Fig. 1From: HR-MPF: high-resolution representation network with multi-scale progressive fusion for pulmonary nodule segmentation and classificationThe architecture of our proposed framework for pulmonary nodule segmentation and classification. HR-MPF is a feature extraction network, and PDM generates segmented pulmonary nodules. Classification module differentiates benign and malignant pulmonary nodules. The discriminator jointly optimizes segmentation and classification of pulmonary nodules in which a loss function with boundary consistency constraint is proposed to calculate inconsistency between the boundaries of prediction and ground truthBack to article page