Virtual Library
Start Your Search
Jindong Guo
Author of
-
+
P3.13 - Radiology/Staging/Screening (ID 729)
- Event: WCLC 2017
- Type: Poster Session with Presenters Present
- Track: Radiology/Staging/Screening
- Presentations: 1
- Moderators:
- Coordinates: 10/18/2017, 09:30 - 16:00, Exhibit Hall (Hall B + C)
-
+
P3.13-010 - Correlation between HRCT Features of Pulmonary Pure Ground-Glass Nodules and the New Pathologic Classification of Lung Adenocarcinoma (ID 8457)
09:30 - 09:30 | Presenting Author(s): Jindong Guo
- Abstract
Background:
We analysed the correlation between the new pathologic classification of lung adenocarcinoma and radiologic findings of early invasive pulmonary adenocarcinomas versus preinvasive lesions appearing as PGGN on HRCT, and evaluated the values in the diagnosis of pathologic classification of lung adenocarcinoma with PGGN on HRCT.
Method:
Retrospective analysis of 123 lesions (16AAH, 35AIS, 35MIA, 37IA) with PGGN on HRCT with T1N0M0 lung adenocarcinoma or AAH from January 2014 to June 2014 in shanghai chest hospital. There were 93 females and 30 males, with a median age of 58 years old. Statistical relationship between the 2015 World Health Organization Classification of the lung adenocarcinoma and radiologic findings of PGGN were analyzed, then sceened out the best predictors, created a modal and verified it.
Result:
The Pearson correlation coefficient( P<0.05) between pathological types and all CT scan morphologic features showed a significant correlation. The logarithm linear correlation cient showed the CT feathures(lobulation, spiculation, pleural indentation, aterial gathering, bubbles/air bronchogram, shape,margin,internal uniformity) had a positive correlation with pathological types excluding tumor-lung interface. These scale variables as maximum lesion area on CT scan, lesion size in cranial-caudal direction, average density of lesion and the corresponding lung’s average background density were significant correlation with pathological types. Multinomial logistic regression analysis showed that the best predictors were spiculation, internal uniformity, lesion size in cranial-caudal direction,average density of lesion, gender in turn. Then the multinomial logistic regression model was built, a likelihood ratio test showed that 70.7% of the cases were classified correctly overall, and the predicted value of AAH was up to 92.9%. Figure 1
Conclusion:
The HRCT characteristics of PGGN were significant correlated with the new pathologic classification of lung adenocarcinoma. The pathologic types of PGGN should be evaluated by HRCT, and the best predictors were speculation, internal uniformity, lesion size in cranial-caudal direction, average density of lesion and gender.
