Stimulated Raman Scattering Microscopy Enables Gleason Scoring of Prostate Core Needle Biopsy by a Convolutional Neural Network

Sep 24, 2026

Focal therapy for clinically localized prostate cancer requires rapid, accurate histology of core needle biopsies to grade tumors and guide treatment while minimizing toxicity to surrounding tissue. Stimulated Raman scattering (SRS) microscopy provides label-free, near real-time histologic imaging of fresh prostate biopsy samples. A convolutional neural network trained on images from 61 patients classified Gleason patterns with 85.7% accuracy, validated on 22 independent cases with 84.4% accuracy. Deep learning SRS grading showed 71% consistency with three pathologists, supporting workflow simplification for FT-compatible evaluation.

LanguageEnglish

CategoryHealthcare

FormatPDF

Pages11

Reading Time28 min

AccessFree to download

Introduction

Focal therapy and the need for rapid, accurate histology

Challenges with conventional intraoperative workflows

SRS microscopy as label-free near real-time imaging

Gleason scoring and interpretation challenges

Stimulated Raman Scattering Microscopy Enables Gleason Scoring of Prostate Core Needle Biopsy by a Convolutional Neural Network Jianpeng Ao1, Xiaoguang Shao2, Zhijie Liu1, Qiang Liu3, Jun Xia3, Yongheng Shi3, Lin Qi4, Jiahua Pan2, and Minbiao Ji1 ◥ Focal therapy(FT)has been proposed asan approach to eradicate clinically signiﬁcant prostate cancer while preserving the normal surrounding tissues to minimize treatment-related toxicity. Rapid histology of core needle biopsies is essential to ensure the precise FT for localized lesions and to determine tumor grades. However, it is difﬁcult to achieve both high accuracy and speed with currently available histopathology methods. Here, we demonstrated that stimulated Raman scattering (SRS) microscopy could reveal the largely heterogeneous histologic features of fresh prostatic biopsy tissues in a label-free and near real-time manner. A diagnostic convolutional neural network(CNN) built based on images from61 patients could classify Gleason patterns of prostate cancer with an accuracy of 85.7% . An additional 22 independent cases introduced as external test dataset validated the CNN performance with 84.4% accuracy. Gleason scores of core needle biopsies from 21 cases were calculated using the deep learning SRS system and showed a 71% diagnostic consistency with grading from three pathologists. This study demonstrates the potential of a deep learning–assisted SRS platform in evaluating the tumor grade of prostate cancer, which could help simplify the diagnostic workﬂow and provide timely histopathology compatible with FT treatment. Signiﬁcance: A platform combining stimulated Raman scattering microscopy and a convolutional neural network provides rapid histopathology and automated Gleason scoring on fresh prostate core needle biopsies without complex tissue processing. Introduction Prostate cancer is the second most diagnosed cancer and the ﬁfth leading cause of cancer-related deaths among men worldwide (1) . Over the past decades, focal therapy (FT) has been an emerging novel option for selected patients with clinically localized cancer like prostate cancer, aiming to eradicate the targeted lesion while preserving the surrounding key functional structures(2–5)However, many urologists are concerned that FT could fail in cancer control for the inability of imaging biopsy to detect multifocal lesions in prostate cancer and the inaccuracy of tumor location during operation (3) . The main concern was derived from the incomplete tumor ablation due to the lack of real-time intraoperative pathologic guidance. The conventional intraoperative frozen section–based workﬂow requires highly effective cooperation during tissue transport and slide processing, which is not 1State Key Laboratory of Surface Physics and Department of Physics, Human Phenome Institute, Academy for Engineering and Technology, Key Laboratory of Micro and Nano Photonic Structures (Ministry of Education), Yiwu Research Institute of Fudan University, Fudan University, Shanghai, P. R. China. 2Department of Urology, Ren Ji Hospital, School of Medicine, Shanghai Jiao Tong University, Shanghai, P. R. China. 3Department of Pathology, Ren Ji Hospital, School of Medicine, Shanghai Jiao Tong University, Shanghai, P. R. China. 4Department of Radiology, Huadong Hospital Afﬁliated to Fudan University, Shanghai, P. R. China. J. Ao, X. Shao, and Z. Liu contributed equally to this article. Corresponding Authors: Minbiao Ji, Fudan University, No 2005 Songhu Rd, Shanghai 200433, P. R. China. Phone: 86-18117163505; E-mail: [minbiaoj@fudan.edu.cn](minbiaoj@fudan.edu.cn); and Jiahua Pan, [jiahua.pan@outlook.com](jiahua.pan@outlook.com) [Cancer Res 2023](Cancer Res 2023);[83:641](83:641)–[51](51) doi: 10.1158/0008-5472 . CAN-22-2146 This open access article is distributed under the Creative Commons AttributionNonCommercial-NoDerivatives 4 . 0 International (CC BY-NC-ND 4 . 0) license. 􀀁2023The Authors;Published by the

FAQs

Why is rapid histology critical for focal therapy prostate cancer treatment?

Focal therapy depends on accurate, timely grading from core needle biopsies to target localized lesions and support intraoperative decision-making. Delays and imprecise diagnostics can undermine treatment control and workflow efficiency.

How does stimulated Raman scattering (SRS) microscopy help in this setting?

SRS microscopy generates label-free histologic images with near real-time speed and submicrometer resolution. It provides chemical contrast linked to biomolecular vibrational properties, avoiding fixation, freezing, sectioning, and staining processes.

What diagnostic performance did the convolutional neural network achieve for Gleason patterns?

The CNN trained on 61 patients classified Gleason patterns with 85.7% accuracy, and external testing on 22 independent cases achieved 84.4% accuracy. For 21 biopsy cases, deep learning SRS grading showed 71% consistency with three pathologists.

Sep 24, 2026

Upload by anakgang17