Lung lesion image segmentation method based on CovSegNet

Abstract

The invention belongs to the technical field of lung lesion image segmentation, and particularly relates to a lung lesion image segmentation method based on CovSegNet, and the method comprises the following steps: data collection, data preprocessing, model construction, model storage, and model evaluation. The data acquisition is used for acquiring various data sets from pulmonary infection, performing data annotation on images in the acquired data set, and constructing a data set required by model training; the data preprocessing is used for data division, normalization and image scaling, and data expansion is carried out; the model construction is based on a CovSegNet segmentation network model, training data are input, and a parameter model is constructed; the model saves the model after the loss function is not reduced any more; the model evaluation is used for evaluating the stored model through a plurality of evaluation indexes and knowing the related performance of the model.

Description

technical field [0001] The invention belongs to the technical field of lung lesion image segmentation, and in particular relates to a lung lesion image segmentation method based on CovSegNet. Background technique [0002] Now with the recent outbreak of Coronavirus Disease - 2019COVID-19, the world has experienced an unprecedented death toll and healthcare systems around the world are severely collapsing. Early diagnosis is the primary issue in controlling this global pandemic at this stage because of its extremely contagious nature. Although reverse transcription-polymerase chain reaction (RT-PCR) is considered the gold standard for diagnosing COVID-19, its long time requirements, low sensitivity, and massive shortage of kits have made alternative automated diagnostic protocols extremely urgent. [0003] Reason for problem or flaw: Automatic segmentation of lung lesions on lung CT scans is a critical stage for accurate diagnosis and severity measurement of COVID-19. Trad...

AI Technical Summary

Problems solved by technology

Traditional encoder-decoder architectures and their variants suffer from reduced contextual information in pooling / upsampling operations, increased semantic gap between encoding and decoding feature maps, and lead to the vanishing-gradient problem of their sequential gradient propagation, resulting in suboptimal performance

In addition, due to the exponential growth of computational complexity, making optimization difficult, there are certain limitations

Images