ÀÇ·á ÀΰøÁö´É¿¡ ´ëÇÑ ´ëÇѹα¹ ¿µ»óÀÇÇаú Àü°øÀÇÀÇ ÀÎ½Ä Á¶»ç ¿¬±¸
Survey of the Knowledge of Korean Radiology Residents on Medical Artificial Intelligence
´ëÇÑ¿µ»óÀÇÇÐȸÁö 2020³â 81±Ç 6È£ p.1397 ~ p.1411
ÀÌÇöºó(Lee Hyeon-Bin) - Korea University Ansan Hospital Department of Radiology
¹Ú¼ºÈ£(Park Seong-Ho) - University of Ulsan College of Medicine Asan Medical Center Department of Radiology
±è丮(Kim Cherry) - Korea University Ansan Hospital Department of Radiology
±è½Â°ü(Kim Seung-Kwan) - Korea University Ansan Hospital Department of Radiology
Â÷ÀçÇü(Cha Jae-Hyung) - Korea University College of Medicine Medical Science Research Center
Abstract
¸ñÀû: ÀÌ ¿¬±¸´Â ÀΰøÁö´É(artificial intelligence; ÀÌÇÏ AI)¿¡ ´ëÇÑ ¿µ»óÀÇÇаú Àü°øÀǵéÀÇ ÀÎ½Ä ¹× ÀÇ°ßÀ» ¾Ë¾Æº¸°íÀÚ ÇÏ¿´´Ù
´ë»ó°ú ¹æ¹ý: 2019³â 6¿ù 4ÀϺÎÅÍ 7ÀϱîÁö AI¿Í °ü·ÃÇÑ 18°³ÀÇ °´°ü½Ä ¹®Ç×°ú 1°³ÀÇ ÁÖ°ü½Ä ¹®Ç×ÀÌ Æ÷ÇÔµÈ ¼³¹®ÀÇ ÀÀ´äÀ» ¹Þ¾Ò´Ù. ¸ðÁýµÈ °á°ú¸¦ ·ÎÁö½ºÆ½ ȸ±ÍºÐ¼®À» ÀÌ¿ëÇÏ¿© Àü°øÀÇ ¿¬Â÷, ¼Ò¼Ó º´¿øÀÇ À§Ä¡ ¹× ±Ô¸ð µîÀÇ ¿äÀο¡ µû¶ó ºÐ¼®ÇÏ¿´´Ù
°á°ú: ÃÑ 101¸í(89.4%)ÀÇ Àü°øÀÇ°¡ ÀÀ´äÇÏ¿´´Ù. AIÀÇ Áö½ÄÀû Ãø¸é¿¡¼ ÀÀ´äÀÚÀÇ 50¸í(49.5%)ÀÌ AI¿¡ ´ëÇØ Æò±Õ ÀÌ»óÀ¸·Î °øºÎÇÏ°í ÀÖÀ¸¸ç, 68¸í(67.3%)ÀÌ AI °ü·Ã ¿ë¾î¿¡ ´ëÇÑ ÀÌÇصµ°¡ Æò±Õ ÀÌ»óÀ̶ó°í ÀÀ´äÇÏ¿´´Ù. ¶ÇÇÑ ¼¿ï ¹× °æ±â Áö¿ª ÀÀ´äÀÚ°¡ ±âŸ Áö¿ª ÀÀ´äÀÚ¿¡ ºñÇÏ¿© AI¿¡ ´ëÇÑ ÀÚ°¡ Æò°¡ ¹× Áö½Ä¼öÁØÀÌ ÀÇ¹Ì ÀÖ°Ô ³ô¾ÒÀ¸¸ç, 4³âÂ÷ Àü°øÀÇ¿¡ ºñÇØ 1~2³âÂ÷ Àü°øÀÇ°¡ AI¿¡ ´ëÇÑ ÀÚ°¡ Æò°¡ ¹× Áö½Ä¼öÁØÀÌ ÀÇ¹Ì ÀÖ°Ô ³·¾Ò´Ù. AI °ü·Ã ¿¬±¸¿¡ Âü¿©Çغ» Àû ÀÖ´Â Àü°øÀÇ´Â 15.8%À̾úÁö¸¸, ÃßÈÄ ¿¬±¸ Âü¿© ÀÇÇâÀÌ ÀÖ´Â Àü°øÀÇ´Â 90%¿¡ ´ÞÇÏ¿´´Ù. Àü°øÀǵéÀº ¶ÇÇÑ ÇÐȸ ÁÖµµÀÇ AI ±³À° ¹× Àû±ØÀû È«º¸¸¦ ¿øÇÏ°í ÀÖ¾ú´Ù.
°á·Ð: ¿µ»óÀÇÇаú Àü°øÀÇÀÇ AI ±³À° ¼ö¿ä¸¦ ÃæÁ·½ÃÅ°°í, ÀÇ·á AI ½Ã´ëÀÇ ¿µ»óÀÇÇаú ÀÇ»çÀÇ ¿ªÇÒÀ» Á¦´ë·Î ¾Ë¸®±â À§ÇØ º¸´Ù ¸¹Àº ÇÐȸ Â÷¿øÀÇ ³ë·ÂÀÌ ¿äûµÈ´Ù.
Purpose: To survey the perception, knowledge, wishes, and expectations of Korean radiology residents regarding artificial intelligence (AI) in radiology.
Materials and Methods: From June 4th to 7th, 2019, questionnaires comprising 19 questions related to AI were distributed to 113 radiology residents. Results were analyzed based on factors such as the year of residency and location and number of beds of the hospital.
Results: A total of 101 (89.4%) residents filled out the questionnaire. Fifty (49.5%) respondents had studied AI harder than the average while 68 (67.3%) had a similar or higher understanding of AI than the average. In addition, the self-evaluation and knowledge level of AI were significantly higher for radiology residents at hospitals located in Seoul and Gyeonggi-do compared to radiology residents at hospitals located in other regions. Furthermore, the self-evaluation and knowledge level of AI were significantly lower in junior residents than in residents in the 4th year of training. Of the 101 respondents, only 16 (15.8%) had experiences in AI-related study while 91 (90%) were willing to participate in AI-related study in the future.
Conclusion: Organizational efforts through a radiology society would be needed to meet the need of radiology trainees for AI education and to promote the role of radiologists more adequately in the era of medical AI.
Å°¿öµå
Surveys and Questionnaires, Artificial Intelligence, Machine Learning
¿ø¹® ¹× ¸µÅ©¾Æ¿ô Á¤º¸
µîÀçÀú³Î Á¤º¸
À¯È¿¼º°á°ú(Recomendation)
A total of 101 (89.4%) residents filled out the questionnaire. Fifty (49.5%) respondents had studied AI harder than the average while 68 (67.3%) had a similar or higher understanding of AI than the average. In addition, the self-evaluation and knowledge level of AI were significantly higher for radiology residents at hospitals located in Seoul and Gyeonggi-do compared to radiology residents at hospitals located in other regions.