الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 193 |  |  |
| | | from | 193 | | |
Abstract
This study aimed to establish the role of artificial intelligence in providing reliable estimates of dominant features of COPD, to guide individualized management strategies and improve disease outcomes in COPD patients. This research was conducted to assess the correlation between AI based COPD parameters by CT and COPD severity using the clinical and spirometric measures and to identify the diagnostic values of CT parameters in discrimination of COPD subtypes. The study was conducted at CT unit – Radiology department at Suez Canal University hospital in Ismailia, Egypt with online remote access to a CAD System (Coreline Soft’s AVIEW). A total number of 80 patients were enrolled in this study.
The mean age of the patients was 60.0±11.7 years, 97.5% were males (97.5%). The patients were classified into mild (n=23), moderate (n=39) and severe + very severe (n=18) according to FEV1. The mean age of the severe group was significantly higher than mild and moderate groups (73.2±10.8 versus 52.6±5.2 and 58.7±9.8 years, respectively) (p<0.0001). All patients in the severe group were heavy smokers (100%), while in the moderate and mild groups, 64.1% and 21.7% were heavy smokers, respectively (p<0.0001).
Exp. LAA -856 (%) and Insp. LAA -950 (%) were the strongest significant parameters (F=37.0, p<0.0001 and F=37.2, p<0.0001, respectively). Additionally, severe group showed significantly higher values of ATI (≤60HU) (%) in comparison to mild and moderate groups (F=32.6, p<0.0001).
The other parameters such as D-value and AWT-Pi10(mm) 6th had lower significant values (p=0.020 and 0.023, respectively) than the previously