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/

Global: Zendy Tools annual

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

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Radiation therapy of thoracic and abdominal tumors requires incorporating therespiratory motion into treatments. To precisely account for the patientrespiratory motions and predict the respiratory signals, a generalized modelfor predictions of different types of respiratory motions is desired. The aimof this study is to explore the feasibility of developing a Long Short-TermMemory (LSTM)-based model for the respiratory signal prediction. To achievethat, 1703 sets of Real-Time Position Management data were collected fromretrospective studies across three clinical institutions. These datasets wereseparated as the training, internal validity and external validity groups. 1187datasets were used for model development and the remaining 516 datasets wereused to test the generality power of the model. Furthermore, an exhaustive gridsearch was implemented to find the optimal hyper-parameters of the LSTM model.The hyper-parameters are the number of LSTM layers, the number of hidden units,the optimizer, the learning rate, the number of epochs, and the length of timelags. Our model achieved superior accuracy over conventional artificial neuralnetwork models: with a prediction window of 500ms, the LSTM model achieved anaverage relative Mean Absolute Error (MAE) of 0.037, an average Root MeanSquare Error (RMSE) of 0.048, and a Maximum Error (ME) of 1.687 in the internalvalidity data, and an average relative MAE of 0.112, an average RMSE of 0.139and an ME of 1.811 in the external validity data. Compared to the LSTM modeltrained with default hyper-parameters, the MAE of the optimized model resultsdecreased by 20%, indicating the importance of tuning the hyper-parameters ofLSTM models to obtain superior accuracy. This study demonstrates the potentialof deep LSTM models for the respiratory signal prediction and illustrates theimpacts of major hyper-parameters in LSTM models.

Towards Real-Time Respiratory Motion Prediction based on Long Short-Term Memory Neural Networks