핵심 개념
Deep learning methods like MEDPSeg enable accurate and efficient segmentation of pulmonary structures on CT scans, improving diagnosis and prognosis of lung diseases.
통계
Over 6000 volumetric CT scans used for training and testing.
인용구
"The COVID-19 pandemic response highlighted the potential of deep learning methods in facilitating the diagnosis, prognosis, and understanding of lung diseases through automated segmentation of pulmonary structures and lesions in chest CT."
"We propose MEDPSeg to tackle the labor-intensive and subjective nature of lung lesion segmentation, resulting in scarce availability of ground truth for supervised learning."