Three-dimensional histopathological reconstruction as a reliable ground truth for prostate cancer studies

Part I: prostate cancer detection, artificial intelligence for prostate cancer and how we measure diagnostic performance: a comprehensive review

MRI has firmly established itself as a mainstay for the detection, staging and surveillance of prostate cancer. Despite its success, prostate MRI continues to suffer from poor inter-reader variability and a low positive predictive value. The recent emergence of Artificial Intelligence (AI) to potentially improve diagnostic performance shows great potential. Understanding and interpreting the AI landscape as well as ever-increasing research literature, however, is difficult. This is in part due to widely varying study design and reporting techniques. This paper aims to address this need by first outlining the different types of AI used for the detection and diagnosis of prostate cancer, next deciphering how data collection methods, statistical analysis metrics (such as ROC and FROC analysis) and end points/outcomes (lesion detection vs. case diagnosis) affect the performance and limit the ability to compare between studies. Finally, this work explores the need for appropriately enriched investigational datasets and proper ground truth, and provides guidance on how to best conduct AI prostate MRI studies. Published in parallel, a clinical study applying this suggested study design was applied to review and report a multiple-reader multiple-case clinical study of 150 bi-parametric prostate MRI studies across nine readers, measuring physician performance both with and without the use of a recently FDA cleared Artificial Intelligence software.1.

Contour interpolation by deep learning approach

Purpose

Contour interpolation is an important tool for expediting manual segmentation of anatomical structures. The process allows users to manually contour on discontinuous slices and then automatically fill in the gaps, therefore saving time and efforts. The most used conventional shape-based interpolation (SBI) algorithm, which operates on shape information, often performs suboptimally near the superior and inferior borders of organs and for the gastrointestinal structures. In this study, we present a generic deep learning solution to improve the robustness and accuracy for contour interpolation, especially for these historically difficult cases.

Approach

A generic deep contour interpolation model was developed and trained using 16,796 publicly available cases from 5 different data libraries, covering 15 organs. The network inputs were a 128 × 128 × 5 image patch and the two-dimensional contour masks for the top and bottom slices of the patch. The outputs were the organ masks for the three middle slices. The performance was evaluated on both dice scores and distance-to-agreement (DTA) values.

Results

The deep contour interpolation model achieved a dice score of 0.95 ± 0.05 and a mean DTA value of 1.09 ± 2.30    mm , averaged on 3167 testing cases of all 15 organs. In a comparison, the results by the conventional SBI method were 0.94 ± 0.08 and 1.50 ± 3.63    mm , respectively. For the difficult cases, the dice score and DTA value were 0.91 ± 0.09 and 1.68 ± 2.28    mm by the deep interpolator, compared with 0.86 ± 0.13 and 3.43 ± 5.89    mm by SBI. The t-test results confirmed that the performance improvements were statistically significant ( p < 0.05 ) for all cases in dice scores and for small organs and difficult cases in DTA values. Ablation studies were also performed.

Conclusions

A deep learning method was developed to enhance the process of contour interpolation. It could be useful for expediting the tasks of manual segmentation of organs and structures in the medical images.

A statistical, voxelised model of prostate cancer for biologically optimised radiotherapy

Background and purpose

Radiation therapy (RT) is commonly indicated for treatment of prostate cancer (PC). Biologicallyoptimised RT for PC may improve disease-free survival. This requires accurate spatial localisation and characterisation of tumour lesions. We aimed to generate a statistical, voxelised biological model to complement in vivomultiparametric MRI data to facilitate biologically-optimised RT.

Material and methods

Ex vivo prostate MRI and histopathological imaging were acquired for 63 PC patients. These data were co-registered to derive three-dimensional distributions of graded tumour lesions and cell density. Novel registration processes were used to map these data to a common reference geometry. Voxelised statistical models of tumour probability and cell density were generated to create the PC biological atlas. Cell density models were analysed using the Kullback–Leibler divergence to compare normal vs. lognormal approximations to empirical data.

Results

A reference geometry was constructed using ex vivo MRI space, patient data were deformably registered using a novel anatomy-guided process. Substructure correspondence was maintained using peripheral zone definitions to address spatial variability in prostate anatomy between patients. Three distinct approaches to interpolation were designed to map contours, tumour annotations and cell density maps from histology into ex vivo MRI space. Analysis suggests a log-normal model provides a more consistent representation of cell density when compared to a linear-normal model.

Conclusion

A biological model has been created that combines spatial distributions of tumour characteristics from a population into three-dimensional, voxelised, statistical models. This tool will be used to aid the development of biologically-optimised RT for PC patients.

Contrast-enhanced ultrasound with dispersion analysis for the localization of prostate cancer: correlation with radical prostatectomy specimens

PURPOSE

To determine the value of two-dimensional (2D) contrast-enhanced ultrasound (CEUS) imaging and the additional value of contrast ultrasound dispersion imaging (CUDI) for the localization of clinically significant prostate cancer (csPCa).

METHODS

In this multicentre study, subjects scheduled for a radical prostatectomy underwent 2D CEUS imaging preoperatively. CUDI maps were generated from the CEUS recordings. Both CEUS recordings and CUDI maps were scored on the likelihood of presenting csPCa (any Gleason ≥ 4 + 3 and Gleason 3 + 4 larger than 0.5 mL) by five observers and compared to radical prostatectomy histopathology. An automated three-dimensional (3D) fusion protocol was used to match imaging with histopathology. Receiver operator curve (ROC) analysis was performed per observer and imaging modality.

RESULTS

133 of 216 (62%) patients were included in the final analysis. Average area under the ROC for all five readers for CEUS, CUDI and the combination was 0.78, 0.79 and 0.78, respectively. This yields a sensitivity and specificity of 81 and 64% for CEUS, 83 and 56% for CUDI and 83 and 55% for the combination. Interobserver agreement for CEUS, CUDI and the combination showed kappa values of 0.20, 0.18 and 0.18 respectively.

CONCLUSION

The sensitivity and specificity of 2D CEUS and CUDI for csPCa localization are moderate. Despite compressing CEUS in one image, CUDI showed a similar performance to 2D CEUS. With a sensitivity of 83% at cutoff point 3, it could become a useful imaging procedure, especially with 4D acquisition, improved quantification and combination with other US imaging techniques such as elastography.

Automated multiparametric localization of prostate cancer based on B-mode, shear-wave elastography, and contrast-enhanced ultrasound radiomics

OBJECTIVES

The aim of this study was to assess the potential of machine learning based on B-mode, shear-wave elastography (SWE), and dynamic contrast-enhanced ultrasound (DCE-US) radiomics for the localization of prostate cancer (PCa) lesions using transrectal ultrasound.

METHODS

This study was approved by the institutional review board and comprised 50 men with biopsy-confirmed PCa that were referred for radical prostatectomy. Prior to surgery, patients received transrectal ultrasound (TRUS), SWE, and DCE-US for three imaging planes. The images were automatically segmented and registered. First, model-based features related to contrast perfusion and dispersion were extracted from the DCE-US videos. Subsequently, radiomics were retrieved from all modalities. Machine learning was applied through a random forest classification algorithm, using the co-registered histopathology from the radical prostatectomy specimens as a reference to draw benign and malignant regions of interest. To avoid overfitting, the performance of the multiparametric classifier was assessed through leave-one-patient-out cross-validation.

RESULTS

The multiparametric classifier reached a region-wise area under the receiver operating characteristics curve (ROC-AUC) of 0.75 and 0.90 for PCa and Gleason > 3 + 4 significant PCa, respectively, thereby outperforming the best-performing single parameter (i.e., contrast velocity) yielding ROC-AUCs of 0.69 and 0.76, respectively. Machine learning revealed that combinations between perfusion-, dispersion-, and elasticity-related features were favored.

CONCLUSIONS

In this paper, technical feasibility of multiparametric machine learning to improve upon single US modalities for the localization of PCa has been demonstrated. Extended datasets for training and testing may establish the clinical value of automatic multiparametric US classification in the early diagnosis of PCa.

KEY POINTS

• Combination of B-mode ultrasound, shear-wave elastography, and contrast ultrasound radiomics through machine learning is technically feasible. • Multiparametric ultrasound demonstrated a higher prostate cancer localization ability than single ultrasound modalities. • Computer-aided multiparametric ultrasound could help clinicians in biopsy targeting.

Convective-Dispersion Modeling in 3D Contrast-Ultrasound Imaging for the Localization of Prostate Cancer

Despite being the solid tumor with the highest incidence in western men, prostate cancer (PCa) still lacks reliable imaging solutions that can overcome the need for systematic biopsies. Dynamic contrast-enhanced ultrasound imaging (DCE-US) allows us to quantitatively characterize the vascular bed in the prostate, due to its ability to visualize an intravenously administered bolus of contrast agents. Previous research has demonstrated that DCE-US parameters related to the vascular architecture are useful markers for the localization of PCa lesions. In this paper, we propose a novel method to assess the convective dispersion (D) and velocity (v) of the contrast bolus spreading through the prostate from three-dimensional (3D) DCE-US recordings. By assuming that D and v are locally constant, we solve the convective-dispersion equation by minimizing the corresponding regularized least-squares problem. 3D multiparametric maps of D and v were compared with 3D histopathology retrieved from the radical prostatectomy specimens of six patients. With a pixel-wise area under the receiver operating characteristic curve of 0.72 and 0.80, respectively, the method shows diagnostic value for the localization of PCa.

Multiparametric ultrasound: evaluation of greyscale, shear wave elastography and contrast-enhanced ultrasound for prostate cancer detection and localization in correlation to radical prostatectomy specimens

BACKGROUND

The diagnostic pathway for prostate cancer (PCa) is advancing towards an imaging-driven approach. Multiparametric magnetic resonance imaging, although increasingly used, has not shown sufficient accuracy to replace biopsy for now. The introduction of new ultrasound (US) modalities, such as quantitative contrast-enhanced US (CEUS) and shear wave elastography (SWE), shows promise but is not evidenced by sufficient high quality studies, especially for the combination of different US modalities. The primary objective of this study is to determine the individual and complementary diagnostic performance of greyscale US (GS), SWE, CEUS and their combination, multiparametric ultrasound (mpUS), for the detection and localization of PCa by comparison with corresponding histopathology.

METHODS/DESIGN

In this prospective clinical trial, US imaging consisting of GS, SWE and CEUS with quantitative mapping on 3 prostate imaging planes (base, mid and apex) will be performed in 50 patients with biopsy-proven PCa before planned radical prostatectomy using a clinical ultrasound scanner. All US imaging will be evaluated by US readers, scoring the four quadrants of each imaging plane for the likelihood of significant PCa based on a 1 to 5 Likert Scale. Following resection, PCa tumour foci will be identified, graded and attributed to the imaging-derived quadrants in each prostate plane for all prostatectomy specimens. Primary outcome measure will be the sensitivity, specificity, negative predictive value and positive predictive value of each US modality and mpUS to detect and localize significant PCa evaluated for different Likert Scale thresholds using receiver operating characteristics curve analyses.

DISCUSSION

In the evaluation of new PCa imaging modalities, a structured comparison with gold standard radical prostatectomy specimens is essential as first step. This trial is the first to combine the most promising ultrasound modalities into mpUS. It complies with the IDEAL stage 2b recommendations and will be an important step towards the evaluation of mpUS as a possible option for accurate detection and localization of PCa.

TRIAL REGISTRATION

The study protocol for multiparametric ultrasound was prospectively registered on Clinicaltrials.gov on 14 March 2017 with the registry name 'Multiparametric Ultrasound-Study for the Detection of Prostate Cancer' and trial registration number NCT03091231.

Accurate validation of ultrasound imaging of prostate cancer: a review of challenges in registration of imaging and histopathology

As the development of modalities for prostate cancer (PCa) imaging advances, the challenge of accurate registration between images and histopathologic ground truth becomes more pressing. Localization of PCa, rather than detection, requires a pixel-to-pixel validation of imaging based on histopathology after radical prostatectomy. Such a registration procedure is challenging for ultrasound modalities; not only the deformations of the prostate after resection have to be taken into account, but also the deformation due to the employed transrectal probe and the mismatch in orientation between imaging planes and pathology slices. In this work, we review the latest techniques to facilitate accurate validation of PCa localization in ultrasound imaging studies and extrapolate a general strategy for implementation of a registration procedure.