Mena Gaed

Prostate Histopathology: Learning Tissue Component Histograms for Cancer Detection and Classification

Radical prostatectomy is performed on approximately 40% of men with organ-confined prostate cancer. Pathologic information obtained from the prostatectomy specimen provides important prognostic information and guides recommendations for adjuvant treatment. The current pathology protocol in most centers involves primarily qualitative assessment. In this paper, we describe and evaluate our system for automatic prostate cancer detection and grading on hematoxylin & eosin-stained tissue images. Our approach is intended to address the dual challenges of large data size and the need for high-level tissue information about the locations and grades of tumors. Our system uses two stages of AdaBoost-based classification. The first provides high-level tissue component labeling of a superpixel image partitioning. The second uses the tissue component labeling to provide a classification of cancer versus noncancer, and low-grade versus high-grade cancer. We evaluated our system using 991 sub-images extracted from digital pathology images of 50 whole-mount tissue sections from 15 prostatectomy patients. We measured accuracies of 90% and 85% for the cancer versus noncancer and high-grade versus low-grade classification tasks, respectively. This system represents a first step toward automated cancer quantification on prostate digital histopathology imaging, which could pave the way for more accurately informed postprostatectomy patient care.

Registration of prostate histology images to ex vivo MR images via strand-shaped fiducials.

To present and evaluate a method for registration of whole‐mount prostate digital histology images to ex vivo magnetic resonance (MR) images.

3D prostate histology image reconstruction: Quantifying the impact of tissue deformation and histology section location

Background: Guidelines for localizing prostate cancer on imaging are ideally informed by registered post-prostatectomy histology. 3D histology reconstruction methods can support this by reintroducing 3D spatial information lost during histology processing. The need to register small, high-grade foci drives a need for high accuracy. Accurate 3D reconstruction method design is impacted by the answers to the following central questions of this work. (1) How does prostate tissue deform during histology processing? (2) What spatial misalignment of the tissue sections is induced by microtome cutting? (3) How does the choice of reconstruction model affect histology reconstruction accuracy? Materials and Methods: Histology, paraffin block face and magnetic resonance images were acquired for 18 whole mid-gland tissue slices from six prostates. 7-15 homologous landmarks were identified on each image. Tissue deformation due to histology processing was characterized using the target registration error (TRE) after landmark-based registration under four deformation models (rigid, similarity, affine and thin-plate-spline [TPS]). The misalignment of histology sections from the front faces of tissue slices was quantified using manually identified landmarks. The impact of reconstruction models on the TRE after landmark-based reconstruction was measured under eight reconstruction models comprising one of four deformation models with and without constraining histology images to the tissue slice front faces. Results: Isotropic scaling improved the mean TRE by 0.8-1.0 mm (all results reported as 95% confidence intervals), while skew or TPS deformation improved the mean TRE by <0.1 mm. The mean misalignment was 1.1-1.9° (angle) and 0.9-1.3 mm (depth). Using isotropic scaling, the front face constraint raised the mean TRE by 0.6-0.8 mm. Conclusions: For sub-millimeter accuracy, 3D reconstruction models should not constrain histology images to the tissue slice front faces and should be flexible enough to model isotropic scaling.

Computer-Aided Prostate Cancer Detection Using Ultrasound RF Time Series: In Vivo Feasibility Study

This paper presents the results of a computer-aided intervention solution to demonstrate the application of RF time series for characterization of prostate cancer, in vivo. Methods: We pre-process RF time series features extracted from 14 patients using hierarchical clustering to remove possible outliers. Then, we demonstrate that the mean central frequency and wavelet features extracted from a group of patients can be used to build a nonlinear classifier which can be applied successfully to differentiate between cancerous and normal tissue regions of an unseen patient. Results: In a cross-validation strategy, we show an average area under receiver operating characteristic curve (AUC) of 0.93 and classification accuracy of 80%. To validate our results, we present a detailed ultrasound to histology registration framework. Conclusion: Ultrasound RF time series results in differentiation of cancerous and normal tissue with high AUC.

3D prostate histology reconstruction: An evaluation of image-based and fiducial-based algorithms

PURPOSE Evaluation of in vivo prostate imaging modalities for determining the spatial distribution and aggressiveness of prostate cancer ideally requires accurate registration of images to an accepted reference standard, such as histopathological examination of radical prostatectomy specimens. Three-dimensional (3D) reconstruction of prostate histology facilitates these registration-based evaluations by reintroducing 3D spatial information lost during histology processing. Because the reconstruction accuracy may constrain the clinical questions that can be answered with these data, it is important to assess the tradeoffs between minimally disruptive methods based on intrinsic image information and potentially more robust methods based on extrinsic fiducial markers. METHODS Ex vivo magnetic resonance (MR) images and digitized whole-mount histology images from 12 radical prostatectomy specimens were used to evaluate four 3D histology reconstruction algorithms. 3D reconstructions were computed by registering each histology image to the corresponding ex vivo MR image using one of two similarity metrics (mutual information or fiducial registration error) and one of two search domains (affine transformations or a constrained subset thereof). The algorithms were evaluated for accuracy using the mean target registration error (TRE) computed from homologous intrinsic point landmarks (3-16 per histology section; 232 total) identified on histology and MR images, and for the sensitivity of TRE to rotational, translational, and scaling initialization errors. RESULTS The algorithms using fiducial registration error and mutual information had mean ± standard deviation TREs of 0.7 ± 0.4 and 1.2 ± 0.7 mm, respectively, and one algorithm using fiducial registration error and affine transforms had negligible sensitivities to initialization errors. The postoptimization values of the mutual information-based metric showed evidence of errors due to both the optimizer and the similarity metric, and variation of parameters of the mutual information-based metric did not improve its performance. CONCLUSIONS The extrinsic fiducial-based algorithm had lower mean TRE and lower sensitivity to initialization than the intrinsic intensity-based algorithm using mutual information. A model relating statistical power to registration error for certain imaging validation study designs estimated that a reconstruction algorithm with a mean TRE of 0.7 mm would require 27% fewer subjects than the method used to initialize the algorithms (mean TRE 1.3 ± 0.7 mm), suggesting the choice of reconstruction technique can have a substantial impact on the design of imaging validation studies, and on their overall cost.

Ultrasound-Based Characterization of Prostate Cancer Using Joint Independent Component Analysis

Objective: This paper presents the results of a new approach for selection of RF time series features based on joint independent component analysis for in vivo characterization of prostate cancer. Methods: We project three sets of RF time series features extracted from the spectrum, fractal dimension, and the wavelet transform of the ultrasound RF data on a space spanned by five joint independent components. Then, we demonstrate that the obtained mixing coefficients from a group of patients can be used to train a classifier, which can be applied to characterize cancerous regions of a test patient. Results: In a leave-one-patient-out cross validation, an area under receiver operating characteristic curve of 0.93 and classification accuracy of 84% are achieved. Conclusion: Ultrasound RF time series can be used to accurately characterize prostate cancer, in vivo without the need for exhaustive search in the feature space. Significance: We use joint independent component analysis for systematic fusion of multiple sets of RF time series features, within a machine learning framework, to characterize PCa in an in vivo study.

Ultrasound-based characterization of prostate cancer: an in vivo clinical feasibility study.

UNLABELLED This paper presents the results of an in vivo clinical study to accurately characterize prostate cancer using new features of ultrasound RF time series. METHODS The mean central frequency and wavelet features of ultrasound RF time series from seven patients are used along with an elaborate framework of ultrasound to histology registration to identify and verify cancer in prostate tissue regions as small as 1.7 mm x 1.7 mm. RESULTS In a leave-one-patient-out cross-validation strategy, an average classification accuracy of 76% and the area under ROC curve of 0.83 are achieved using two proposed RF time series features. The results statistically significantly outperform those achieved by previously reported features in the literature. The proposed features show the clinical relevance of RF time series for in vivo characterization of cancer.

Toward quantitative digital histopathology for prostate cancer: comparison of inter-slide interpolation methods for tumour measurement

Accurate pathology assessment of post-prostatectomy specimens is important to determine the need for and to guide potentially life-saving adjuvant therapy. Digital pathology imaging is enabling a transition to a more objective quantification of some surgical pathology assessments, such as tumour volume, that are currently visually estimated by pathologists and subject to inter-observer variability. One challenge for tumour volume quantification is the traditional 3–5 mm spacing of images acquired from sections of radical prostatectomy specimens. Tumour volume estimates may benefit from a well-motivated approach to inter-slide tumour boundary interpolation. We implemented and tested a level set-based interpolation method and found that it produced 3D tumour surfaces that may be more biologically plausible than those produced via a simpler nearest-slide interpolation. We found that the simpler method produced larger tumour volumes, compared to the level set method, by a median factor of 2.3. For contexts where only tumour volume is of interest, we determined that the volumes produced via the simpler method can be linearly adjusted to the level setproduced volumes. The smoother surfaces from level set interpolation yielded measurable differences in tumour boundary location; this may be important in several clinical/research contexts (e.g. pathology-based imaging validation for focal therapy planning).

Using Hidden Markov Models to capture temporal aspects of ultrasound data in prostate cancer

Recent studies highlight temporal ultrasound data as highly promising in differentiating between malignant and benign tissues in prostate cancer patients. Since Hidden Markov Models can be used for capturing order and patterns in time varying signals, we employ them to model temporal aspects of ultrasound data that are typically not incorporated in existing models. By comparing order-preserving and order-altering models, we demonstrate that the order encoded in the series is necessary to model the variability in ultrasound data of prostate tissues. In future studies, we will investigate the influence of order on the differentiation between malignant and benign tissues.

Multiparametric MR imaging of prostate cancer foci: assessing the detectability and localizability of Gleason 7 peripheral zone cancers based on image contrasts

Purpose: Multiparametric magnetic resonance imaging (MPMRI) supports detection and staging of prostate cancer, but the image characteristics needed for tumor boundary delineation to support focal therapy have not been widely investigated. We quantified the detectability (image contrast between tumor and non-cancerous contralateral tissue) and the localizability (image contrast between tumor and non-cancerous neighboring tissue) of Gleason score 7 (GS7) peripheral zone (PZ) tumors on MPMRI using tumor contours mapped from histology using accurate 2D–3D registration. Methods: MPMRI [comprising T2-weighted (T2W), dynamic-contrast-enhanced (DCE), apparent diffusion coefficient (ADC) and contrast transfer coefficient images] and post-prostatectomy digitized histology images were acquired for 6 subjects. Histology contouring and grading (approved by a genitourinary pathologist) identified 7 GS7 PZ tumors. Contours were mapped to MPMRI images using semi-automated registration algorithms (combined target registration error: 2 mm). For each focus, three measurements of mean ± standard deviation of image intensity were taken on each image: tumor tissue (mT±sT), non-cancerous PZ tissue < 5 mm from the tumor (mN±sN), and non-cancerous contralateral PZ tissue (mC±sC). Detectability [D, defined as mT-mC normalized by sT and sC added in quadrature] and localizability [L, defined as mT-mN normalized by sT and sN added in quadrature] were quantified for each focus on each image. Results: T2W images showed the strongest detectability, although detectability |D|≥1 was observed on either ADC or DCE images, or both, for all foci. Localizability on all modalities was variable; however, ADC images showed localizability |L|≥1 for 3 foci. Conclusions: Delineation of GS7 PZ tumors on individual MPMRI images faces challenges; however, images may contain complementary information, suggesting a role for fusion of information across MPMRI images for delineation.