Chemoradiotherapy standardized quality control quantitative method based on neural network

Abstract

The invention discloses a chemoradiotherapy standardized quality control quantitative method based on a neural network. The method comprises the following steps that a quality control index is quantized to obtain a quality control vector quantity, namely, quality control index quantification and dimensionality reduction and normalization of the quality control vector quantity are conducted; the neural network is formed, namely, a network topology structure, nerve cell quantity and a transfer function are confirmed; the neural network is trained, namely, a training sample is selected, errors are counted, and network parameters are adjusted; the trained neural network is applied to the quantization of chemoradiotherapy standardized quality control. Due to the fact that a clear relation does exist between the quality control vector quantity and a quality control score, the relation is usually obtained through comprehensive judgment by a specialist according to experience, the chemoradiotherapy standardized quality control quantitative method based on the neural network can well simulate the nonlinear implication relation with the utilization of the neural network through study, and therefore chemoradiotherapy standardized quality control degree in follow-up therapeutic processes can be systematically and standardly judged.

Description

technical field [0001] The invention relates to a quantification method for standardized quality control of radiotherapy and chemotherapy, in particular to a quantification method for standardized quality control of radiotherapy and chemotherapy based on a neural network. Background technique [0002] Artificial neural network, referred to as neural network for short, is a mathematical model that simulates a biological neural network. It is composed of multiple artificial neurons connected to each other. Each neuron performs a weighted summation of the outputs of neurons connected to its input, and Through a transfer function, the calculation result of itself is output to the neuron connected to its output, so that the neural network can respond to the given excitation input from the outside world. Through training, the connection strength between each neuron can be changed, that is, the weight of input weighting, so as to achieve different degrees of activation or inhibitio...

AI Technical Summary

Problems solved by technology

However, while chemotherapy and radiotherapy kill tumor cells, it also kills normal cells in the human body, which leads to complications such as physical weakness, decreased immunity, bone marrow suppression, and digestive disorders, and may have toxic side effects on various organs and the nervous system.

Images