State estimation in dynamic MRI

We propose a state estimation approach for functional magnetic resonance imaging (fMRI). In state estimation, time-dependent image reconstruction problem is modeled by separate state evolution and observation models, and the objective is to estimate the time series of system states, given the models and the time-dependent measurement data. Our method computes the state estimates by using the Kalman filter (KF) and Kalman smoother (KS) algorithms. We propose to complement the state estimation formulation with a structural prior which can be derived from the anatomical MRI, acquired as a part of the fMRI protocol. Two different constructions of the structural prior are considered. The first one is a structured smoothness prior where the state observation matrix is augmented with a spatially weighted regularization matrix which promotes structural similarity of the gradient of the unknown image with the gradient of the anatomical image. The second approach is based on applying structured total variation denoising to the KF estimate at each time step of the Kalman recursions. The proposed approaches are evaluated using simulated and experimental, radially sampled, small animal fMRI data from a rat brain. In our method, the state estimates are updated after each new spoke of radial data becomes available, leading to faster frame rate compared to the conventional image reconstruction approaches. The results are compared to a sliding window method and a conventional reconstruction which produces new image only after a full circle of k-space spokes becomes available. The results suggest that the state estimation approach with the structural prior can improve both spatial and temporal resolution of fMRI.

In this paper, a new neural network model inspired by the biological immune system functions is presented. The model, termed Artificial Immune-Activated Neural Network (AIANN), extracts classification knowledge from a training data set, which is then used to classify input patterns or vectors. The AIANN is based on a neuron activation function whose behavior is conceptually modeled after the chemical bonds between the receptors and epitopes in the biological immune system. The bonding is controlled through an energy measure to ensure accurate recognition. The AIANN model was applied to the segmentation of 3-dimensional magnetic resonance imaging (MRI) data of the brain and a contextual basis was developed for the segmentation problem. Evaluation of the segmentation results was performed using both real MRI data obtained from the Center for Morphometric Analysis at Massachusetts General Hospital and simulated MRI data generated using the McGill University BrainWeb MRI simulator. Experimental results demonstrated that the AIANN model attained higher average results than those obtained using published methods for real MRI data and simulated MRI data, especially at low levels of noise.

Accelerating Dynamic Magnetic Resonance Imaging (MRI) for Lung Tumor Tracking Based on Low-Rank Decomposition in the Spatial–Temporal Domain: A Feasibility Study Based on Simulation and Preliminary Prospective Undersampled MRI

Dynamic MR imaging in Tolosa–Hunt syndrome

Dynamic contrast-enhanced magnetic resonance imaging (MRI) is the modality of choice for the diagnosis of pituitary microadenomas; however, it may be associated with a relatively high false-negative rate, especially in small lesions. To evaluate the usefulness of subtraction images for enhancing the visual detection of pituitary microadenomas. In total, 50 patients with clinically established diagnosis hyperprolactinemia, acromegaly, and Cushing's disease were enrolled. Ten patients referred for brain MRI for reasons other than pituitary abnormality were selected as control group. Routine dynamic MRI of the pituitary gland and obtained subtraction MRI scans were scrutinized separately on different sessions by an experienced radiologist blinded to the study design and patient's data. The investigator's opinion on the presence or absence of a lesion and lesion size were collected. In patients with pituitary microadenoma, dynamic MRI images were reported positive in 42 (84%) patients and negative in 8 (16%). Subtraction images were described as positive in all patients (100%)-including all patients with negative dynamic MRI-and the difference was statistically significant (P=0.016). Undetected lesions on dynamic MRI had a mean size of 2.84 ± 1.79 mm (median= 2.20 mm, interquartile range=1.62-4.62 mm) and a significant inverse correlation was noted between lesion size and negative report of dynamic MRI (P=0.018). Brain MRI scans in the control group were reported negative for pituitary microadenoma in both dynamic contrast-enhanced and subtraction images. Subtraction images can successfully identify all lesions detectable with conventional dynamic MRI as well as improving visualization of lesions undetected on dynamic MRI, especially in small lesions.

e17105 Background: In patients suspected of prostate cancer, the decision to perform a biopsy hinges on available clinical data. Magnetic resonance imaging (MRI) and digital rectal exam (DRE) data are informative but may not be available. We created accurate models for clinically significant prostate cancer (csPCa), flexible to MRI and DRE data availability. Methods: Optimized ensembles of calibrated random forest models predicting csPCa (Grade Group ≥2) used total PSA, free PSA, prior negative biopsy status, and age, with or without DRE and MRI data (prostate volume and PI-RADS score). Risk models were derived (training cohorts n=1257 to 2191) and validated (validation cohorts n=317 to 1257) from different clinical sites. Models were evaluated by the area under the receiver operating characteristic curve (ROC AUC), sensitivity, specificity, positive predictive value, and negative predictive value, using thresholds providing ~ 95% sensitivity. Feature importance was determined by SHAP analysis. Results: All models had an AUC of at least 0.80, showing that predicting csPCa can be accurate without MRI or DRE data. Including MRI data significantly increased the AUC in the validation cohort (ClarityDX Prostate 0.80 vs ClarityDX Prostate +MRI 0.87). DRE had moderate value for models without MRI data (ClarityDX Prostate 0.80 vs ClarityDX Prostate +DRE 0.82) and minor value with MRI data (ClarityDX Prostate +MRI vs ClarityDX Prostate +DRE+MRI; AUC 0.87 vs 0.87; specificity 45% vs 47%, Table). Mean absolute SHAP values were highest for PI-RADS and prostate volume. Conclusions: These optimized risk models provide high accuracy for predicting csPCa in various clinical settings. Including MRI data greatly increases model accuracy, while DRE has a smaller effect on model accuracy. [Table: see text]

We assessed the imaging features of intraductal papillary mucinous neoplasms (IPMNs) of the pancreas paying special attention to underlying pancreatic fibrosis on three-phase helical computed tomography (CT) and dynamic magnetic resonance (MR) imaging. Sixteen patients with histopathologically proven IPMNs underwent three-phase helical CT and dynamic MR imaging. IPMNs were seen as a cluster of cyst-like structures in branch duct (n = 5) and combined types (n = 10), and as a fusiform appearance in the main duct type (n = 1). IPMN shape was most easily visualized at the portal venous dominant phase or delayed phase owing to rim-like enhancement of the dilated ducts. Pathologically mild to severe fibrosis was seen on this enhanced rim replacing the surrounding pancreatic parenchyma. Communication between the dilated branch ducts and main pancreatic duct was identified in 15 patients on helical CT and 14 patients on dynamic MR imaging. In patients with fibrosis of pancreatic parenchyma surrounding this, communication was most easily visualized at the later phase on CT and MR imaging. Adenocarcinomas were depicted as papillary projections in eight of nine patients on CT and MR imaging. Invasion of the pancreatic parenchyma was seen in five of six patients as a solid mass in the pancreatic parenchyma. These masses were most easily visualized at the arterial dominant phase on both CT and MR imaging. Three-phase helical CT and dynamic MR imaging were useful in the diagnosis of IPMN of the pancreas.

Electrocardiographic mapping (ECGM) is to estimate the cardiac activities from the measured body surface potentials (BSPs), in which the epicardial potentials (EPs) is often reconstructed. One of the challenges in ECGM problem is its ill-posedness, and regularization techniques are needed to obtain the clinically reasonable solutions. The total variation (TV) method has been validated in keeping the sharp edges and has found some preliminary applications in ECG inverse problem. In this study, we applied and compared two algorithms: lagged diffusivity (LD) fixed point iteration and primal dual-interior point method (PD-IPM), to implement TV regularization method in ECGM problem. With a realistic heart-lung-torso model, the TV methods are tested and compared to the L2-norm regularization methods in zero- and first-order. The simulation results demonstrate that the TV method can generate better EPs compared to the zero-order Tikhonov method. Compared to the first-order Tikhonov method, the TV’s results are much sharper. For the two algorithms in TV method, the LD algorithm seems more robust than the PD-IPM in ECGM problem, though the PD-IPM converges faster.KeywordsBoundary Element MethodElectrical Impedance TomographyFixed Point IterationTotal Variation RegularizationTikhonov MethodThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Magnetic resonance (MR) imaging is rapidly gaining acceptance as an accurate, reproducible, noninvasive method for optimal assessment of structural and functional parameters in patients with valvular heart disease. The severity of valvular regurgitation can be evaluated with cine gradient-echo MR imaging, which allows measurement of the area of the signal void corresponding to the abnormal flow jet. Alternatively, this modality can be used to obtain ventricular volumetric measurements and calculate the regurgitant fraction, or velocity-encoded cine (VEC) MR imaging can be used to quantify regurgitant blood flow. The severity of valvular stenosis can be determined by evaluating the flow jet and associated findings with either modality or by using VEC MR imaging to calculate the transvalvular pressure gradient and valve area. Dynamic MR imaging allows accurate assessment of ventricular function and comprehensive evaluation of pathophysiologic changes. In addition, good interstudy reproducibility suggests a role for VEC MR imaging in assessing the effects of therapeutic intervention and monitoring regurgitant fraction, thereby helping in surgical planning and the prevention of ventricular dysfunction. With greater cost-effectiveness and the increasing availability of new hardware and more advanced techniques, MR imaging will become a routine procedure in valvular heart disease.

Background: This study evaluates the ability to predict the response of locally advanced breast cancers to neoadjuvant therapy (NAT) using patient-specific magnetic resonance imaging (MRI) data and a biophysical mathematical model. The 3D mathematical model consists of three parts: tumor cell proliferation, tumor spread (diffusion), and treatment. In particular, the tumor cells proliferate according to logistic growth, and the diffusion term is coupled to the mechanical properties of the surrounding fibroglandular and adipose tissues to inform individual tumor growth patterns (specific to each patient’s anatomy). The model’s treatment term accounts for tumor cell reduction according to approximate local drug delivery for each patient. Methods: Patients (N = 21) with intermediate to high grade invasive breast cancers with varying receptor status, who were eligible for NAT as a component of their clinical care, were recruited. Each patient was treated with standard-of-care consisting of one or two NAT regimens in sequence followed by surgical resection of any residual tumor. MRI data are acquired at four time points: 1) prior to initiation of NAT, 2) after 1 cycle of NAT, 3) after 2-4 cycles of NAT, and 4) 1 cycle after scan 3. The MRI data is processed and evaluated using our semi-automated pipeline. Specifically, diffusion-weighted MRI data is utilized to characterize the cellularity throughout the tumor tissue, and dynamic contrast-enhanced (DCE-) MRI data is used to segment the breast tissue and analyze the local drug delivery using pharmacokinetic analysis and population-derived plasma curves of drug concentrations. The model’s predictive ability is assessed using three different strategies. First, the model is calibrated using each patient’s first two scans to enable predictions of the total tumor cellularity, volume, and longest axis that are directly compared to the values measured from their third scan. Second, the model’s predictions for tumor response are compared to the corresponding response evaluation criteria in solid tumors (RECIST) results. Third, the model is re-calibrated using scans 3 and 4 and simulated to the time of surgery to compare the model’s predictions to each patient’s response status determined by surgical pathology. Results: Calibrating the model with MRI data for one cycle of therapy yields predictions strongly correlated with tumor response measured from each patient’s third scan, concordance correlation coefficients of 0.91, 0.90, and 0.86 for total cellularity, volume, and longest axis, respectively (p < 0.01, N = 18). The model’s predictions are significantly (p < 0.01) correlated with tumor response as designated by RECIST for the cohort. Specifically, the model predicts greater percent reduction in the longest axis for the RECIST designated responder group (i.e., complete response and partial response) compared to non-responders. At the time of surgery, the model predicts changes in total tumor cellularity from baseline that are significantly (p < 0.01) correlated with pathological response status—an area under the receiver operator characteristic curve of 0.92 and a sensitivity and specificity of 1.0 and 0.74, respectively. Discussion: These preliminary results suggest that this clinical-mathematical approach can be predictive of tumor response very early in the course of NAT on a patient-specific basis. Moreover, the study was performed in the community-care setting across a heterogenous group of patients, indicating the approach may be practical for wide-spread application. NCI U01 CA174706, NCI U01 CA154602, CPRIT RR160005, ACS-RSG-18-006-01-CCE, R01CA240589, NCI-U24CA226110, CPRIT RR160093 Citation Format: Angela M Jarrett, David A. Hormuth, II, Anum K Syed, Chengyue Wu, John Virostko, Anna G Sorace, Julie C DiCarlo, Jeanne Kowalski, Debra Patt, Boone Goodgame, Sarah Avery, Thomas E Yankeelov. Predicting breast cancer response to neoadjuvant therapies using a mathematical model individualized with patient-specific magnetic resonance imaging data: Preliminary Results [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PS13-18.

Unpaired multi-modal training and single-modal testing for detecting signs of endometriosis.

To prospectively determine clinical value of dynamic contrast material-enhanced magnetic resonance (MR) imaging in the evaluation of disease activity in perianal Crohn disease (CD). Patients provided written informed consent. Study approval was waived by an institutional review board. Thirty-three patients with perianal CD underwent pelvic MR imaging; 17 were male (mean age, 37.4 years +/- 10.8 [standard deviation]; age range, 18-54 years) and 16 were female (mean age, 32.0 years +/- 8.3; age range, 16-43 years). Dynamic contrast-enhanced MR imaging was performed; time-intensity curves (TICs) were obtained. Each pixel was classified as one of six predefined TIC shape types. For each MR imaging examination, a region of interest (ROI) was drawn around the fistula on the single section corresponding to the most extensive and most hyperintense lesion; maximum enhancement (ME), slope of enhancement, and TIC shapes were calculated. Absolute and relative numbers of pixels for each curve type were calculated in a two-dimensional ROI. These results were compared with Perianal Disease Activity Index (PDAI), C-reactive protein (CRP) level, an MR imaging-based severity score, and clinical outcome. A Spearman rank correlation test was used to calculate correlation coefficients between dynamic contrast-enhanced MR imaging parameters and reference parameters. A Mann-Whitney U test was used to calculate differences in dynamic contrast-enhanced MR imaging parameters between predefined groups of patients. Significant correlations were found between the absolute amounts of the TIC shape types and PDAI and between ROI volume and PDAI. The ratio of quickly enhancing versus slowly enhancing pixels correlated with higher MR imaging scores as did the ROI volume. The absolute amounts of pixels displaying TIC types 2, 3, 4, and 5 correlated significantly with MR imaging score. CRP level showed a significant correlation with mean ME. Larger numbers of quickly enhancing pixels were observed in patients who needed medication changes or developed new abscesses during follow-up. Dynamic contrast-enhanced MR imaging can help determine disease activity in perianal CD and might be helpful in selecting a subpopulation of patients who should be monitored more closely for development of more extensive disease.

3D Reconstructions of the Cerebral Ventricles and Volume Quantification in Children with Brain Malformations

BackgroundBecause of superior soft tissue contrast, the use of magnetic resonance imaging (MRI) as a complement to computed tomography (CT) in the target definition procedure for radiotherapy is increasing. To keep the workflow simple and cost effective and to reduce patient dose, it is natural to strive for a treatment planning procedure based entirely on MRI. In the present study, we investigate the dose calculation accuracy for different treatment regions when using bulk density assignments on MRI data and compare it to treatment planning that uses CT data.MethodsMR and CT data were collected retrospectively for 40 patients with prostate, lung, head and neck, or brain cancers. Comparisons were made between calculations on CT data with and without inhomogeneity corrections and on MRI or CT data with bulk density assignments. The bulk densities were assigned using manual segmentation of tissue, bone, lung, and air cavities.ResultsThe deviations between calculations on CT data with inhomogeneity correction and on bulk density assigned MR data were small. The maximum difference in the number of monitor units required to reach the prescribed dose was 1.6%. This result also includes effects of possible geometrical distortions.ConclusionsThe dose calculation accuracy at the investigated treatment sites is not significantly compromised when using MRI data when adequate bulk density assignments are made. With respect to treatment planning, MRI can replace CT in all steps of the treatment workflow, reducing the radiation exposure to the patient, removing any systematic registration errors that may occur when combining MR and CT, and decreasing time and cost for the extra CT investigation.

Segmentation of brain magnetic resonance imaging (MRI) data plays an important role in the computer-aided diagnosis and neuroscience research. Fuzzy c-means (FCM) clustering algorithm is one of the most usually used techniques for brain MRI image segmentation because of its fuzzy nature. However, the conventional FCM method fails to carry out segmentation well enough due to intensity inhomogeneity in MRI data. To overcome this issue, we propose an improved algorithm based on FCM clustering for segmentation of brain MRI data. Specifically, we modify the conventional FCM algorithm to allow for intensity inhomogeneity by introducing the regularization of the neighborhood influence and bias field. Results show that our proposed algorithm obtains reasonable segmentation of white matter (WM), gray matter (GM), and cerebrospinal fluid (CSF) from MRI data, which is superior to the expectation-maximization (EM) and conventional FCM methods.