OptraSCAN Introduces AI & ML Based Scoring Algorithms for Analysis of Prostate Cancer
San Jose, CA, October 10th, 2018: OptraSCAN®, the leading end-to-end digital pathology solution provider has introduced an artificial intelligence and machine learning based scoring algorithms that provides accurate, rapid and reproducible Prostate Cancer analysis.
The classical Gleason grading system exclusively based on visual assessment of architectural patterns of prostate adenocarcinoma acknowledged by WHO and further modified and revised in 2004 and 2016 by the International Society of Urological Pathology (ISUP), is a powerful prognostic predictor for patients with prostate carcinoma. However, the histological assessment by pathologists of these tissue slides to evaluate the morphological patterns is often time-consuming and suffers from limited reproducibility. Studies published have reported Intra and Interobserver agreement variability. With the FDA approval for whole slide imaging systems and with rapid technological advances and the advent of computer-aided platforms, development of machine-based scoring algorithms show promise in overcoming these challenges and provide a possible approach to accurate, rapid and reproducible analysis.
“Prostate cancer is the most prevalent form of cancer and the second most common cause of death among men in the United States. Automated and deep learning approaches by comparing morphological and quantitative information on digital pathology images have the potential to overcome the above-mentioned limitations to quantitatively characterize the Gleason spectrum and generate reproducible results”, said Anand Maiskar, VP – Sales ROW, OptraScan®.
OptraSCAN’s AI & ML based scoring algorithms:
The proposed automated grading of prostate cancer for histopathological images is an end to end solution based on Artificial Neural Networks which intelligently segments cell nuclei into recognizable patterns like area, shape, intensity etc. The convolutional neural networks which combine human engineered feature 
extraction modules and deep neural networks can automatically detect glandular lumens and architecture. Feature fusion and feature ranking modules along with Convolutional Neural Network based techniques used, are trained to identify gland region formation, distinguish between well, moderately and poorly differentiated glands and further classify individual cell characteristics like mean intensity, area, standard deviation of stain intensity, etc. The state-of-the-art technology optimized for tissue pathology with combined feature extraction and easily retrainable machine learning system demonstrates the potential as an assistance to pathologists in grading prostate cancer.
Thus, this user-friendly deep learning technology will make it possible for the pathologists to automatically compute Gleason grade of a digitally imaged prostate tissue histopathology slide and generate highly accurate and reproducible results.
About OptraSCAN®:
OptraSCAN® (www.optrascan.com), an ISO 13485 certified is an end-to-end digital pathology solution provider. OptraSCAN is focused on delivering fully integrated, affordable solutions that will maximize your return on investment and improve the performance of your pathology services. OptraSCAN® are pioneers in the On-Demand Digital Pathology® System and is working to eliminate the barriers to “Go Digital” no matter the size of the pathology lab, the lab’s throughput or global location.
OptraSCAN’s end-to-end solution provides effective acquisition of whole slide images, viewing, storing, sharing, reporting, analysis and management of digital slides and associated metadata, via On-Demand solution or outright purchase model. Follow Us on LinkedIn and Twitter.
Contacts:
Media Inquiries:
Shruti Soman | Manager – Marketing Communications | 91.20.66540900 x 231 | s.soman@optraventures.com
Source: OptraSCAN
