AUTHOR=Mayer Rulon , Turkbey Baris , Choyke Peter , Simone Charles B. 

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

JOURNAL=Frontiers in Oncology

VOLUME=Volume 12 - 2022

YEAR=2023

URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2022.1033323

DOI=10.3389/fonc.2022.1033323

ISSN=2234-943X

ABSTRACT=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 an anomaly detector (RX) (unsupervised target detector), which searches for voxels that depart from the background normal tissue, and detect 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 ( 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 the images derived from threshold Adaptive Cosine Estimator (ACE) and quantitative color analysis. Receiver Operator Characteristic (ROC) curves used the RX and reference images. To quantitatively assess algorithm performance, the Area Under the Curve (AUC) and the Maximum Accuracy (MA) point for the ROC curves were computed. 
Results: The patient average for the AUC and [MA] 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[MA] 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 ground truth.  The AUC[MA] 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[MA] 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 ground truth.  All standard errors were less than 0.020.
Conclusions: This first-ever study of spatially registered MP-MRI applied anomaly detection that does not use a tumor signature. For RX, filtering out PC and applied Regularization achieved higher AUC and MA using ACE and color images as references than unprocessed CM, modified Regularization, and elliptical envelope minimization.