DOGS-2: Digital Pathology for Optimized Gleason Score-2

Reference number
Coordinator Region Skåne - Skånes universitetssjukhus Malmö, VO Urologi
Funding from Vinnova SEK 1 000 000
Project duration September 2018 - September 2020
Status Ongoing
Venture Medtech4Health innovators
Call Medtech4Health: Support for Innovators in Care and Healthcare 2018

Purpose and goal

Prostate cancer is the most common cancer in men. Correct characterization of the tumor by grade and stage is important for selecting the best treatment. A Gleason grade, assigned by a pathologist based on the tumor´s pattern of growth in prostate biopsies, is the best biomarker for prostate cancer prognosis today. However, the assessment is subjective and labor-intensive. In this project, we will develop an algorithm for automated assessment of Gleason grading for fast, reproducible and objective tumor assessment.

Expected results and effects

Computerized image analysis of Gleason grade will be a tool for pathologists for fast and accurate diagnosis of prostate cancer. It can lead to increased reproducibility and less variation between different pathologists regarding the aggressiveness of the tumor. The ultimate goal is to increase the precision of individualized diagnosis and treatment for prostate cancer as well as reduced costs for the society.

Planned approach and implementation

Our group, consisting of pathologists, mathematicians and software engineers, has developed a prototype algorithm which distinguishes cancerous from benign patterns on prostate biopsy images with the aid of deep learning. In the present project we propose to improve the current prototype with artificial intelligence methods by retraining it with additional convolutional neural network layers to fine tune the accuracy of the final result and validate it on two independent cohorts to demonstrate its clinical value.

The project description has been provided by the project members themselves and the text has not been looked at by our editors.

Last updated 26 April 2019

Reference number 2018-02271

Page statistics