Temporal analysis of lung lesions through dynamic shape features. (March 2019)

• Record Type: Journal Article
• Authors: Quintanilha, Darlan Bruno Pontes; Diniz, João Otávio Bandeira; Silva, Aristófanes Corrêa; de Paiva, Anselmo Cardoso; Gattass, Marcelo
• Is Part Of: Computers & electrical engineering. Volume 74(2019)
• Page Start: 245
• Page End: 258
• Publication Date: 2019-03
• Subjects: Medical image -- Lung lesion -- Change detection -- Temporal analysis -- Dynamic shape features; Computer engineering -- Periodicals; Electrical engineering -- Periodicals; Electrical engineering -- Data processing -- Periodicals; Ordinateurs -- Conception et construction -- Périodiques; Électrotechnique -- Périodiques; Électrotechnique -- Informatique -- Périodiques; Computer engineering; Electrical engineering; Electrical engineering -- Data processing; Periodicals; Electronic journals; 621.302854
• DOI: 10.1016/j.compeleceng.2019.02.001
• Languages: English
• ISSNs: 0045-7906

Abstract

Highlights: Proposed approach is used to analyze shape changes of lung lesions over time. Our methodology uses dynamic shape features to quantify changes of lung lesions. Provide the specialist bigger support in studies on the biological activity of pulmonary nodules. Abstract: Lung cancer remains one of the highest occurring cancers worldwide. Temporal assessment is a useful tool for analyzing the malignant behavior of a lesion during treatment, or for classifying previously unclassified lesions. This paper provides relatively detailed information about the changes in lesions, complementing the results of studies on the biological activity of pulmonary nodules. The methodology presented uses dynamic shape features (DSF) to quantify and analyze the changes in lung lesions over time as detected in computed tomography (CT) images. The study was conducted using two chest CT databases: the Public Lung Database (PLD), which contains images of lesions undergoing evaluation related to the patient's drug therapy, and a private database of images of lung lesions of undetermined diagnosis. The DSF parameters presented variance values in the range from 8.66 × 10 − 05 to 2.58 × 10 06 in lesions in images in the private database, and in the range from 6.72 × 10 − 04 to 5.90 × 10 10 in lesions images in the PLD.