Segmenting thoracic cavities with neoplastic lesions: A head-to-head benchmark with fully convolutional neural networks

Keyphrases

• Right Lung 100%
• Thoracic Cavity 100%
• Neoplastic Lesions 100%
• U-Net 100%
• Fully Convolutional Neural Network 100%
• Lung 66%
• Left Lung 66%
• Healthy Lung 66%
• Imaging Biomarkers 66%
• Dice Coefficient 66%
• Cancer Patients 33%
• Radiotherapy Planning 33%
• Neoplastic Disease 33%
• Survival Prediction 33%
• Precancerous Lesions 33%
• Evaluation Criteria 33%
• Comparative Analysis 33%
• Clinical Prognosis 33%
• Chest CT 33%
• Computerized Tomography 33%
• Treatment Planning 33%
• Automatic Segmentation 33%
• Cavity Structure 33%
• Prognosis Prediction 33%
• Lung Parenchyma 33%
• Biomarker Discovery 33%
• Treatment Survival 33%
• Computer Tomography Image 33%
• Downstream Analysis 33%
• Surface Distance 33%
• Hausdorff Distance 33%
• Deep Learning Model 33%
• Radiomics Approach 33%
• Segmentation Performance 33%
• Processing pipeline 33%
• Seal Segmentation 33%
• Symmetric Surfaces 33%
• Fully Convolutional 33%
• Convolutional Architecture 33%

Medicine and Dentistry

• Right Lung 100%
• Thoracic Cavity 100%
• Lung 100%
• Biological Marker 66%
• Computer Assisted Tomography 66%
• Left Lung 66%
• Malignant Neoplasm 33%
• Neoplasm 33%
• Treatment Planning 33%
• Lung Parenchyma 33%
• Radiomics 33%
• Radiation Therapy Planning 33%
• Survival Prediction 33%
• Evaluation Criterion 33%

Biochemistry, Genetics and Molecular Biology

• Computer Assisted Tomography 100%
• Biomarker Discovery 50%
• Radiomics 50%
• Survival Prediction 50%