Technology | Treatment Planning | December 27, 2018

RayStation 8B Released With Machine Learning Applications for Treatment Planning

New features allow automated treatment planning and organ segmentation

RayStation 8B Released With Machine Learning Applications for Treatment Planning

December 27, 2018 — RaySearch Laboratories released RayStation 8B, the latest version of the radiation therapy treatment planning system (TPS), with new advances including some automation applications using machine learning and deep learning. Other major news in RayStation 8B include a new module for evaluation of robustness of treatment plans and photon Monte Carlo dose.

Machine learning means that the algorithms in these features learn from the data they are trained on, and they learn in a way that resembles human logic. The applications for automated treatment planning and automated organ segmentation will help improve efficiency and consistency in the clinic. The machine learning framework delivered with RayStation 8B allows for models to be trained on the clinic’s available data, and it is also possible to use pre-trained models provided by RaySearch.

The new features have been developed by RaySearch’s in-house machine learning department in collaboration with Princess Margaret Cancer Centre in Toronto, Canada, and are the first machine learning applications in a TPS on the radiation oncology market today, according to the company.

Robust evaluation is a new module in Plan Evaluation, which enables efficient evaluation of robustness of treatment plans with respect to uncertainties in patient setup and density interpretation of computed tomography (CT). Multiple scenarios with different uncertainty settings are easily created and the scenarios can be evaluated simultaneously for quick decision support. This is achieved through display of different robustness metrics connected to the clinical goals of the treatment.

Improvements for photon planning include a Monte Carlo dose engine. The Monte Carlo dose algorithm brings improved accuracy and it is utilizing the graphics processing unit (GPU) to enable fast dose computation. The dose for a dual arc volumetric modulated arc therapy (VMAT) plan can be computed in less than 60 seconds, which is at least one order of magnitude faster than any other system on the market, according to RaySearch. The Monte Carlo dose engine can be used also during optimization.

Boron neutron capture therapy (BNCT) planning has been under development since 2017 and is now available in RayStation 8B. BNCT is a type of radiation therapy that enables targeting of cancer at the cellular level. BNCT planning in RayStation is developed together with Sumitomo Heavy Industries Ltd and Neutron Therapeutics Inc.

RayStation 8B also adds various improvements and enhancements, including support for directly deliverable multicriteria optimization (MCO) for VMAT, collimation of individual energy layers with the Adaptive Aperture of the Mevion S250i Hyperscan proton therapy system, and handling of relative biological effectiveness for proton dose.

For more information: www.raysearchlabs.com

Related Content

One in three adults treated for cancer may experience side effects from treatment they wish they had known more about, according to a new survey published today in the Journal of Oncology Practice
News | Clinical Trials | November 21, 2019
November 21, 2019 — One in three adults treated for cancer may experience side effects from treatment they wish they
Infinitt PACS 7.0 is a faster, more powerful viewer that was built from the ground up to support AI for image analysis and for operational/ workflow improvements in radiology
News | PACS | November 19, 2019
November 19, 2019 — Infinitt North America will be highlighting a next generation,...
 LifeImage Baystate
News | Artificial Intelligence | November 18, 2019
November 18, 2019 — Baystate Health, the premier integrated health syst
News | Artificial Intelligence | November 18, 2019
November 18, 2019 — Visage Imaging, Inc. (Visage), a wholly owned subsidi
The study finds it's possible to use commercial facial recognition software to identify people from brain MRI that includes imagery of the face
News | Magnetic Resonance Imaging (MRI) | November 15, 2019
November 15, 2019 — Though identifying data typically are removed from medical image files before they are shared for
EchoGo uses artificial intelligence (AI) to calculate cardiac ultrasound left ventricular ejection fraction (EF), the most frequently used measurement of heart function, left ventricular volumes (LV) and, for the first time for an AI application, automated cardiac strain.

EchoGo uses artificial intelligence (AI) to calculate cardiac ultrasound left ventricular ejection fraction (EF), the most frequently used measurement of heart function, left ventricular volumes (LV) and, for the first time for an AI application, automated cardiac strain.

News | Cardiovascular Ultrasound | November 14, 2019
November 14, 2019 — Ultromics has received 510(k) clearance from the U.S.
 MaxQ AI
News | Artificial Intelligence | November 13, 2019
November 13, 2019 – MaxQ AI announced a new partnership agreement with...