Study on Predicting Tumor Motion via Memory-Based Learning

Abstract:

Prediction is necessary to compensate system latency in real-time tracking radiotherapy for thoracic and abdominal cancers. The paper proposes a memory-based learning method to predict respiratory tumor motion. The method first stores the training data in memory and then finds relevant data to answer a particular query. By fitting relatively simple models to local patches instead of fitting one single global model, it is able to capture highly nonlinear and complex relations between the internal tumor motion and external surrogates accurately and immediately. Furthermore, due to the local nature of weighting functions, it is inherently robust to outliers in the training data.

• Periodical: Advanced Materials Research (Volumes 760-762)
• Pages: 2068-2071
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.760-762.2068
• Online since: September 2013
• Authors: Wei Quan Wan, Zhi Qiang He, Chao Min Chen
• Keywords: Memory-Based Learning, Prediction, Respiratory Surrogate, Tumor Motion

References

[1] Jin JY and Yin FF, Time delay measurement for linac based treatment delivery in synchronized respiratory gating radiotherapy, Medical Physics, 32(2005) 1293-1296.

DOI: 10.1118/1.1896452

[2] Martin J Murphy, Tracking moving organs in real time, Seminars in radiation oncology, 14(2004)91-100.

DOI: 10.1053/j.semradonc.2003.10.005

[3] A Krauss, S Nill and U Oelfke, The comparative performance of four respiratory motion predictors for real-time tumour tracking, Physics in Medicine and Biology, 56 (2011) 5303-5317.

DOI: 10.1088/0031-9155/56/16/015

[4] R. Laurent, J. Henriet, M. Salomon, M. Sauget, R. Gschwind and L. Makovicka, Respiratory lung motion using an artificial neural network, Neural Computing and Applications, 21 (2012) 929-934.

DOI: 10.1007/s00521-011-0727-y

[5] Dan Ruan, Kernel density estimation-based real-time prediction for respiratory motion, Physics in Medicine and Biology, 55 (2010) 1311-1326.

DOI: 10.1088/0031-9155/55/5/004

[6] Ernst, Floris, Schweikard and Achim, Forecasting respiratory motion with accurate online support vector regression (SVRpred), Cardiovascular Research, 4 (2009) 439-447.

DOI: 10.1007/s11548-009-0355-5

[7] Ruijiang Li, John H Lewis, Ross I Berbeco and Lei Xing. Real-time tumor motion estimation using respiratory surrogate via memory-based learning, Physics in Medicine and Biology, 57 (2012) 4771-4786.

DOI: 10.1088/0031-9155/57/15/4771

[8] Vapnik V, Principles of Risk Minimization for Learning Theory. San Mateo, CA: Morgan Kaufman, (1992).

[9] Atkeson C G, Moore A W and Schaal S, Locally weighted learning, Artificial Intelligence Review, 11 (1997) 11-73.

DOI: 10.1007/978-94-017-2053-3_2

[10] B. W. Silverman, Density Estimation for Statistics and Data Analysis. London: Chapman and Hall/CRC, (1986).

[11] R Li and L Xing, SU-D-BRA-01: Accurate Real-time Tumor Motion Estimation From Respiratory Surrogates Via Memory-Based Learning, Medical Physics, 39 (2012) 3616.

DOI: 10.1118/1.4734679