English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

Due to the large combinatorial problem, current beam orientation optimizationalgorithms for radiotherapy, such as column generation (CG), are typicallyheuristic or greedy in nature, leading to suboptimal solutions. We propose areinforcement learning strategy using Monte Carlo Tree Search capable offinding a superior beam orientation set and in less time than CG.We utilized areinforcement learning structure involving a supervised learning network toguide Monte Carlo tree search (GTS) to explore the decision space of beamorientation selection problem. We have previously trained a deep neural network(DNN) that takes in the patient anatomy, organ weights, and current beams, andthen approximates beam fitness values, indicating the next best beam to add.This DNN is used to probabilistically guide the traversal of the branches ofthe Monte Carlo decision tree to add a new beam to the plan. To test thefeasibility of the algorithm, we solved for 5-beam plans, using 13 testprostate cancer patients, different from the 57 training and validationpatients originally trained the DNN. To show the strength of GTS to othersearch methods, performances of three other search methods including a guidedsearch, uniform tree search and random search algorithms are also provided. Onaverage GTS outperforms all other methods, it find a solution better than CG in237 seconds on average, compared to CG which takes 360 seconds, and outperformsall other methods in finding a solution with lower objective function value inless than 1000 seconds. Using our guided tree search (GTS) method we were ableto maintain a similar planning target volume (PTV) coverage within 1% error,and reduce the organ at risk (OAR) mean dose for body, rectum, left and rightfemoral heads, but a slight increase of 1% in bladder mean dose.

A reinforcement learning application of guided Monte Carlo Tree Search algorithm for beam orientation selection in radiation therapy