Àá½Ã¸¸ ±â´Ù·Á ÁÖ¼¼¿ä. ·ÎµùÁßÀÔ´Ï´Ù.

±â°è ÇнÀ ¾îÇø®ÄÉÀ̼ÇÀ» È°¿ëÇÑ Æijë¶ó¸¶ ¿µ»ó¿¡¼­ÀÇ Á¤Áß °úÀ×Ä¡ ½Äº°

Identification of Mesiodens Using Machine Learning Application in Panoramic Images

´ëÇѼҾÆÄ¡°úÇÐȸÁö 2021³â 48±Ç 2È£ p.221 ~ 228
½ÂÀç±¹, ±èÀç°ï, ¾ç¿¬¹Ì, ÀÓÇüºó, ·¹¹Ý³´ÅÁ, ÀÌ´ë¿ì,
¼Ò¼Ó »ó¼¼Á¤º¸
½ÂÀç±¹ ( Seung Jae-Gook ) - Jeonbuk National University School of Dentistry Department of Pediatric Dentistry
±èÀç°ï ( Kim Jae-Gon ) - Jeonbuk National University School of Dentistry Department of Pediatric Dentistry
¾ç¿¬¹Ì ( Yang Yeon-Mi ) - Jeonbuk National University School of Dentistry Department of Pediatric Dentistry
ÀÓÇüºó ( Lim Hyung-Bin ) - Jeonbuk National University School of Dentistry Department of Pediatric Dentistry
·¹¹Ý³´ÅÁ ( Le Van Nhat Thang ) - Jeonbuk National University School of Dentistry Department of Pediatric Dentistry
ÀÌ´ë¿ì ( Lee Dae-Woo ) - Jeonbuk National University School of Dentistry Department of Pediatric Dentistry

Abstract

À̹ø ¿¬±¸´Â ¼Õ½±°Ô Á¢±Ù °¡´ÉÇÑ À¥»çÀÌÆ® ±â¹Ý ±â°è ÇнÀ ¾îÇø®ÄÉÀ̼ÇÀ» È°¿ëÇÏ¿© Æijë¶ó¸¶ ¹æ»ç¼± ¿µ»ó¿¡¼­ °úÀ×Ä¡ ½Äº° ¸ðµ¨À»ÇнÀ½ÃÅ°°í, ÇнÀµÈ ¸ðµ¨ÀÇ °úÀ×Ä¡¸¦ ½Äº°ÇÏ´Â ¼º´ÉÀ» Æò°¡ÇÏ°íÀÚ ÇÏ¿´À¸¸ç, Àΰ£ Áý´Ü°úÀÇ ¼º´ÉÀ» ºñ±³Çϱâ À§ÇÑ ¿¬±¸¸¦ ÁøÇàÇÏ¿´´Ù.
ÃÑ 1604ÀåÀÇ 5 - 7¼¼ ȯÀÚÀÇ Æijë¶ó¸¶ À̹ÌÁö°¡ À̹ø ¿¬±¸¿¡¼­ »ç¿ëµÇ¾ú´Ù. ¿¬±¸¿¡ »ç¿ëµÈ ¸ðµ¨Àº Google¿¡¼­ °³¹ßÇÑ ±â°èÇнÀ ¸ðµ¨ÀÎ Teachable MachineÀ» »ç¿ëÇÏ¿´´Ù. °úÀ×Ä¡ ½Äº° ¸ðµ¨À» ÈƷýÃÅ°°í ¼º´ÉÀ» Æò°¡Çϱâ À§ÇØ data set 1À» ¼³Á¤ÇÏ¿´´Ù. Data set 2´ÂÇнÀ¸ðµ¨°ú Àΰ£ Áý´Ü °£ÀÇ Á¤È®µµ ºñ±³¸¦ À§ÇØ ¼³Á¤ÇÏ¿´´Ù. ÇнÀ¸ðµ¨ ¹× Àΰ£ Áý´ÜÀÇ °úÀ×Ä¡ ½Äº° ´É·ÂÀ» Æò°¡Çϱâ À§ÇØ Á¤È®µµ(accuracy), ¹Î°¨µµ(sensitivity), ƯÀ̵µ(specificity) °ªÀ» »ç¿ëÇÏ¿´´Ù.
Data set 1ÀÇ °ËÁõ °á°ú, Æò±Õ 0.82ÀÇ ºÐ·ù Á¤È®µµ¸¦ ¾ò¾ú´Ù. Data set 2ÀÇ Å×½ºÆ® °á°ú, ¸ðµ¨ÀÇ Á¤È®µµ´Â 0.78À̾ú´Ù. Àü°øÀDZº°ú Çлý±ºÀÇ Æò±Õ Á¤È®µµ´Â °¢°¢ 0.82, 0.69¿´´Ù.
À̹ø ¿¬±¸´Â À¯Ä¡¿­±â ¹× Ãʱâ È¥ÇÕÄ¡¿­±â ¾î¸°ÀÌÀÇ Æijë¶ó¸¶ ¹æ»ç¼± ¿µ»ó°ú À¥ ±â¹Ý ±â°è ÇнÀ ¾îÇø®ÄÉÀÌ¼Ç ÀÌ¿ëÇÏ¿© °úÀ×Ä¡ ½Äº° ¸ðµ¨À» °³¹ßÇÏ¿´°í ÇнÀµÈ ¸ðµ¨°ú Àΰ£ ÀÇ»ç Áý´Ü(Àü°øÀÇ ¹× Çлý) °£ÀÇ °úÀ×Ä¡ ½Äº° Á¤µµ¸¦ ºñ±³ ¿¬±¸ÇÏ¿´´Ù. ÈƷøðµ¨ÀÇ ºÐ·ù Á¤È®µµ´Â Àü°øÀDZº°ú ºñ±³ ½Ã ³·¾ÒÁö¸¸ ÈƷùÞÁö ¾ÊÀº Ä¡°ú ´ëÇÐ Çлý±ºº¸´Ù ºÐ·ù Á¤È®µµ°¡ ³ô¾Æ ºñÀü¹®°¡ Çлý ¶Ç´Â ÀϹÝÀǻ翡°Ô °úÀ×Ä¡ Áø´Ü Á¤È®µµ¸¦ ³ôÀÌ´Â µ¥ È°¿ëµÉ °¡´É¼ºÀÌ ÀÖÀ½À» È®ÀÎÇÏ¿´´Ù.

The aim of this study was to evaluate the use of easily accessible machine learning application to identify mesiodens, and to compare the ability to identify mesiodens between trained model and human.
A total of 1604 panoramic images (805 images with mesiodens, 799 images without mesiodens) of patients aged 5 ? 7 years were used for this study. The model used for machine learning was Google¡¯s teachable machine. Data set 1 was used to train model and to verify the model. Data set 2 was used to compare the ability between the learning model and human group.
As a result of data set 1, the average accuracy of the model was 0.82. After testing data set 2, the accuracy of the model was 0.78. From the resident group and the student group, the accuracy was 0.82, 0.69.
This study developed a model for identifying mesiodens using panoramic radiographs of children in primary and early mixed dentition. The classification accuracy of the model was lower than that of the resident group. However, the classification accuracy (0.78) was higher than that of dental students (0.69), so it could be used to assist the diagnosis of mesiodens for non-expert students or general dentists.

Å°¿öµå

Mesiodens; Machine learning; Artificial Intelligence; Deep learning; Panoramic radiography

¿ø¹® ¹× ¸µÅ©¾Æ¿ô Á¤º¸

   

µîÀçÀú³Î Á¤º¸

KCI