is based on Compressed Sensing (CS) framework and works on the hypothesis that limiting CS to an ROI can achieve superior CS performance. In this work, ROICS has been demonstrated on breast DCE-MRI data at chosen acceleration factors and the results are compared with conventional CS implementation. Normalized Root Mean Square Error (NRMSE) was calculated to compare ROICS with CS quantitatively. CS and ROICS reconstructed images were used to compare Ktrans and ve values derived using standard Tofts Model (TM). This also validated the superior performance of ROICS over conventional CS. ROICS generated Concentration Time Curves (CTC's) at chosen acceleration factors follow similar trend as the ground truth data as compared to CS. Both qualitative and quantitative analyses show that ROICS outperforms CS particularly at acceleration factors of 5× and above.

Repeatability of Quantitative Diffusion-Weighted Imaging Metrics in Phantoms, Head-and-Neck and Thyroid Cancers: Preliminary Findings.

The aim of this study was to establish the repeatability measures of quantitative Gaussian and non-Gaussian diffusion metrics using diffusion-weighted imaging (DWI) data from phantoms and patients with head-and-neck and papillary thyroid cancers. The Quantitative Imaging Biomarker Alliance (QIBA) DWI phantom and a novel isotropic diffusion kurtosis imaging phantom were scanned at 3 different sites, on 1.5T and 3T magnetic resonance imaging systems, using standardized multiple b-value DWI acquisition protocol. In the clinical component of this study, a total of 60 multiple b-value DWI data sets were analyzed for test-retest, obtained from 14 patients (9 head-and-neck squamous cell carcinoma and 5 papillary thyroid cancers). Repeatability of quantitative DWI measurements was assessed by within-subject coefficient of variation (wCV%) and Bland-Altman analysis. In isotropic diffusion kurtosis imaging phantom vial with 2% ceteryl alcohol and behentrimonium chloride solution, the mean apparent diffusion (Dapp × 10-3 mm2/s) and kurtosis (Kapp, unitless) coefficient values were 1.02 and 1.68 respectively, capturing in vivo tumor cellularity and tissue microstructure. For the same vial, Dapp and Kapp mean wCVs (%) were ≤1.41% and ≤0.43% for 1.5T and 3T across 3 sites. For pretreatment head-and-neck squamous cell carcinoma, apparent diffusion coefficient, D, D*, K, and f mean wCVs (%) were 2.38%, 3.55%, 3.88%, 8.0%, and 9.92%, respectively; wCVs exhibited a higher trend for papillary thyroid cancers. Knowledge of technical precision and bias of quantitative imaging metrics enables investigators to properly design and power clinical trials and better discern between measurement variability versus biological change.

Application of Region of Interest Compressed Sensing to accelerate magnetic resonance angiography.

Magnetic Resonance Angiography (MRA) is a group of techniques based on Magnetic Resonance Imaging (MRI) to image blood vessels. Compressed Sensing (CS) is a mathematical framework to reconstruct MR images from sparse data to minimize the data acquisition time. Image sparsity is the key in CS to reconstruct MR images. CS technique allows reconstruction from significantly fewer k-space samples as compared to full k-space acquisition, which results in reduced MRI data acquisition time. The images resulting from MRA are sparse in native representation, hence yielding themselves well to CS. Recently our group has proposed a novel CS method called Region of Interest Compressed Sensing (ROICS) as a part of Region of Interest (ROI) weighted CS. This work aims at the implementation of ROICS for the first time on MRA data to reduce MR data acquisition time. It has been demonstrated qualitatively and quantitatively that ROICS outperforms CS at higher acceleration factors. ROICS technique has been applied to 3D angiograms of the brain data acquired at 1.5T. It helps to reduce the MRA data acquisition time and improves the visualization of arteries. ROICS technique has been applied on 4 brain angiogram data sets at different acceleration factors from 2× to 10×. Reconstructed images show ROICS technique performs better than conventional CS technique and is quantified by the comparative Signal to Noise Ratio (SNR) in the ROI.

Compressed sensing MRI: a review.

Compressed sensing (CS) is a mathematical framework that reconstructs data from highly undersampled measurements. To gain acceleration in acquisition time, CS has been applied to MRI and has been demonstrated on diverse MRI methods. This review discusses the important requirements to qualify MRI to become an optimal application of CS, namely, sparsity, pseudo-random undersampling, and nonlinear reconstruction. By utilizing concepts of transform sparsity and compression, CS allows acquisition of only the important coefficients of the signal during the acquisition. A priori knowledge of MR images specifically related to transform sparsity is required for the application of CS. In this paper, Section I introduces the fundamentals of CS and the idea of CS as applied to MRI. The requirements for application of CS to MRI is discussed in Section II, while the various acquisition techniques, reconstruction techniques, the advantages of combining CS and parallel imaging, and sampling mask design problems are discussed in Section III. Numerous applications of CS in MRI due to its ability to improve imaging speed are reviewed in section IV. Clinical evaluations of some of the CS applications recently published are discussed in Section V. Section VI provides information on available open source software that could be used for CS implementations.