Antonio Oseas de Carvalho Filho1, Wener Borges de Sampaio2, Aristófanes Corrêa Silva3, Anselmo Cardoso de Paiva4, Rodolfo Acatauassú Nunes5, Marcelo Gattass6. 1. Federal University of Maranhão, Av. dos Portugueses, SN, Campus do Bacanga, Bacanga, 65085-580 São Luís, MA, Brazil. Electronic address: antoniooseas@gmail.com. 2. Federal University of Maranhão, Av. dos Portugueses, SN, Campus do Bacanga, Bacanga, 65085-580 São Luís, MA, Brazil. Electronic address: wenersampaio@gmail.com. 3. Federal University of Maranhão, Av. dos Portugueses, SN, Campus do Bacanga, Bacanga, 65085-580 São Luís, MA, Brazil. Electronic address: ari@dee.ufma.br. 4. Federal University of Maranhão, Av. dos Portugueses, SN, Campus do Bacanga, Bacanga, 65085-580 São Luís, MA, Brazil. Electronic address: paiva@deinf.ufma.br. 5. State University of Rio de Janeiro, São Francisco de Xavier, 524, Maracanã, 20550-900 Rio de Janeiro, RJ, Brazil. Electronic address: rodolfoacatauassu@yahoo.com.br. 6. Pontifical Catholic University of Rio de Janeiro, R. São Vicente, 225, Gávea, 22453-900 Rio de Janeiro, RJ, Brazil. Electronic address: mgattass@tecgraf.puc-rio.br.
Abstract
OBJECTIVE: The present work has the objective of developing an automatic methodology for the detection of lung nodules. METHODOLOGY: The proposed methodology is based on image processing and pattern recognition techniques and can be summarized in three stages. In the first stage, the extraction and reconstruction of the pulmonary parenchyma is carried out and then enhanced to highlight its structures. In the second stage, nodule candidates are segmented. Finally, in the third stage, shape and texture features are extracted, selected and then classified using a support vector machine. RESULTS: In the testing stage, with 140 new exams from the Lung Image Database Consortium image collection, 80% of which are for training and 20% are for testing, good results were achieved, as indicated by a sensitivity of 85.91%, a specificity of 97.70% and an accuracy of 97.55%, with a false positive rate of 1.82 per exam and 0.008 per slice and an area under the free response operating characteristic of 0.8062. CONCLUSION: Lung cancer presents the highest mortality rate in addition to one of the smallest survival rates after diagnosis. An early diagnosis considerably increases the survival chance of patients. The methodology proposed herein contributes to this diagnosis by being a useful tool for specialists who are attempting to detect nodules.
OBJECTIVE: The present work has the objective of developing an automatic methodology for the detection of lung nodules. METHODOLOGY: The proposed methodology is based on image processing and pattern recognition techniques and can be summarized in three stages. In the first stage, the extraction and reconstruction of the pulmonary parenchyma is carried out and then enhanced to highlight its structures. In the second stage, nodule candidates are segmented. Finally, in the third stage, shape and texture features are extracted, selected and then classified using a support vector machine. RESULTS: In the testing stage, with 140 new exams from the Lung Image Database Consortium image collection, 80% of which are for training and 20% are for testing, good results were achieved, as indicated by a sensitivity of 85.91%, a specificity of 97.70% and an accuracy of 97.55%, with a false positive rate of 1.82 per exam and 0.008 per slice and an area under the free response operating characteristic of 0.8062. CONCLUSION: Lung cancer presents the highest mortality rate in addition to one of the smallest survival rates after diagnosis. An early diagnosis considerably increases the survival chance of patients. The methodology proposed herein contributes to this diagnosis by being a useful tool for specialists who are attempting to detect nodules.
Authors: Robherson Wector de Sousa Costa; Giovanni Lucca França da Silva; Antonio Oseas de Carvalho Filho; Aristófanes Corrêa Silva; Anselmo Cardoso de Paiva; Marcelo Gattass Journal: Med Biol Eng Comput Date: 2018-05-23 Impact factor: 2.602
Authors: Vitória de Carvalho Brito; Patrick Ryan Sales Dos Santos; Nonato Rodrigues de Sales Carvalho; Antonio Oseas de Carvalho Filho Journal: Pattern Recognit Date: 2021-06-06 Impact factor: 7.740