Quantification and Classification
Lung Cancer Prognostication Using CT Image-Based Features
Other issues in the implementation of LDCT screening for lung cancer are overdiagnosis and disease recurrence after curative surgery on nodules detected at an early stage. Pulmonary nodules detected on lung cancer screening include not only aggressive lesions but also indolent lesions that may not progress to significant disease. This overdiagnosis is a potential drawback of screening because of unnecessary treatment, additional follow-up cost, and patient anxiety [96-98]. Patz et al. examined data from NLST to estimate the overdiagnosis rate and reported that more than 18% of all lung cancers detected by LDCT appeared to be indolent . The authors emphasize the need for better biomarkers and imaging techniques to determine which lung cancers are more or less aggressive for optimization of patient care and enhancement of the value of screening. LDCT screening for lung cancer increases the rate of detection of early-stage lung cancer [13, 14]. In NLST, 51.8% of lung cancers detected were reportedly at stage IA . However, early detection does not always guarantee cure. A significant number of early-stage lung cancers recur even after complete surgical resection of the primary tumor and pathological confirmation of absence of any regional lymph node metastasis . In a Japanese lung cancer registry study, the five-year survival rate of patients with stage IA lung cancer was under 80% . Recent studies have demonstrated that the use of adjuvant chemotherapy improves the survival of patients with early-stage lung cancers . A proportion of patients with stage I lung cancer have a poorer prognosis and may benefit significantly from adjuvant chemotherapy. Therefore, the identification of patients with early-stage lung cancer who have a higher risk for recurrence and who require more aggressive surveillance, or who may benefit from adjuvant therapy, has been another target of intense investigation [102-104, 116].
In this subsection, we present an example of 3D computerized quantification and show its association with histopathologic findings and postoperative outcomes [102, 105, 106]. The focus of our approach is on lung cancer prognostication on the basis of the analysis of the internal structure of a solitary pulmonary nodule. The relationship between the subjective CT appearance of pulmonary nodules and histopathologic findings, including the Noguchi classification  and postoperative outcomes, has been investigated in several studies [108-116].