Current Result Document :
ÇѱÛÁ¦¸ñ(Korean Title) |
µö ·¯´×À» »ç¿ëÇÑ X¼± ¿µ»óÀÇ Æó Áúȯ ºÐ·ù |
¿µ¹®Á¦¸ñ(English Title) |
Lung Diseases Classification on X-Ray images using Deep Learning |
ÀúÀÚ(Author) |
±è¼º¿±
¸²ºóº¸´ÏÄ«
À̼ºÁø
¸¶¿À¸¶Ä«¶ó
È«¹Î
Sungyeup Kim
Beanbonyka Rim
Sungjin Lee
Makara Mao
and Min Hong
|
¿ø¹®¼ö·Ïó(Citation) |
VOL 23 NO. 01 PP. 0203 ~ 0204 (2022. 04) |
Çѱ۳»¿ë (Korean Abstract) |
|
¿µ¹®³»¿ë (English Abstract) |
Lung disease is one of the most causes of death in the world such as pneumonia, and pneumothorax. Chest X-Ray images are scanned to diagnosis lung diseases in the early stage. The computer-aided diagnostic systems (CADs) are used to assist doctors on chest X-Ray images earlier and easier. Therefore, in this paper, we proposed a deep learning approach to classify lung diseases into pneumonia, pneumothorax, and normal. This classification would help to improve the performance of CADs for doctors to diagnosis on X-Ray images more precisely. We exploited EfficientNet-V2-M model to classify lung diseases on the NIH dataset and achieved an accuracy, sensitivity, and specificity of 82.00%, 80.60%, and 92.12% on validation set, respectively.
|
Å°¿öµå(Keyword) |
|
ÆÄÀÏ÷ºÎ |
PDF ´Ù¿î·Îµå
|