Multi-center collaborative cancer prognosis prediction system based on multi-source transfer learning

Abstract

The invention discloses a multi-center collaborative cancer prognosis prediction system based on multi-source transfer learning. The system includes a model parameter setting module, a data screening module and a multi-source transfer learning module; the model parameter setting module is responsible for setting cancer prognosis prediction model parameters; The data screening module is arranged in the clinical center, and the management center transmits the set model parameters to each clinical center, and each clinical center queries the sample characteristics and prognostic index data from the local database according to the model parameters, and preprocesses the data; multi-source transfer learning The module includes source model training, transfer weight calculation and target model calculation unit. The invention uses multi-source transfer learning to solve the problem of data heterogeneity between the source center and the target center and the problem of insufficient label data of the target center, and builds a more accurate prediction model under the premise of considering the multi-center data heterogeneity. At the same time, during the model training process, the original data of each institution is complementary and shared to avoid leakage of patient privacy.

Description

technical field [0001] The invention belongs to the field of medical treatment and machine learning, and in particular relates to a multi-center collaborative cancer prognosis prediction system based on multi-source transfer learning. Background technique [0002] Cancer has a high mortality rate, and with its increasing incidence, it has become one of the leading causes of human death. High-quality cancer prognosis prediction can provide a basis for doctors' clinical decision-making, which is of great significance for cancer control and treatment. [0003] Traditional prognosis prediction is based on expert clinical experience (such as TNM model) and lacks evidence-based support. With the development of medical information technology, especially electronic medical records, medical big data analysis and mining and other technologies, data-driven prognosis prediction models have attracted more and more attention. These prediction models require large-scale clinical data, bu...

AI Technical Summary

Problems solved by technology

Existing methods are difficult to apply in the absence of local labels

Moreover, large-scale data requires the joint participation of multiple institutions, and there is a risk of patient privacy leakage

Images