A segmentation tool for pulmonary nodules in lung cancer screening: Testing and clinical usage |
| |
| Affiliation: | 1. Dipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano, Milano, Italy;2. Division of Radiology, IEO, European Institute of Oncology IRCCS, Milan, Italy;3. School of Medicine, University of Milan, Milan, Italy;4. Postgraduate School of Diagnostic and Interventional Radiology, University of Milan, Milan, Italy;5. Bioengineering Unit, CNAO Foundation, Pavia, Italy;1. Section of Radiological Sciences, Department of Biomedical and Dental Sciences and Morphofunctional Imaging, University of Messina, Italy;2. INFN, National Institute for Nuclear Physics, Section of Catania, Italy;3. Institute of Radiation Physics, Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland;4. MIFT Department, University of Messina, Italy;1. Medical Physics, Centre Oscar Lambret, Lille, France;2. Academic Department of Radiotherapy, Centre Oscar Lambret, Lille, France;3. Radiology Department, Centre Oscar Lambret, Lille, France;1. Shanghai Jiao Tong University, Shanghai, China;2. Hunan University, Changsha, Hunan, China;3. Shanghai United Imaging Intelligence Co. Ltd, Shanghai, China;1. Federal University of Maranhão, Av. dos Portugueses, SN, Campus do Bacanga, Bacanga, 65085-580 São Luís, MA, Brazil;2. State University of Rio de Janeiro, São Francisco de Xavier, 524, Maracanã, 20550-900 Rio de Janeiro, RJ, Brazil;3. Pontifical Catholic University of Rio de Janeiro, R. São Vicente, 225, Gávea, 22453-900 Rio de Janeiro, RJ, Brazil;1. Instituto Federal do Ceará, Campus Maracanaú, Av. Parque Central, S/N, Distrito Industrial I, 61939-140 Maracanaú, Ceará, Brazil;2. Universidade Federal do Ceará, Departamento de Engenharia de Teleinformática, Av. Mister Hull, S/N, Campus do Pici, 6005, 60455-760 Fortaleza, Ceará, Brazil;3. Programa de Pós-Graduacão em Informática Aplicada, Universidade de Fortaleza, Av. Washington Soares, 1321, Edson Queiroz, 60811341, CEP 608113-41 Fortaleza, Ceará, Brazil;4. Instituto de Ciência e Inovacão em Engenharia Mecânica e Engenharia Industrial, Departamento de Engenharia Mecânica, Faculdade de Engenharia, Universidade do Porto, Rua Dr. Roberto Frias, S/N, 4200-465 Porto, Portugal;1. Institute of Electronics and Telecommunications, Peter the Great St.Petersburg Polytechnic University, Saint-Petersburg, Russia;2. Department of Electrical and Computer Engineering, University of California, Davis, United States |
| |
| Abstract: | PurposeWith the future goal of defining a large dataset based on low-dose CT with labelled pulmonary lesions for lung cancer screening (LCS) research, the aim of this work is to propose and evaluate into a clinical context a tool for semi-automatic segmentation able to facilitate the process of labels collection from a LCS study (COSMOS, Continuous Observation of SMOking Subjects).MethodsConsidering a preliminary set of manual annotations, a segmentation model based on a 2D-Unet was trained from scratch. Contour quality of the final 2D-Unet was assessed on an internal test set of manual annotations and on a subset of the public available LIDC dataset used as external test set. The tool for semi-automatic segmentation was then designed integrating the tested model into a Graphical User Interface. According to the opinion of two clinical users, the percentage of lesions properly contoured through the tool was quantified (Acceptance Rate, AR). The variability between segmentations derived by the two readers was estimated as mean percentage of difference (MPD) between the two sets of volumes and comparing the likelihood of malignancy derived from Volume Doubling Time (VDT).ResultsPerformance in test sets were found similar (DICE ~ 0.75(0.15)). Accordingly, a good mean AR (80.1%) resulted from the two readers. Variability in terms of MPD was equal to 23.6% while 2.7% was the VDTs percentage of disagreement.ConclusionsA semi-automatic segmentation tool was developed and its applicability evaluated into a clinical context demonstrating the efficacy of the tool in facilitating the collection of labelled data. |
| |
| Keywords: | Deep learning Automatic segmentation Lung cancer screening Clinical validation |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
