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

»çÀÌÆ®¸Ê

Loading..

Please wait....

±¹³» ÇÐȸÁö

Ȩ Ȩ > ¿¬±¸¹®Çå > ±¹³» ÇÐȸÁö > µ¥ÀÌÅͺ£À̽º ¿¬±¸È¸Áö(SIGDB)

µ¥ÀÌÅͺ£À̽º ¿¬±¸È¸Áö(SIGDB)

Current Result Document : 1 / 1

ÇѱÛÁ¦¸ñ(Korean Title) ÀÌÁ¾ÀÇ OCT ±â±â·ÎºÎÅÍ »ý¼ºµÈ º¼·ý µ¥ÀÌÅͷκÎÅÍ ½ÉÃþ ÄÁº¼·ç¼Ç ½Å°æ¸ÁÀ» ÀÌ¿ëÇÑ AMD Áø´Ü
¿µ¹®Á¦¸ñ(English Title) AMD Identification from OCT Volume Data Acquired from Heterogeneous OCT Machines using Deep Convolutional Neural Network
ÀúÀÚ(Author) ±Ç¿ÀÈì   Á¤À¯Áø   ±Ç±â·æ   ¼ÛÇÏÁÖ   Oh-Heum Kwon   Yoo Jin Jung   Ki-Ryong Kwon   Ha-Joo Song  
¿ø¹®¼ö·Ïó(Citation) VOL 34 NO. 03 PP. 0124 ~ 0136 (2018. 12)
Çѱ۳»¿ë
(Korean Abstract)
½Å°æ¸ÁÀ» ÀÌ¿ëÇÏ¿© OCT ¿µ»óÀ» ºÐ¼®ÇÏ°í ´Ù¾çÇÑ ¸Á¸· ÁúȯÀ» ÀÚµ¿ Áø´ÜÇÏ´Â °Í¿¡ °üÇÑ ¿¬±¸µéÀÌ È°¹ßÇÏ°Ô ÀÌ·ç¾îÁö°í ÀÖ´Ù. ÀÌ·¯ÇÑ ¿¬±¸°¡ Çö½Ç¿¡ Àû¿ëµÇ±â À§ÇÑ ÇϳªÀÇ Áß¿äÇÑ ¿ä°ÇÀº ÇнÀµÈ ½Å°æ¸ÁÀÌ ÇнÀ¿¡ »ç¿ëµÈ µ¥ÀÌÅÍ¿Í´Â ´Ù¸¥ ±â±â¿¡¼­ »ý¼ºµÈ µ¥ÀÌÅÍ¿¡ ´ëÇؼ­µµ ¼º´ÉÀÇ Å« Ç϶ô ¾øÀÌ ÀϹÝÈ­µÉ ¼ö ÀÖ¾î¾ß ÇÑ´Ù´Â °ÍÀÌ´Ù. º» ³í¹®¿¡¼­´Â ½ÉÃþ CNNÀ» ÀÌ¿ëÇÏ¿© OCT ¿µ»óÀ¸·ÎºÎÅÍ ³ë³â±âȲ¹Ýº¯¼º(AMD)À» ÀÚµ¿ Áø´ÜÇÏ´Â °ÍÀ» ´Ù·é´Ù. ÇϳªÀÇ OCT ±â±â·ÎºÎÅÍ È¹µæÇÑ µ¥ÀÌÅÍ ¼ÂÀ» ÀÌ¿ëÇÏ¿© ½Å°æ¸ÁÀ» ÇнÀ½ÃŲ ÈÄ ´Ù¸¥ OCT ±â±â·ÎºÎÅÍ »ý»êµÈ À̹ÌÁö¸¦ Å×½ºÆ®ÇÑ °á°ú »ó´çÇÑ ¼º´ÉÀÇ Ç϶ôÀ» °üÂûÇÒ ¼ö ÀÖ¾ú´Ù. ÀÌ·¯ÇÑ ¼º´ÉÀÇ Ç϶ôÀ» ¹æÁöÇϱâ À§Çؼ­ OCT À̹ÌÁö¸¦ Á¤±ÔÈ­ ÇÏ´Â ±â¹ýÀ» Á¦¾ÈÇÏ°í ½ÇÇèÀ» ÅëÇØ ±× È¿°ú¸¦ ºÐ¼®ÇÏ¿´´Ù. Á¦¾ÈÇÑ ±â¹ýÀº OCT À̹ÌÁö¸¦ ºÐÇÒÇÏ¿© ¸Á¸·¿¡ ÇØ´çÇÏ´Â ¿µ¿ªÀ» ã¾Æ³½ ÈÄ À̹ÌÁö ³»¿¡¼­ ¸Á¸· ¿µ¿ªÀÌ ¼öÆò¿¡ °¡±î¿î ±â¿ï±â¸¦ °¡Áöµµ·Ï Á¤·Ä(align)ÇÏ¿© ÇüÅÂÀûÀÎ Ãø¸é¿¡¼­ OCT À̹ÌÁö¸¦ Á¤±ÔÈ­ ÇÏ´Â °ÍÀ» ¸ñÀûÀ¸·Î ÇÑ´Ù. ½ÇÇèÀ» ÅëÇÏ¿© Á¦¾ÈÇÑ ±â¹ýÀÌ ÀÌÁ¾ÀÇ ±â±â¿¡¼­ »ý¼ºµÈ OCT À̹ÌÁö·ÎºÎÅÍ AMD¸¦ ÀÚµ¿Áø´Ü Çϴµ¥ À־ »ó´çÇÑ ¼º´ÉÀÇ Çâ»óÀ» ´Þ¼ºÇÔÀ» º¸¿´´Ù.
¿µ¹®³»¿ë
(English Abstract)
There have been active research activities to use neural networks to analyze OCT images and make medical decisions. One requirement for these approaches to be promising solutions is that the trained network must be generalized to new devices without a substantial loss of performance. In this paper, we use a deep convolutional neural network to distinguish AMD from normal patients. The network was trained using a data set generated from an OCT device. We observed a significant performance degradation when it was applied to a new data set obtained from a different OCT device. To overcome this performance degradation, we propose an image normalization method which performs segmentation of OCT images to identify the retina area and aligns images so that the retina region lies horizontally in the image. We experimentally evaluated the performance of the proposed method. The experiment confirmed a significant performance improvement of our approach.
Å°¿öµå(Keyword) OCT   ³ë³â±âȲ¹Ýº¯¼º(AMD)   ±â°è ÇнÀ   ÄÁº¼·ç¼Ç ½Å°æ¸Á   À̹ÌÁö ºÐÇÒ   Optical Coherence Tomography   Age-Related Macular Degeneration   Machine Learning   Convolutional Neural Network   Image Segmentation  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå