• Àüü
  • ÀüÀÚ/Àü±â
  • Åë½Å
  • ÄÄÇ»ÅÍ
´Ý±â

»çÀÌÆ®¸Ê

Loading..

Please wait....

Ȩ Ȩ > ¿¬±¸¹®Çå >

Current Result Document :

ÇѱÛÁ¦¸ñ(Korean Title) ÀÇ·á ¿µ»ó ºÐÇÒÀ» À§ÇÑ ºñÀü Æ®·£½ºÆ÷¸Ó¿¡ °üÇÑ ¿¬±¸
¿µ¹®Á¦¸ñ(English Title) A Study on Vision Transformer for Medical Image Segmentation
ÀúÀÚ(Author) Tuan Le Dinh   ±Ç¼º±Ù   À̼®È¯   ±Ç±â·æ   Tuan Le Dinh   Seong-Geun Kwon   Suk-Hwan Lee      Ki-Ryong Kwon  
¿ø¹®¼ö·Ïó(Citation) VOL 48 NO. 02 PP. 0833 ~ 0835 (2021. 12)
Çѱ۳»¿ë
(Korean Abstract)
ÃÖ±Ù ¸î ³â µ¿¾È ÀÇ·á ¹× »ý¹°ÀÇÇÐ À̹ÌÁö ºÐÇÒ ¹®Á¦¿¡ ´ëÇÑ ¸¹Àº º¥Ä¡¸¶Å© Â÷Æ®¿¡¼­ Æ®·£½ºÆ÷¸Ó(transformer)¸¦ ±â¹ÝÀ¸·Î ÇÏ´Â ½Å°æ¸ÁÀÌ °¢±¤¹Þ°í ÀÖ´Ù. 2017³â Æ®·£½ºÆ÷¸Ó°¡ µµÀÔµÈ ÀÌÈÄ »ç½Ç»ó NLP(ÀÚ¿¬¾î ó¸®) °ü·Ã ±â¼úÀÇ Ç¥ÁØÀÌ µÇ¾úÀ¸¸ç, NLP¿ë Æ®·£½ºÆ÷¸ÓÀÇ ¼º°ø ÀÌÈÄ ½Ã°¢ °ü·Ã ±â¼ú ¹× ÀÀ¿ë ºÐ¾ß¿¡ Æ®·£½ºÆ÷¸Ó¸¦ Àû¿ëÇÏ·Á´Â ¸¹Àº ½Ãµµ°¡ ÀÖ¾ú°í ÇöÀå¿¡¼­ ¸¹Àº SOTA(state-of-the-art) ±â¹ýÀ» ´É°¡ÇÏ´Â °á°ú¸¦ º¸¿©ÁÖ°í ÀÖ´Ù. º» ³í¹®¿¡¼­´Â ÀÇ·á ¹× »ý¹° ÀÇÇÐ °ü·Ã ºÐ¾ß¿¡ Àû¿ëµÇ´Â ÃֽŠƮ·£½ºÆ÷¸Ó ¸ðµ¨À» »ìÆ캸°íÀÚµ¿ ½ÉÀå Áø´Ü(ACDC) µ¥ÀÌÅÍ ¼¼Æ®¿Í ½Ã³À½º ´Ù±â°ü CT µ¥ÀÌÅÍ ¼¼Æ®¿¡¼­ Average Dice Score (DSC) Çà·Ä ¹× Average Hausdorff Distance metric (HD) Çà·Ä°ú °ü·ÃÇÏ¿© °¢ ¸ðµ¨ÀÇ °á°ú¸¦ Æò°¡ÇÏ¿´´Ù.
¿µ¹®³»¿ë
(English Abstract)
In recent years, neural networks based on transformers have been dominating many benchmark charts of medical and biomedical image segmentation challenges. Since the introduction of the transformer in 2017, the network has become the de facto standard for Natural Language Processing (NLP) tasks. After the success of transformers for NLP, many attempts have been made to apply transformers for Vision tasks and have surpassed many state-of-the-art methods in the field. In this paper, we will walk through many modern vision transformer models which are applied for medical and biomedical tasks. And we demonstrate many numerical evaluations of those models on two public datasets which are the Automatic Cardiac Diagnosis (ACDC) dataset and the Synapse Multi-organ CT dataset with respect to Average Dice Score (DSC) metric and Average Hausdorff Distance metric (HD).
Å°¿öµå(Keyword) ÀÇ·á ¹× »ý¹°ÀÇÇÐ À̹ÌÁö ºÐÇÒ   ÀÚ¿¬¾î 󸮠  ºñÀü Æ®·£½ºÆ÷¸Ó   Average Dice Score   Average Hausdorff Distance   medical and biomedical image segmentation   Natural Language Processing   vision transformer   Average Dice Score   Average Hausdorff Distance.  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå