Deep learning for automatic Gleason pattern classification for grade group determination of prostate biopsies

Abstract

Histopathologic grading of prostate cancer using Gleason patterns (GPs) is subject to a large inter-observer variability, which may result in suboptimal treatment of patients. With the introduction of digitization and whole-slide images of prostate biopsies, computer-aided grading becomes feasible. Computer-aided grading has the potential to improve histopathological grading and treatment selection for prostate cancer. Automated detection of GPs and determination of the grade groups (GG) using a convolutional neural network. In total, 96 prostate biopsies from 38 patients are annotated on pixel-level. Automated detection of GP 3 and GP ≥ 4 in digitized prostate biopsies is performed by re-training the Inception-v3 convolutional neural network (CNN). The outcome of the CNN is subsequently converted into probability maps of GP ≥ 3 and GP ≥ 4, and the GG of the whole biopsy is obtained according to these probability maps. Differentiation between non-atypical and malignant (GP ≥ 3) areas resulted in an accuracy of 92% with a sensitivity and specificity of 90 and 93%, respectively. The differentiation between GP ≥ 4 and GP ≤ 3 was accurate for 90%, with a sensitivity and specificity of 77 and 94%, respectively . Concordance of our automated GG determination method with a genitourinary pathologist was obtained in 65% ( κ = 0.70), indicating substantial agreement. A CNN allows for accurate differentiation between non-atypical and malignant areas as defined by GPs, leading to a substantial agreement with the pathologist in defining the GG. Electronic supplementary material The online version of this article (10.1007/s00428-019-02577-x) contains supplementary material, which is available to authorized users.

Digital Biopsy with Fluorescence Confocal Microscope for Effective Real-time Diagnosis of Prostate Cancer: A Prospective, Comparative Study

“…The resulting moderate level of inter-rater agreement that we found for both HE and FCM is consistent with other findings, as far as grading is concerned. Actually, an interobserver variability in Gleason attribution has already been documented for HE analyses [4,6,18]. In the Ozkan et al's study [19], the interobserver concordance was reported to be 58% (K = 0.43) with respect to the Gleason sum.…”

“…The ISUP 2014 update aimed to improve the reproducibility of grading; however, a certain degree of interobserver variability still persists and is similar to the one described for conventional Gleason scoring [18]. This is the basis for most of the second opinion requirement, which can lead to a degree of discrepancy close to 45% (K = 0.46) [20,21].…”

“…Owing to the documented variability even in HE reporting between pathologists [6,18], the current study also analyzed the level of inter-rater agreement, which was also high for both HE and FCM with regard to cancer detection (presence vs absence of cancer).…”

The Application of Artificial Intelligence in Prostate Cancer Management—What Improvements Can Be Expected? A Systematic Review

“…Working speed of the machine allowed the analysis of hundreds of slides in a few hours. With different AI models, [45][46][47] found identical results with a good concordance between pathologist and a CNN model, with high accuracy, sensibility, and specificity in diagnosing and differentiating low and high-grade PCa. Arvaniti et al [48] provided additional information.…”

En‐face optical coherence tomography for the detection of cancer in prostatectomy specimens: Quantitative analysis in 20 patients

“…One option could be image recognition by finding comparable visual characteristics by texture analysis . Another possibility is to use tissue recognition using similar deep learning algorithms now used in histology . OCT images with a higher resolution could improve these two options, allowing more subtle differences to be detected.…”