利用经常性神经网络对无偏见在线培训的神经网络对外部标志者职位的预测 (Prediction of the Position of External Markers Using a Recurrent Neural Network Trained With Unbiased Online Recurrent Optimization for Safe Lung Cancer Radiotherapy)

无偏 · 循环神经网络 · 优化器 · Performer · Neural Networks ·

2022 年 5 月 24 日

Prediction of the Position of External Markers Using a Recurrent Neural Network Trained With Unbiased Online Recurrent Optimization for Safe Lung Cancer Radiotherapy

翻译：利用经常性神经网络对无偏见在线培训的神经网络对外部标志者职位的预测

Michel Pohl,Mitsuru Uesaka,Hiroyuki Takahashi,Kazuyuki Demachi,Ritu Bhusal Chhatkuli

from arxiv, 24 pages, 16 figures; Fig. 2 and acknowledgments section added, abstract, introduction, section 3.2, and conclusion section updated, formatting issues solved, English language and scientific unit mistakes corrected; accepted manuscript version

During lung radiotherapy, the position of infrared reflective objects on the chest can be recorded to estimate the tumor location. However, radiotherapy systems have a latency inherent to robot control limitations that impedes the radiation delivery precision. Prediction with online learning of recurrent neural networks (RNN) allows for adaptation to non-stationary respiratory signals, but classical methods such as RTRL and truncated BPTT are respectively slow and biased. This study investigates the capabilities of unbiased online recurrent optimization (UORO) to forecast respiratory motion and enhance safety in lung radiotherapy. We used 9 observation records of the 3D position of 3 external markers on the chest and abdomen of healthy individuals breathing during intervals from 73s to 222s. The sampling frequency was 10Hz, and the amplitudes of the recorded trajectories range from 6mm to 40mm in the superior-inferior direction. We forecast the 3D location of each marker simultaneously with a horizon value between 0.1s and 2.0s, using an RNN trained with UORO. We compare its performance with an RNN trained with RTRL, LMS, and offline linear regression. We provide closed-form expressions for quantities involved in the gradient loss calculation in UORO, thereby making its implementation efficient. Training and cross-validation were performed during the first minute of each sequence. On average over the horizon values considered and the 9 sequences, UORO achieves the lowest root-mean-square (RMS) error and maximum error among the compared algorithms. These errors are respectively equal to 1.3mm and 8.8mm, and the prediction time per time step was lower than 2.8ms (Dell Intel core i9-9900K 3.60 GHz). Linear regression has the lowest RMS error for the horizon values 0.1s and 0.2s, followed by LMS for horizon values between 0.3s and 0.5s, and UORO for horizon values greater than 0.6s.

翻译：在肺部放射治疗期间,可以记录胸前红外线反射物体的位置,以估计肿瘤的位置。然而,放射治疗系统具有机器人控制限制所固有的潜值,从而妨碍辐射输送的精确度。通过在线学习经常性神经网络(RNN)的预测,可以适应非静止呼吸信号,但传统方法,如RTRL和短程BPT,分别是缓慢和偏颇的。本项研究调查了公正在线经常性优化(UORO)预测呼吸运动和加强肺部放射治疗安全的能力。我们使用了3D在胸部和腹部呼吸的3个外部标记位置的9个观察记录。在从73到222的间隔期间健康个人的直径位值。取样频率为10Hz,记录轨迹的振荡度范围从6毫米到40毫米不等。我们同时预测每个标记的3D位置与直径值在0.1和2.0之间。我们将其表现与RNER8的3的外部标记和直径直径的直径值作了比较,在IMS和直径轨道上进行最短的计算。