ÇÕ¼º°ö½Å°æ¸ÁÀÇ ÇнÀ ¹× Å×½ºÆ®ÀÚ·á¿¡ µû¸¥ °ñ´Ù°øÁõ Æǵ¶¿¡ ¹ÌÄ¡´Â ¿µÇâ
Effect of Training and Testing Condition of Convolutional Neural Network on evaluating Osteoporosis
±èÀçÀ±, ÀÌÀç¼, °º´Ã¶, ±èÇü¼®, Shyam Adhikan, Liu Liu, À±¼÷ÀÚ,
¼Ò¼Ó »ó¼¼Á¤º¸
±èÀçÀ± ( Kim jae-Yun ) - Àü³²´ëÇб³ Ä¡ÀÇÇÐÀü¹®´ëÇпø
ÀÌÀç¼ ( Lee Jae-Seo ) - Àü³²´ëÇб³ Ä¡ÀÇÇÐÀü¹®´ëÇпø ±¸°¾Ç¾È¸é¹æ»ç¼±Çб³½Ç
°º´Ã¶ ( Kang Byung-Cheol ) - Àü³²´ëÇб³ Ä¡ÀÇÇÐÀü¹®´ëÇпø ±¸°¾Ç¾È¸é¹æ»ç¼±Çб³½Ç
±èÇü¼® ( Kim Hyong-Suk ) - Àü³²´ëÇб³ ÀüÀÚ°øÇкÎ
( Shyam Adhikan ) - Àü³²´ëÇб³ ÀüÀÚ°øÇкÎ
( Liu Liu ) - Àü³²´ëÇб³ ÀüÀÚ°øÇкÎ
À±¼÷ÀÚ ( Yoon Suk-Ja ) - Àü³²´ëÇб³ Ä¡ÀÇÇÐÀü¹®´ëÇпø ±¸°¾Ç¾È¸é¹æ»ç¼±Çб³½Ç
Abstract
This study aimed to test a convolutional neural network (CNN) in two different settings of training and testing data. Panoramic radiographs were selected from 1170 female dental patients (mean age 49.19 ¡¾ 21.91 yr). The cortical bone of the mandible inferior border was evaluated for osteoporosis or normal condition on the panoramic radiographs. Among them, 586 patients (mean age 27.46 ¡¾ 6.73 yr) had normal condition, and osteoporosis was interpreted on 584 patients (mean age 71.00 ¡¾ 7.64 yr). Among them, one data set of 569 normal patients (mean age 26.61 ¡¾ 4.60 yr) and 502 osteoporosis patients (mean age 72.37 ¡¾ 7.10 yr) was used for training CNN, and the other data set of 17 normal patients (mean age 55.94 ¡¾ 4.0 yr) and 82 osteoporosis patients (mean age 62.60 ¡¾ 5.00 yr) for testing CNN in the first experiment, while the latter was used for training CNN and the former for testing CNN in the second experiment.
The error rate was 15.15% in the first experiment and 5.14% in the second experiment. This study suggests that age-matched training data make more accurate testing results.
Å°¿öµå
Mandible; Osteoporosis; Panoramic radiograph; Computer
¿ø¹® ¹× ¸µÅ©¾Æ¿ô Á¤º¸
µîÀçÀú³Î Á¤º¸