Lung adenocarcinoma Ki67 expression level noninvasive detection method and device based on depth radiomics

Abstract

The invention discloses a lung adenocarcinoma Ki67 expression level noninvasive detection method and device based on depth radiomics. The method comprises the following steps: S1, preprocessing data: obtaining basic clinical information and a CT image of a lung adenocarcinoma patient; S2, extracting deep learning features of the CT image: extracting deep radiomics features from the CT image, feeding back the region of interest to the network, and adaptively learning potential features in a supervised manner; S3, extracting radiomics characteristics: extracting the radiomics characteristics from the tumor image by adopting a handmade radiology signature method (HCR); and S4, constructing a prediction model by using image features of multi-source feature fusion: fusing deep learning features, manually made image omics features and clinical features, performing connection and fusion by using three FC layers, and outputting probability of a Ki67 state by a softmax function. According to the invention, the state of Ki67 in lung adenocarcinoma can be noninvasively detected, and a clinician is assisted to make a clinical decision on a lung adenocarcinoma patient.

Description

technical field [0001] The invention relates to a non-invasive detection method and device for Ki67 expression level of lung adenocarcinoma based on deep radiomics. Background technique [0002] Lung adenocarcinoma is one of the most common histopathological diseases, and the logical subtype of non-small cell lung cancer (NSCLC) accounts for 85% of all lung cancers. As a marker of tumor cell proliferation, Ki67 has been widely used in cancer research and clinical applications. [0003] More and more evidences show the clinical value of Ki67 as a predictor of curative effect, and it is expected to become an important molecular marker for diagnosis and prediction of tumors. High expression of Ki67 has been reported to be an indicator of poor prognosis in patients with non-small cell lung cancer. Therefore, Ki67 index can be used clinically to measure the extent of tumor proliferation and predict survival and recurrence. However, determination of the Ki67 index is often requ...

AI Technical Summary

Problems solved by technology

However, traditional methods for determining the Ki67 index often require sampling tumor tissue, which is invasive for patients

Images