BACKGROUND:Signal-to-noise ratio (SNR) measurements are important for quality assurance, especially when quantitative MRI markers or new hardware are being developed or evaluated. Regions of low signal or assumed uniform signal are used for SNR estimation in a variety of MR image types, but this is unreliable due to spatial heterogeneity of noise arising from the organization of multiple-detector arrays and sensitivity encoding (SENSE) reconstruction. PURPOSE:We present a universally applicable method for empirical noise mapping by propagating simulated noise through the image-generating reconstruction pipeline, to identify optimal regions of interest (ROIs) for determining noise levels in prostate diffusion-weighted imaging (DWI). METHODS:Thirty-six men were imaged at 3T with DWI. K-space data were propagated through an off-line reconstruction pipeline including sensitivity encoding (SENSE), to generate DW images. Simulated complex Gaussian noise generated from prescan per-channel measurements was similarly processed to produce pure noise maps. Per-voxel standard deviations (SD) of spatial distributions of noise, averaged over prostate, internal obturator muscle (IOM), low signal (anterior and posterior), rectum, and bladder ROIs were calculated and compared to evaluate suitability for SNR estimation in prostate DWI. RESULTS:IOM ROI noise levels, averaged bilaterally, were not significantly different from those in the prostate. Bladder ROI noise levels were comparable. Noise levels measured in low signal areas were highly dependent on SENSE algorithm regularization strength. DISCUSSION:The proposed method can be used-on any scanner-to accurately calculate noise levels in MR images, without repeated acquisition or interference from SENSE use. CONCLUSION:Bladder and averaged IOM ROIs are appropriate for estimating DWI noise levels in the prostate. Low signal ROIs should not be used, due to the strong dependence on the SENSE regularization strength, and by extension, on the SENSE reconstruction algorithm. The presented method can be used instead to obtain accurate spatial maps of noise levels.

Magnetic Resonance Imaging (MRI) is vital for prostate cancer (PCa) diagnosis. While advanced techniques such as Hybrid Multi-dimensional MRI (HM-MRI) have enhanced diagnostic capabilities, the significant need remains for robust, automated Artificial Intelligence (AI)-based detection methods. In this study, we combine quantitative HM-MRI of tissue composition with an AI-based neural network. We propose the Hadamard-Bias Network plus ResNet18 (HBR-Net-18), a two-stage AI framework for PCa detection. In the first stage, a Hadamard U-Net-based algorithm suppresses intensity inhomogeneities (bias fields) across six parametric HM-MRI maps generated via a Physics-Informed Autoencoder (PIA). In the second stage, a Residual Network (ResNet-18) performs patch-level classification. The framework utilizes overlapping 11-by-11 patches, incorporating both 2D intra-slice and 3D inter-slice (adjacent-slice) information to improve spatial consistency. Our experimental results demonstrate that HB-Net achieves balanced sensitivity and specificity, significantly outperforming conventional radiomics-based approaches and baseline CNN models, highlighting its potential for clinical deployment.

This study demonstrates that hybrid multidimensional MRI has the potential to improve MRI/US fusion biopsy results, provide more reliable detection of prostate cancer, and add complementary information to Prostate Imaging and Reporting Data System–based evaluation by expert radiologists.

Prostate cancer is one of the leading causes of cancer-related morbidity and mortality worldwide, and imaging plays a critical role in its detection, localization, staging, treatment, and management. The advent of artificial intelligence (AI) has introduced transformative possibilities in prostate imaging, offering enhanced accuracy, efficiency, and consistency. This review explores the integration of AI in prostate cancer diagnostics across key imaging modalities, including multiparametric MRI (mpMRI), PSMA PET/CT, and transrectal ultrasound (TRUS). Advanced AI technologies, such as machine learning, deep learning, and radiomics, are being applied for lesion detection, risk stratification, segmentation, biopsy targeting, and treatment planning. AI-augmented systems have demonstrated the ability to support PI-RADS scoring, automate prostate and tumor segmentation, guide targeted biopsies, and optimize radiation therapy. Despite promising performance, challenges persist regarding data heterogeneity, algorithm generalizability, ethical considerations, and clinical implementation. Looking ahead, multimodal AI models integrating imaging, genomics, and clinical data hold promise for advancing precision medicine in prostate cancer care and assisting clinicians, particularly in underserved regions with limited access to specialists. Continued multidisciplinary collaboration will be essential to translate these innovations into evidence-based practice. This article explores current AI applications and future directions that are transforming prostate imaging and patient care.

Purpose: To evaluate the performance of Physics-Informed Autoencoder (PIA), a self-supervised deep learning model, in measuring tissue-based biomarkers for prostate cancer (PCa) using hybrid multidimensional MRI. Materials and Methods: This retrospective study introduces PIA, an emerging self-supervised deep learning model that integrates a three-compartment diffusion-relaxation model with hybrid multidimensional MRI. PIA was trained to encode the biophysical model into a deep neural network to predict measurements of tissue-specific biomarkers for PCa without extensive training data requirements. Comprehensive in silico and in vivo experiments, using histopathology measurements as the reference standard, were conducted to validate the model's efficacy in comparison to the traditional nonlinear least squares (NLLS) algorithm. PIA's robustness to noise was tested in in silico experiments with varying signal-to-noise ratio (SNR) conditions, and in vivo performance for estimating volume fractions was evaluated in 21 patients (mean age, 60 years +/- 6.6 [SD]; all male) with PCa (71 regions of interest). Evaluation metrics included the intraclass correlation coefficient (ICC) and Pearson correlation coefficient. Results: PIA predicted the reference standard tissue parameters with high accuracy, outperforming conventional NLLS methods, especially under noisy conditions (rs = 0.80 vs 0.65, P < .001 for epithelium volume at SNR of 20:1). In in vivo validation, PIA's noninvasive volume fraction estimates matched quantitative histology (ICC, 0.94, 0.85, and 0.92 for epithelium, stroma, and lumen compartments, respectively; P < .001 for all). PIA's measurements strongly correlated with PCa aggressiveness (r = 0.75, P < .001). Furthermore, PIA ran 10 000 faster than NLLS (0.18 second vs 40 minutes per image). Conclusion: PIA provided accurate prostate tissue biomarker measurements from MRI data with better robustness to noise and computational efficiency compared with the NLLS algorithm. The results demonstrate the potential of PIA as an accurate, noninvasive, and explainable artificial intelligence method for PCa detection.

Despite the growing interest in human-AI decision making, experimental studies with domain experts remain rare, largely due to the complexity of working with domain experts and the challenges in setting up realistic experiments. In this work, we conduct an in-depth collaboration with radiologists in prostate cancer diagnosis based on MRI images. Building on existing tools for teaching prostate cancer diagnosis, we develop an interface and conduct two experiments to study how AI assistance and performance feedback shape the decision making of domain experts. In Study 1, clinicians were asked to provide an initial diagnosis (human), then view the AI's prediction, and subsequently finalize their decision (human-AI team). In Study 2 (after a memory wash-out period), the same participants first received aggregated performance statistics from Study 1, specifically their own performance, the AI's performance, and their human-AI team performance, and then directly viewed the AI's prediction before making their diagnosis (i.e., no independent initial diagnosis). These two workflows represent realistic ways that clinical AI tools might be used in practice, where the second study simulates a scenario where doctors can adjust their reliance and trust on AI based on prior performance feedback. Our findings show that, while human-AI teams consistently outperform humans alone, they still underperform the AI due to under-reliance, similar to prior studies with crowdworkers. Providing clinicians with performance feedback did not significantly improve the performance of human-AI teams, although showing AI decisions in advance nudges people to follow AI more. Meanwhile, we observe that the ensemble of human-AI teams can outperform AI alone, suggesting promising directions for human-AI collaboration.

Objectives: The primary objective of this study was to identify whether patients with prostate cancer (PCa) could progress to denervation-resistant prostate cancer (CRPC) after 12 months of hormone therapy. Methods: A total of 96 PCa patients with baseline clinical data who underwent multiparametric magnetic resonance imaging (MRI) between September 2018 and September 2022 were included in this retrospective study. Patients were classified as progressing or not progressing to CRPC on the basis of their outcome after 12 months of hormone therapy. A dense multimodal fusion artificial intelligence (Dense-MFAI) model was constructed by incorporating a squeeze-and-excitation block and a spatial pyramid pooling layer into a dense convolutional network (DenseNet), as well as integrating the eXtreme Gradient Boosting machine learning algorithm. The accuracy, sensitivity, specificity, positive predictive value, negative predictive value, receiver operating characteristic curves, area under the curve (AUC) and confusion matrices were used as classification performance metrics. Results: The Dense-MFAI model demonstrated an accuracy of 94.2%, with an AUC of 0.945, when predicting the progression of patients with PCa to CRPC after 12 months of hormone therapy. The experimental validation demonstrated that combining radiomics feature mapping with baseline clinical characteristics significantly improved the model’s classification performance, confirming the importance of multimodal data. Conclusions: The Dense-MFAI model proposed in this study has the ability to more accurately predict whether a PCa patient could progress to CRPC. This model can assist urologists in developing the most appropriate treatment plan and prognostic measures.

Accurate segmentation of various regions within the prostate is pivotal for diagnosing and treating prostate-related diseases. However, the scarcity of labeled data, particularly in specialized medical fields like prostate imaging, poses a significant challenge. Segment Anything Model (SAM) is a new large model for natural image segmentation, but there are some challenges in medical imaging. In order to better utilize the powerful feature extraction capability of SAM as well as to address the problem of low data volume for medical image annotation, we use Low-Rank Adaptation (LoRA) and semi-supervised learning methods of Monte Carlo guided interpolation consistency (MCIC) to enhance the fine-tuned SAM. We propose Monte Carlo-guided Interpolation Consistency Segment Anything Model (MCICSAM) for application to semi-supervised learning based prostate region segmentation. In the unlabeled data section, MCIC performs two different interpolation transformations on the input data and incorporates Monte Carlo uncertainty analysis in the output, forcing the model to be consistent in its predictions. The consistency constraints imposed on these interpolated samples allow the model to fit the distribution of unlabeled data better, ultimately improving its performance in semi-supervised scenarios. We use Dice and Hausdorff Distance at 95th percentile (HD95) to validate model performance. MCICSAM yieldes Dice with 79.38 HD95 values of 3.12 and 2.27 for transition zone and transition zone. At the same time MCICSAM demonstrates strong generalizability. This method is expected to bring new possibilities in the field of prostate image segmentation.

A new approach to analysis of prostate hybrid multidimensional MRI (HM-MRI) data was introduced in this study. HM-MRI data were acquired for a combination of a few echo times (TEs) and a few b-values. Naturally, there is a matrix associated with HM-MRI data for each image pixel. To process the data, we first linearized HM-MRI data by taking the natural logarithm of the imaging signal intensity. Subsequently, a hybrid symmetric matrix was constructed by multiplying the matrix for each pixel by its own transpose. The eigenvalues for each pixel could then be calculated from the hybrid symmetric matrix. In order to compare eigenvalues between patients, three b-values and three TEs were used, because this was smallest number of b-values and TEs among all patients. The results of eigenvalues were displayed as qualitative color maps for easier visualization. For quantitative analysis, the ratio (λr) of eigenvalues (λ1, λ2, λ3) was defined as λr = (λ1/λ2)/λ3 to compare region of interest (ROI) between prostate cancer (PCa) and normal tissue. The results show that the combined eigenvalue maps show PCas clearly and these maps are quite different from apparent diffusion coefficient (ADC) and T2 maps of the same prostate. The PCa has significant larger λr, smaller ADC and smaller T2 values than normal prostate tissue (p < 0.001). This suggests that the matrix-based method for analyzing HM-MRI data provides new information that may be clinically useful. The method is easy to use and could be easily implemented in clinical practice. The eigenvalues are associated with combination of ADC and T2 values, and could aid in the identification and staging of PCa.

Prostate MRI has traditionally relied on qualitative interpretation. However, quantitative components hold the potential to markedly improve performance. The ADC from DWI is probably the most widely recognized quantitative MRI biomarker and has shown strong discriminatory value for clinically significant prostate cancer as well as for recurrent cancer after treatment. Advanced diffusion techniques, including intravoxel incoherent motion imaging, diffusion kurtosis imaging, diffusion-tensor imaging, and specific implementations such as restriction spectrum imaging, purport even better discrimination but are more technically challenging. The inherent T1 and T2 of tissue also provide diagnostic value, with more advanced techniques deriving luminal water fraction and hybrid multidimensional MRI metrics. Dynamic contrast-enhanced imaging, primarily using a modified Tofts model, also shows independent discriminatory value. Finally, quantitative lesion size and shape features can be combined with the aforementioned techniques and can be further refined using radiomics, texture analysis, and artificial intelligence. Which technique will ultimately find widespread clinical use will depend on validation across a myriad of platforms and use cases.

Background:High b -value diffusion-weighted images (DWI) are used for detection of clinically significant prostate cancer (csPCa). To decrease scan time and improve signal-to-noise ratio, high b -value (>1000 s/mm 2 ) images are often synthesized instead of acquired.Purpose:Qualitatively and quantitatively compare synthesized DWI (sDWI) to acquired (aDWI) for detection of csPCa.Study Type:Retrospective.Subjects:151 consecutive patients who underwent prostate MRI and biopsy.Sequence:Axial DWI with b =0, 500, 1000, and 2000 s/mm 2 using a 3T clinical scanner using a 32-channel phased-array body coil.Assessment:We synthesized DWI for b =2000 s/mm 2 via extrapolation based on monoexponential decay, using b =0 and b =500 s/mm 2 (sDWI 500 ) and b =0, b =500, and b =1000 s/mm 2 (sDWI 1000 ). Differences between sDWI and aDWI were evaluated within regions of interest (ROIs). The maximum DWI value within each ROI was evaluated for prediction of csPCa. Classification accuracy was also compared to Restriction Spectrum Imaging restriction score (RSIrs), a previously validated biomarker based on multi-exponential DWI.Statistical Tests:Discrimination of csPCa was evaluated via area under the receiver operating characteristic curve (AUC). Statistical significance was assessed using bootstrap difference (two-sided α=0.05).Results:Within the prostate, mean ± standard deviation of percent mean differences between sDWI and aDWI signal were -46±35% for sDWI 1000 and -67±24% for sDWI 500 . AUC for aDWI, sDWI 500, sDWI 1000 , and RSIrs within the prostate 0.62[95% confidence interval: 0.53, 0.71], 0.63[0.54, 0.72], 0.65[0.56, 0.73] and 0.78[0.71, 0.86], respectively. When considering the whole field of view, classification accuracy and qualitative image quality decreased notably for sDWI compared to aDWI and RSIrs.Data Conclusion:sDWI is qualitatively comparable to aDWI within the prostate. However, hyperintense artifacts are introduced with sDWI in the surrounding pelvic tissue that interfere with quantitative cancer detection and might mask metastases. In the prostate, RSIrs yields superior quantitative csPCa detection than sDWI or aDWI.

Purpose: This study investigates whether quantitative MRI and histology of the prostate reveal differences between races, specifically African Americans (AAs) and Caucasian Americans (CAs), that can affect diagnosis. Materials and Methods: Patients (98 CAs, 47 AAs) with known or suspected prostate cancer (PCa) underwent 3T MRI (T2W, DWI, and DCE-MRI) prior to biopsy or prostatectomy. Quantitative mpMRI metrics: ADC, T2, and DCE empirical mathematical model parameters were calculated. Results: AAs had a greater percentage of higher Gleason-grade lesions compared to CAs. There were no significant differences in the quantitative ADC and T2 values between AAs and CAs. The cancer signal enhancement rate (α) on DCE-MRI was significantly higher for AAs compared to CAs (AAs: 13.3 ± 9.3 vs. CAs: 6.1 ± 4.7 s−1, p < 0.001). The DCE signal washout rate (β) was significantly lower in benign tissue of AAs (AAs: 0.01 ± 0.09 s−1 vs. CAs: 0.07 ± 0.07 s−1, p < 0.001) and significantly elevated in cancer tissue in AAs (AAs: 0.12 ± 0.07 s−1 vs. CAs: 0.07 ± 0.08 s−1, p = 0.02). DCE significantly improves the differentiation of PCa from benign in AAs (α: 52%, β: 62% more effective in AAs compared to CAs). Histologic analysis showed cancers have a greater proportion (p = 0.04) of epithelium (50.9 ± 12.3 vs. 44.7 ± 12.8%) and lower lumen (10.5 ± 6.9 vs. 16.2 ± 6.8%) in CAs compared to AAs. Conclusions: This study shows that AAs have different quantitative DCE-MRI values for benign prostate and prostate cancer and different histologic makeup in PCa compared to CAs. Quantitative DCE-MRI can significantly improve the performance of MRI for PCa diagnosis in African Americans but is much less effective for Caucasian Americans.

PURPOSE:This study addresses the challenge of low resolution and signal-to-noise ratio (SNR) in diffusion-weighted images (DWI), which are pivotal for cancer detection. Traditional methods increase SNR at high b-values through multiple acquisitions, but this results in diminished image resolution due to motion-induced variations. Our research aims to enhance spatial resolution by exploiting the global structure within multicontrast DWI scans and millimetric motion between acquisitions. METHODS:We introduce a novel approach employing a "Perturbation Network" to learn subvoxel-size motions between scans, trained jointly with an implicit neural representation (INR) network. INR encodes the DWI as a continuous volumetric function, treating voxel intensities of low-resolution acquisitions as discrete samples. By evaluating this function with a finer grid, our model predicts higher-resolution signal intensities for intermediate voxel locations. The Perturbation Network's motion-correction efficacy was validated through experiments on biological phantoms and in vivo prostate scans. RESULTS:Quantitative analyses revealed significantly higher structural similarity measures of super-resolution images to ground truth high-resolution images compared to high-order interpolation (p < $$ < $$ 0.005). In blind qualitative experiments, 96 . 1 % $$ 96.1\% $$ of super-resolution images were assessed to have superior diagnostic quality compared to interpolated images. CONCLUSION:High-resolution details in DWI can be obtained without the need for high-resolution training data. One notable advantage of the proposed method is that it does not require a super-resolution training set. This is important in clinical practice because the proposed method can easily be adapted to images with different scanner settings or body parts, whereas the supervised methods do not offer such an option.

1. In a retrospective review of 61 men by four readers, two inexperienced-readers had higher accuracy using multiparametric MRI+Hybrid-Multidimensional MRI (mpMRI+HM-MRI)=82%,81% versus Multiparametric MRI=77%, 71% (p=.006,<.001) and higher specificity=89%,88% versus 84%,75% (p=.009, p<.001) on a per-sextant analysis. 2. Multiparametric MRI+Hybrid-Multidimensional MRI increased trainee specificity=46% versus Multiparametric MRI=7% (p<.001) on a per-patient analysis. 3. Multiparametric MRI+Hybrid-Multidimensional MRI increased interreader Kappa agreement=0.36 versus Multiparametric MRI (mpMRI) Kappa=0.17 (p=0.009) in diagnosing clinically significant prostate cancer (CS PCa).

Physics model-based approaches using Magnetic Resonance Imaging (MRI) have been developed to provide a non-invasive tool to estimate tissue composition by fitting the notoriously ill-conditioned sum-of-decaying-exponentials functions. However, even modest levels of noise can negatively impact the accuracy of such models. This study introduces a new paradigm that integrates the strengths of physics model-based and deep learning-based methods, while overcoming their respective weaknesses. We reformulated the MRI-based tissue composition estimation using a physics-informed autoencoder (PIA), transforming the problem from a least-squares fit into a robust deep learning framework. The PIA model consists of a trainable multi-head neural network encoder and a fixed, physics-informed decoder. Without a need for further fine-tuning for each new patient, the encoder's latent activations are utilized to derive prostate tissue composition estimates, such as volume fractions for each compartment along with their diffusivity and T2. PIA's estimates of tissue parameters are juxtaposed with the least-squares solution through both Monte Carlo simulations and the quantitative histology of in- vivo prostate scans. In-vivo evaluations, performed on 19 pathology-proven prostate cancer patients, have been validated with quantitative histology-based true tissue compositions. Both Monte Carlo and in-vivo evaluations reveal that PIA significantly outperforms the conventional least-squares solution on Pearson correlation and mean absolute error metrics. Notably, PIA was able to predict the diffusivity and T2 of each tissue compartment more accurately, whereas the least-squares method tends to pin these values to the extremities.

PURPOSE:The interpretation of prostate multiparametric magnetic resonance imaging (MRI) is subjective in nature, and there is large inter-observer variability among radiologists and up to 30% of clinically significant cancers are missed. This has motivated the development of new MRI techniques and sequences, especially quantitative approaches to improve prostate cancer diagnosis. Using hybrid multidimensional MRI, apparent diffusion coefficient (ADC) and T2 have been shown to change as a function of echo time (TE) and b-values, and that this dependence is different for cancer and benign tissue, which can be exploited for prostate cancer diagnosis. The purpose of this study is to investigate whether four-quadrant vector mapping of hybrid multidimensional MRI (HM-MRI) data can be used to diagnose prostate cancer (PCa) and determine cancer aggressiveness. METHODS:Twenty-one patients with confirmed PCa underwent preoperative MRI prior to radical prostatectomy. Axial HM-MRI were acquired with all combinations of TE = 47, 75, 100 ms and b-values of 0, 750, 1500 s/mm2 , resulting in a 3 × 3 data matrix associated with each voxel. Prostate Quadrant (PQ) mapping analysis represents HM-MRI data for each voxel as a color-coded vector in the four-quadrant space of HM-MRI parameters (a 2D matrix of signal values for each combination of b-value and TE) with associated amplitude and angle information representing the change in T2 and ADC as a function of b-value and TE, respectively. RESULTS:Cancers have a higher PQ4 (22.50% ± 21.27%) and lower PQ2 (69.86% ± 28.24%) compared to benign tissue: peripheral, transition, and central zone (PQ4 = 0.13% ± 0.56%, 5.73% ± 15.07%, 2.66% ± 4.05%, and PQ2 = 98.51% ± 3.05%, 86.18% ± 21.75%, 93.38% ± 9.88%, respectively). Cancers have a higher vector angle (206.5 ± 41.8°) and amplitude (0.017 ± 0.013) compared to benign tissue. PQ metrics showed moderate correlation with Gleason score (|ρ| = 0.388-0.609), with more aggressive cancers being associated with increased PQ4 and angle and reduced PQ2 and amplitude. A combination of four-quadrant analysis metrics provided an area under the curve of 0.904 (p < 0.001) for the differentiation of prostate cancer from benign prostatic tissue. CONCLUSIONS:Four-quadrant vector mapping of HM-MRI data provides effective cancer markers, with cancers associated with high PQ4 and high vector angle and lower PQ2 and vector amplitude.

Prostate cancer poses significant diagnostic challenges, with conventional methods like prostate-specific antigen (PSA) screening and transrectal ultrasound (TRUS)-guided biopsies often leading to overdiagnosis or miss clinically significant cancers. Multiparametric MRI (mpMRI) has emerged as a more reliable tool. However, it is limited by high inter-observer variability and radiologists missing up to 30% of clinically significant cancers. This article summarizes a few of these recent advancements in quantitative MRI techniques that look at the "Virtual Pathology" of the prostate with an aim to enhance prostate cancer detection and characterization. These techniques include T2 relaxation-based techniques such as luminal water imaging, diffusion based such as vascular, extracellular, and restricted diffusion for cytometry in tumors (VERDICT) and restriction spectrum imaging or combined relaxation-diffusion techniques such as hybrid multi-dimensional MRI (HM-MRI), time-dependent diffusion imaging, and diffusion-relaxation correlation spectrum imaging. These methods provide detailed insights into underlying prostate microstructure and tissue composition and have shown improved diagnostic accuracy over conventional MRI. These innovative MRI methods hold potential for augmenting mpMRI, reducing variability in diagnosis, and paving the way for MRI as a 'virtual histology' tool in prostate cancer diagnosis. However, they require further validation in larger multi-center clinical settings and rigorous in-depth radiological-pathology correlation are needed for broader implementation.

High b-value diffusion-weighted imaging (DWI) plays an important role for accurate detection of clinically significant prostate cancer (csPCa). Synthesizing high b-value images decreases scan time. We compared (normalized) acquired and synthesized high b-value (2000s/mm2) DWI for detection of csPCa in 151 patients who underwent MRI and biopsy. We also compared these to an advanced DWI biomarker called the Restriction Spectrum Imaging restriction score (RSIrs). Synthesized images yielded similar results to acquired images within the prostate but failed to accurately represent the surrounding pelvic tissue. RSIrs was superior to synthesized and acquired high b-value DWI for detection of csPCa.

We propose a general method for combining multiple models to predict tissue microstructure, with an exemplar using in vivo diffusion-relaxation MRI data. The proposed method obviates the need to select a single ’optimum’ structure model for data analysis in heterogeneous tissues where the best model varies according to local environment. We break signal interpretation into a three-stage sequence: (1) application of multiple semi-phenomenological models to predict the physical properties of tissue water pools contributing to the observed signal; (2) from each Stage-1 semi-phenomenological model, application of a tissue microstructure model to predict the relative volumes of tissue structure components that make up each water pool; and (3) aggregation of the predictions of tissue structure, with weightings based on model likelihood and fractional volumes of the water pools from Stage-1. The multiple model approach is expected to reduce prediction variance in tissue regions where a complex model is overparameterised, and bias where a model is underparameterised. The separation of signal characterisation (Stage-1) from biological assignment (Stage-2) enables alternative biological interpretations of the observed physical properties of the system, by application of different tissue structure models. The proposed method is exemplified with human prostate diffusion-relaxation MRI data, but has potential application to a wide range of analyses where a single model may not be optimal throughout the sampled domain.

The spatial two-tissue compartment model (2TCM) was used to analyze prostate dynamic contrast enhanced (DCE) MRI data and compared with the standard Tofts model. A total of 29 patients with biopsy-confirmed prostate cancer were included in this IRB-approved study. MRI data were acquired on a Philips Achieva 3T-TX scanner. After T2-weighted and diffusion-weighted imaging, DCE data using 3D T1-FFE mDIXON sequence were acquired pre- and post-contrast media injection (0.1 mmol/kg Multihance) for 60 dynamic scans with temporal resolution of 8.3 s/image. The 2TCM has one fast (K 1trans and k 1ep ) and one slow (K 2trans and k 2ep ) exchanging compartment, compared with the standard Tofts model parameters (K trans and k ep ). On average, prostate cancer had significantly higher values (p < 0.007) than normal prostate tissue for all calculated parameters. There was a strong correlation (r = 0.94, p < 0.0001) between K trans and K 1trans for cancer, but weak correlation (r = 0.28, p < 0.05) between k ep and k 1ep . Average root-mean-square error (RMSE) in fits from the 2TCM was significantly smaller (p < 0.001) than the RMSE in fits from the Tofts model. Receiver operating characteristic (ROC) analysis showed that fast K 1trans had the highest area under the curve (AUC) than any other individual parameter. The combined four parameters from the 2TCM had a considerably higher AUC value than the combined two parameters from the Tofts model. The 2TCM may be useful for quantitative analysis of prostate DCE-MRI data and may provide new information in the diagnosis of prostate cancer.