INCREMENTAL AND DENSE-MRRN FOR AUTOMATIC LUNG TUMOR SEGMENTATION

In the medical field, it is essential to monitor lung tumor response for doing therapy Segmentation and accurate longitudinal tracing of tumor volume gets changed from computed tomography (CT) images actual taken. Hence, this paper aims to develop a two Multiple Resolution Residually connected Network (MRRN) formulations. This formulation includes two networks via., incremental-MRRN and dense-MRRN. This networks concurrently associates the features across multiple image resolution and feature levels through residual connections for lung tumor and segmentation. ResNet is used for residual connection but it don't eradicate poor localization and blurring issues in the image from consecutive pooling processes. There was no substantial variance in the assessments of volumetric tumor changes figured using the incremental-MRRN method with expert segmentation. This paper developed MRRN approach segmenting lung tumors volumetrically which enables precise, serial measurement of lung tumor volume by automated identification.