Conclusions and Contributions - Prediction and Classification of Respiratory Motion - page 132

The first contribution of this study is to present a new approach to split the
whole RMLP with the complicated neuron number into a couple of RMLPs with
the simple neuron number to adjust separate input channels. It also comprehen-
sively organizes the multiple channel sensory process by adapting the coupling
technique using multiple channel inputs for the mutually exclusive groups to
compensate the computational accuracy.
The second contribution is to adopt a clustering method for multiple patients to
get more practical breathing pattern information and to find an accurate prediction
process for an individual class. With the clustering based on breathing patterns, we
can get appropriate parameter selections with respect to each class—e.g., optimal
neuron number for the prediction process of the neural network and/or interactive
(coupling) degree for the multiple breathing information. It can yield a more
accurate

prediction performance than when the clustering in not based on

breathing patterns.
The third contribution is to propose a new approach to detect abnormal
breathing patterns with multiple patient-breathing data. We retrospectively cate-
gorized breathing data into several classes based on the extracted feature metrics
derived from breathing data of multiple patients. The newly proposed statistical
classification may provide clinically significant contributions to optimize the
safety margin during external beam radiotherapy based on the breathing regularity
classification for the individual patient.
The prediction of respiratory motion traces has become an important research
area due to the compensation for uncertainty and irregularity originating from
technical limitations or physiological phenomena. So far, investigations on the
prediction of respiratory motion have been limited to estimates of respiratory
motion, probably due to immature development of medical systems. This leads to
further investigations for adequate and sophisticated radiotherapy technology.
Radiation therapy is one of the most advanced treatment techniques for macro-
scopic cancers. For the accurate and precise delivery of radiation therapy, the
prediction of respiratory motion is important. Collaborative research activities
with various disciplines including biomedical, engineering, and medical physics
are required.