Algorithms applied to spatially registered multi-parametric MRI for prostate tumor volume measurement

Autonomous Tumor Signature Extraction Applied to Spatially Registered Bi-Parametric MRI to Predict Prostate Tumor Aggressiveness: A Pilot Study

BACKGROUND

Accurate, reliable, non-invasive assessment of patients diagnosed with prostate cancer is essential for proper disease management. Quantitative assessment of multi-parametric MRI, such as through artificial intelligence or spectral/statistical approaches, can provide a non-invasive objective determination of the prostate tumor aggressiveness without side effects or potential poor sampling from needle biopsy or overdiagnosis from prostate serum antigen measurements. To simplify and expedite prostate tumor evaluation, this study examined the efficacy of autonomously extracting tumor spectral signatures for spectral/statistical algorithms for spatially registered bi-parametric MRI.

METHODS

Spatially registered hypercubes were digitally constructed by resizing, translating, and cropping from the image sequences (Apparent Diffusion Coefficient (ADC), High B-value, T2) from 42 consecutive patients in the bi-parametric MRI PI-CAI dataset. Prostate cancer blobs exceeded a threshold applied to the registered set from normalizing the registered set into an image that maximizes High B-value, but minimizes the ADC and T2 images, appearing "green" in the color composite. Clinically significant blobs were selected based on size, average normalized green value, sliding window statistics within a blob, and position within the hypercube. The center of mass and maximized sliding window statistics within the blobs identified voxels associated with tumor signatures. We used correlation coefficients (R) and p-values, to evaluate the linear regression fits of the z-score and SCR (with processed covariance matrix) to tumor aggressiveness, as well as Area Under the Curves (AUC) for Receiver Operator Curves (ROC) from logistic probability fits to clinically significant prostate cancer.

RESULTS

The highest R (R > 0.45), AUC (>0.90), and lowest p-values (<0.01) were achieved using z-score and modified registration applied to the covariance matrix and tumor signatures selected from the "greenest" parts from the selected blob.

CONCLUSIONS

The first autonomous tumor signature applied to spatially registered bi-parametric MRI shows promise for determining prostate tumor aggressiveness.

Using an artificial intelligence model to detect and localize visible clinically significant prostate cancer in prostate magnetic resonance imaging: a multicenter external validation study

Background

An increasing number of patients with suspected clinically significant prostate cancer (csPCa) are undergoing prostate multiparametric magnetic resonance imaging (mpMRI). The role of artificial intelligence (AI) algorithms in interpreting prostate mpMRI needs to be tested with multicenter external data. This study aimed to investigate the diagnostic efficacy of an AI model in detecting and localizing visible csPCa on mpMRI a multicenter external data set.

Methods

The data of 2,105 patients suspected of having prostate cancer from four hospitals were retrospectively collected to develop an AI model to detect and localize suspicious csPCa. The lesions were annotated based on pathology records by two radiologists. Diffusion-weighted imaging (DWI) and apparent diffusion coefficient (ADC) values were used as the input for the three-dimensional U-Net framework. Subsequently, the model was validated using an external data set comprising the data of 557 patients from three hospitals. Sensitivity, specificity, and accuracy were employed to evaluate the diagnostic efficacy of the model.

Results

At the lesion level, the model had a sensitivity of 0.654. At the overall sextant level, the model had a sensitivity, specificity, and accuracy of 0.846, 0.884, and 0.874, respectively. At the patient level, the model had a sensitivity, specificity, and accuracy of 0.943, 0.776, and 0.849, respectively. The AI-predicted accuracy for the csPCa patients (231/245, 0.943) was significantly higher than that for the non-csPCa patients (242/312, 0.776) (P<0.001). The lesion number and tumor volume were greater in the correctly diagnosed patients than the incorrectly diagnosed patients (both P<0.001). Among the positive patients, those with lower average ADC values had a higher rate of correct diagnosis than those with higher average ADC values (P=0.01).

Conclusions

The AI model exhibited acceptable accuracy in detecting and localizing visible csPCa at the patient and sextant levels. However, further improvements need to be made to enhance the sensitivity of the model at the lesion level.

Relationship between Eccentricity and Volume Determined by Spectral Algorithms Applied to Spatially Registered Bi-Parametric MRI and Prostate Tumor Aggressiveness: A Pilot Study

(1) Background: Non-invasive prostate cancer assessments using multi-parametric MRI are essential to the reliable detection of lesions and proper management of patients. While current guidelines call for the administration of Gadolinium-containing intravenous contrast injections, eliminating such injections would simplify scanning and reduce patient risk and costs. However, augmented image analysis is necessary to extract important diagnostic information from MRIs. Purpose: This study aims to extend previous work on the signal to clutter ratio and test whether prostate tumor eccentricity and volume are indicators of tumor aggressiveness using bi-parametric (BP)-MRI. (2) Methods: This study retrospectively processed 42 consecutive prostate cancer patients from the PI-CAI data collection. BP-MRIs (apparent diffusion coefficient, high b-value, and T2 images) were resized, translated, cropped, and stitched to form spatially registered BP-MRIs. The International Society of Urological Pathology (ISUP) grade was used to judge cases of prostate cancer as either clinically significant prostate cancer (CsPCa) (ISUP ≥ 2) or clinically insignificant prostate cancer (CiPCa) (ISUP < 2). The Adaptive Cosine Estimator (ACE) algorithm was applied to the BP-MRIs, followed by thresholding, and then eccentricity and volume computations, from the labeled and blobbed detection maps. Then, univariate and multivariate linear regression fittings of eccentricity and volume were applied to the ISUP grade. The fits were quantitatively evaluated by computing correlation coefficients (R) and p-values. Area under the curve (AUC) and receiver operator characteristic (ROC) curve scores were used to assess the logistic fitting to CsPCa/CiPCa. (3) Results: Modest correlation coefficients (R) (>0.35) and AUC scores (0.70) for the linear and/or logistic fits from the processed prostate tumor eccentricity and volume computations for the spatially registered BP-MRIs exceeded fits using the parameters of prostate serum antigen, prostate volume, and patient age (R~0.17). (4) Conclusions: This is the first study that applied spectral approaches to BP-MRIs to generate tumor eccentricity and volume metrics to assess tumor aggressiveness. This study found significant values of R and AUC (albeit below those from multi-parametric MRI) to fit and relate the metrics to the ISUP grade and CsPCA/CiPCA, respectively.

Pilot study for generating and assessing nomograms and decision curves analysis to predict clinically significant prostate cancer using only spatially registered multi-parametric MRI

Background

Current prostate cancer evaluation can be inaccurate and burdensome. To help non-invasive prostate tumor assessment, recent algorithms applied to spatially registered multi-parametric (SRMP) MRI extracted novel clinically relevant metrics, namely the tumor's eccentricity (shape), signal-to-clutter ratio (SCR), and volume.

Purpose

Conduct a pilot study to predict the risk of developing clinically significant prostate cancer using nomograms and employing Decision Curves Analysis (DCA) from the SRMP MRI-based features to help clinicians non-invasively manage prostate cancer.

Methods

This study retrospectively analyzed 25 prostate cancer patients. MP-MRI (T1, T2, diffusion, dynamic contrast-enhanced) were resized, translated, and stitched to form SRMP MRI. Target detection algorithm [adaptive cosine estimator (ACE)] applied to SRMP MRI determines tumor's eccentricity, noise reduced SCR (by regularizing or eliminating principal components (PC) from the covariance matrix), and volume. Pathology assessed wholemount prostatectomy for Gleason score (GS). Tumors with GS >=4+3 (<=3+4) were judged as "Clinically Significant" ("Insignificant"). Logistic regression combined eccentricity, SCR, volume to generate probability distribution. Nomograms, DCA used all patients plus training (13 patients) and test (12 patients) sets. Area Under the Curves for (AUC) for Receiver Operator Curves (ROC) and p-values evaluated the performance.

Results

Combining eccentricity (0.45 ACE threshold), SCR (3, 4 PCs), SCR (regularized, modified regularization) with tumor volume (0.65 ACE threshold) improved AUC (>0.70) for ROC curves and p-values (<0.05) for logistic fit. DCA showed greater net benefit from model fit than univariate analysis, treating "all," or "none." Training/test sets achieved comparable AUC but with higher p-values.

Conclusions

Performance of nomograms and DCA based on metrics derived from SRMP-MRI in this pilot study were comparable to those using prostate serum antigen, age, and PI-RADS.

Assessing and testing anomaly detection for finding prostate cancer in spatially registered multi-parametric MRI

Background

Evaluating and displaying prostate cancer through non-invasive imagery such as Multi-Parametric MRI (MP-MRI) bolsters management of patients. Recent research quantitatively applied supervised target algorithms using vectoral tumor signatures to spatially registered T1, T2, Diffusion, and Dynamic Contrast Enhancement images. This is the first study to apply the Reed-Xiaoli (RX) multi-spectral anomaly detector (unsupervised target detector) to prostate cancer, which searches for voxels that depart from the background normal tissue, and detects aberrant voxels, presumably tumors.

Methods

MP-MRI (T1, T2, diffusion, dynamic contrast-enhanced images, or seven components) were prospectively collected from 26 patients and then resized, translated, and stitched to form spatially registered multi-parametric cubes. The covariance matrix (CM) and mean μ were computed from background normal tissue. For RX, noise was reduced for the CM by filtering out principal components (PC), regularization, and elliptical envelope minimization. The RX images were compared to images derived from the threshold Adaptive Cosine Estimator (ACE) and quantitative color analysis. Receiver Operator Characteristic (ROC) curves were used for RX and reference images. To quantitatively assess algorithm performance, the Area Under the Curve (AUC) and the Youden Index (YI) points for the ROC curves were computed.

Results

The patient average for the AUC and [YI] from ROC curves for RX from filtering 3 and 4 PC was 0.734[0.706] and 0.727[0.703], respectively, relative to the ACE images. The AUC[YI] for RX from modified Regularization was 0.638[0.639], Regularization 0.716[0.690], elliptical envelope minimization 0.544[0.597], and unprocessed CM 0.581[0.608] using the ACE images as Reference Image. The AUC[YI] for RX from filtering 3 and 4 PC was 0.742[0.711] and 0.740[0.708], respectively, relative to the quantitative color images. The AUC[YI] for RX from modified Regularization was 0.643[0.648], Regularization 0.722[0.695], elliptical envelope minimization 0.508[0.605], and unprocessed CM 0.569[0.615] using the color images as Reference Image. All standard errors were less than 0.020.

Conclusions

This first study of spatially registered MP-MRI applied anomaly detection using RX, an unsupervised target detection algorithm for prostate cancer. For RX, filtering out PC and applying Regularization achieved higher AUC and YI using ACE and color images as references than unprocessed CM, modified Regularization, and elliptical envelope minimization.

Combining and analyzing novel multi-parametric magnetic resonance imaging metrics for predicting Gleason score

BACKGROUND

Radiologists currently subjectively examine multi-parametric magnetic resonance imaging (MP-MRI) to determine prostate tumor aggressiveness using the Prostate Imaging Reporting and Data System scoring system (PI-RADS). Recent studies showed that modified signal to clutter ratio (SCR), tumor volume, and eccentricity (elongation or roundness) of prostate tumors correlated with Gleason score (GS). No previous studies have combined the prostate tumor's shape, SCR, tumor volume, in order to predict potential tumor aggressiveness and GS.

METHODS

MP-MRI (T1, T2, diffusion, dynamic contrast-enhanced images) were obtained, resized, translated, and stitched to form spatially registered multi-parametric cubes. Multi-parametric signatures that characterize prostate tumors were inserted into a target detection algorithm [adaptive cosine estimator (ACE)]. Pixel-based blobbing, and labeling were applied to the threshold ACE images. Eccentricity calculation used moments of inertia from the blobs. Tumor volume was computed by counting pixels within multi parametric MRI blobs and tumor outlines based on pathologist assessment of whole mount histology. Pathology assessment of GS was performed on whole mount prostatectomy. The covariance matrix and mean of normal tissue background was computed from normal prostate. Using signatures and normal tissue statistics, the z-score, noise corrected SCR [principal component (PC), modified regularization] from each patient was computed. Eccentricity, tumor volume, and SCR were fitted to GS. Analysis of variance assesses the relationship among the variables.

RESULTS

A multivariate analysis generated correlation coefficient (0.60 to 0.784) and P value (0.00741 to <0.0001) from fitting two sets of independent variates, namely, tumor eccentricity (the eccentricity for the largest blob, weighted average for the eccentricity) and SCR (removing 3 PCs, removing 4 PCs, modified regularization, and z-score) to GS. The eccentricity t-statistic exceeded the SCR t-statistic. The three-variable fit to GS using tumor volume (histology, MRI) yielded correlation coefficients ranging from 0.724 to 0.819 (P value <<0.05). Tumor volumes generated from histology yielded higher correlation coefficients than MRI volumes. Adding volume to eccentricity and SCR adds little improvement for fitting GS due to higher correlation coefficients among independent variables and little additional, independent information.

CONCLUSIONS

Combining prostate tumors eccentricity with SCR relatively highly correlates with GS.

Development and testing quantitative metrics from multi-parametric magnetic resonance imaging that predict Gleason score for prostate tumors

BACKGROUND

Radiologists currently subjectively examine multi-parametric magnetic resonance imaging (MRI) to detect possible clinically significant lesions using the Prostate Imaging Reporting and Data System (PI-RADS) protocol. The assessment of imaging, however, relies on the experience and judgement of radiologists creating opportunity for inter-reader variability. Quantitative metrics, such as z-score and signal to clutter ratio (SCR), are therefore needed.

METHODS

Multi-parametric MRI (T1, T2, diffusion, dynamic contrast-enhanced images) were resampled, rescaled, translated, and stitched to form spatially registered multi-parametric cubes for patients undergoing radical prostatectomy. Multi-parametric signatures that characterize prostate tumors were inserted into z-score and SCR. The multispectral covariance matrix was computed for the outlined normal prostate. The z-score from each MRI image was computed and summed. To reduce noise in the covariance matrix, following matrix decomposition, the noisy eigenvectors were removed. Also, regularization and modified regularization was applied to the covariance matrix by minimizing the discrimination score. The filtered and regularized covariance matrices were inserted into the SCR calculation. The z-score and SCR were quantitatively compared to Gleason scores from clinical pathology assessment of the histology of sectioned wholemount prostates.

RESULTS

Twenty-six consecutive patients were enrolled in this retrospective study. Median patient age was 60 years (range, 49 to 75 years), median prostate-specific antigen (PSA) was 5.8 ng/mL (range, 2.3 to 23.7 ng/mL), and median Gleason score was 7 (range, 6 to 9). A linear fit of the summed z-score against Gleason score found a correlation of R=0.48 and a P value of 0.015. A linear fit of the SCR from regularizing covariance matrix against Gleason score found a correlation of R=0.39 and a P value of 0.058. The SCR employing the modified regularizing covariance matrix against Gleason score found a correlation of R=0.52 and a P value of 0.007. A linear fit of the SCR from filtering out 3 and 4 eigenvectors from the covariance matrix against Gleason score found correlations of R=0.50 and 0.44, respectively, and P values of 0.011 and 0.027, respectively.

CONCLUSIONS

Z-score and SCR using filtered and regularized covariance matrices derived from spatially registered multi-parametric MRI correlates with Gleason score with highly significant P values.

Prostate tumor eccentricity predicts Gleason score better than prostate tumor volume

Background

Prostate tumor volume predicts biochemical recurrence, metastases, and tumor proliferation. A recent study showed that prostate tumor eccentricity (elongation or roundness) correlated with Gleason score. No studies examined the relationship among the prostate tumor's shape, volume, and potential aggressiveness.

Methods

Of the 26 patients that were analyzed, 18 had volumes >1 cc for the histology-based study, and 25 took up contrast material for the MRI portion of this study. This retrospective study quantitatively compared tumor eccentricity and volume measurements from pathology assessment sectioned wholemount prostates and multi-parametric MRI to Gleason scores. Multi-parametric MRI (T1, T2, diffusion, dynamic contrast-enhanced images) were resized, translated, and stitched to form spatially registered multi-parametric cubes. Multi-parametric signatures that characterize prostate tumors were inserted into a target detection algorithm (Adaptive Cosine Estimator, ACE). Various detection thresholds were applied to discriminate tumor from normal tissue. Pixel-based blobbing, and labeling were applied to digitized pathology slides and threshold ACE images. Tumor volumes were measured by counting voxels within the blob. Eccentricity calculation used moments of inertia from the blobs.

Results

From wholemount prostatectomy slides, fitting two sets of independent variables, prostate tumor eccentricity (largest blob eccentricity, weighted eccentricity, filtered weighted eccentricity) and tumor volume (largest blob volume, average blob volume, filtered average blob volume) to Gleason score in a multivariate analysis, yields correlation coefficient R=0.798 to 0.879 with P<0.01. The eccentricity t-statistic exceeded the volume t-statistic. Fitting histology-based total prostate tumor volume against Gleason score yields R=0.498, P=0.0098. From multi-parametric MRI, the correlation coefficient R between the Gleason score and the largest blob eccentricity for varying thresholds (0.30 to 0.55) ranged from -0.51 to -0.672 (P<0.01). For varying thresholds (0.60 to 0.80) for MRI detection, the R between the largest blob volume eccentricity against the Gleason score ranged from 0.46 to 0.50 (P<0.03). Combining tumor eccentricity and tumor volume in multivariate analysis failed to increase Gleason score prediction.

Conclusions

Prostate tumor eccentricity, determined by histology or MRI, more accurately predicted Gleason score than prostate tumor volume. Combining tumor eccentricity with volume from histology-based analysis enhanced Gleason score prediction, unlike MRI.

Correlation of prostate tumor eccentricity and Gleason scoring from prostatectomy and multi-parametric-magnetic resonance imaging

Background

Proliferating cancer cells interacting with their microenvironment affects a tumor's spatial shape. Elongation or roundness (eccentricity) of lung, skin, and breast cancers indicates the cancer's relative aggressiveness. Non-invasive determination of the prostate tumor's shape should provide meaningful input for prognostication and clinical management. There are currently few studies of prostate tumor shape, therefore this study examines the relationship between a prostate tumor's eccentricity, derived from spatially registered multi-parametric MRI and histology slides, and Gleason scores.

Methods

A total of 26 consecutive patients were enrolled in the study. Median patient age was 60 years (range, 49 to 75 years), median PSA was 5.8 ng/mL (range, 2.3 to 23.7 ng/mL, and median Gleason score was 7 (range, 6 to 9). Multi-parametric MRI (T1, T2, Diffusion, Dynamic Contrast Enhanced) were resampled, rescaled, translated, and stitched to form spatially registered multi-parametric cubes. Multi-parametric signatures that characterize prostate tumors were inserted into a target detection algorithm (Adaptive Cosine Estimator, ACE). Various detection thresholds were applied to discriminate tumor from normal tissue. Also, tumor shape was computed from the histology slides. Blobbing, labeling, and calculation of eccentricity using moments of inertia were applied to the multi-parametric MRI and histology slides. The eccentricity measurements were compared to the Gleason scores from 25 patients.

Results

From histology slides analysis: the correlation coefficient between the eccentricity for the largest blob and a weighted average eccentricity against the Gleason score ranged from -0.67 to -0.78 for all 18 patients whose tumor volume exceeded 1.0 cc. From multi-parametric MRI analysis: the correlation coefficient between the eccentricity for the largest blob for varying thresholds against the Gleason score ranged from -0.60 to -0.66 for all 25 patients showing contrast uptake in the Dynamic Contrast Enhancement (DCE) MRI.

Conclusions

Spherical shape prostate adenocarcinoma shows a propensity for higher Gleason score. This novel finding follows lung and breast adenocarcinomas but depart from other primary tumor types. Analysis of multi-parametric MRI can non-invasively determine the prostate tumor's morphology and add critical information for prognostication and disease management. Eccentricity of smaller tumors (<1.0 cc) from MP-MRI correlates well with Gleason score, unlike eccentricity measured using histology of wholemount prostatectomy.