ÀΰøÁö´É ±â¹Ý ÀÓ»óÀÇÇÐ °áÁ¤ Áö¿ø ½Ã½ºÅÛ ÀÇ·á±â±âÀÇ ¼º´É ¹× ¾ÈÀü¼º °ËÁõÀ» À§ÇÑ °£ Á¾¾ç Ç¥ÁØ µ¥ÀÌÅͼ ±¸Ãà
Construction of a Standard Dataset for Liver Tumors for Testing the Performance and Safety of Artificial Intelligence-Based Clinical Decision Support Systems
´ëÇÑ¿µ»óÀÇÇÐȸÁö 2021³â 82±Ç 5È£ p.1196 ~ p.1206
±è½Â¼·(Kim Seung-Seob) - Yonsei University College of Medicine Severance Hospital Department of Radiology
À̵¿È£(Lee Dong-Ho) - Seoul National University College of Medicine Seoul National University Hospital Department of Radiology
À̹οì(Lee Min-Woo) - Sungkyunkwan University School of Medicine Samsung Medical Center Department of Radiology
±è¼Ò¿¬(Kim So-Yeon) - University of Ulsan College of Medicine Asan Medical Center Department of Radiology
½ÅÀç½Â(Shin Jae-Seung) - Yonsei University College of Medicine Severance Hospital Department of Radiology
ÃÖÁø¿µ(Choi Jin-Young) - Yonsei University College of Medicine Severance Hospital Department of Radiology
ÃÖº´¿í(Choi Byoung-Wook) - Yonsei University College of Medicine Severance Hospital Department of Radiology
Abstract
¸ñÀû: °£ Á¾¾çÀÇ Á¶¿µÁõ° ÄÄÇ»ÅÍ´ÜÃþÃÔ¿µ(ÀÌÇÏ CT) ¿µ»ó¿¡ °üÇÑ ÀΰøÁö´É ¾Ë°í¸®ÁòÀÇ ¼º´É°ú ¾ÈÀü¼ºÀ» °ËÁõÇÒ ¼ö Àִ ǥÁØ Å×½ºÆà µ¥ÀÌÅͼÂÀ» ±¸ÃàÇÏ°íÀÚ ÇÏ¿´´Ù.
´ë»ó°ú ¹æ¹ý: ±¹³» 4°³ 3Â÷ ÀÇ·á±â°üÀÇ º¹ºÎ ¿µ»óÀÇÇÐ Àü¹®°¡ 4ÀÎÀÌ ¸ð¿© °£ Á¾¾ç Áø´Ü ¾Ë°í¸®ÁòÀÇ ¼º´É°ú ¾ÈÀü¼ºÀ» °ËÁõÇϱâ À§ÇØ Ç¥ÁØ µ¥ÀÌÅͼÂÀÌ °®Ãç¾ß ÇÒ Á¶°ÇÀ» ³íÀÇÇÏ¿´´Ù. °¢ ±â°ü¸¶´Ù °£¼¼Æ÷¾Ï 75¿¹, ÀüÀÌ¾Ï 75¿¹, ±×¸®°í ¾ç¼º º´º¯ 30-50¿¹¾¿ ¼öÁýÇÏ¿©, ÃÑ 783¸í ȯÀÚÀÇ CT ¿µ»óÀ» ´ë»óÀ¸·Î ÇÏ¿´´Ù. °£¼¼Æ÷¾Ï°ú ÀüÀ̾ÏÀÇ °æ¿ì º´¸®ÇÐÀûÀ¸·Î È®ÁøµÈ °æ¿ì¸¸À» ´ë»óÀ¸·Î ÇÏ¿´´Ù. °¢ ±â°üÀÇ º¹ºÎ ¿µ»óÀÇÇÐ Àü¹®°¡µéÀÌ Á÷Á¢ ȯÀÚÀÇ ÀÓ»óÁ¤º¸¸¦ ÃßÃâÇÏ°í CT ¿µ»ó¿¡ °üÇÑ µ¥ÀÌÅÍ ¶óº§¸µ(labeling)À» ¼ö±â·Î ½ÃÇàÇÏ¿´´Ù. CT ¿µ»óÀº ÀÇ·á¿ë µðÁöÅÐ ¿µ»ó ¹× Åë½Å(Digital Imaging and Communications in Medicine, DICOM) ÆÄÀÏ·Î ÀúÀåÇÏ¿´´Ù.
°á°ú: º¹ºÎ ¿µ»óÀÇÇÐ Àü¹®°¡µéÀÌ ¼ö±â µ¥ÀÌÅÍ ¶óº§¸µÀ» ½ÃÇàÇÑ ÃÑ 783 Áõ·ÊÀÇ °£ Á¾¾ç Á¶¿µÁõ° CTÀÇ Ç¥ÁØ µ¥ÀÌÅͼÂÀ» ±¸ÃàÇÏ¿´´Ù. ¾Ë°í¸®ÁòÀÇ ¼º´É ¹× ¾ÈÀü¼ºÀº º´º¯ÀÇ ¹ß°ß ¿©ºÎ ¹× Ư¼ºÈÀÇ Á¤È®µµ¿¡ ´ëÇØ ¹Î°¨µµ¿Í ƯÀ̵µ¸¦ °è»êÇÏ¿© Æò°¡ÇÒ ¼ö ÀÖ´Ù.
°á·Ð: º» ¿¬±¸¿¡¼ ±¸ÃàÇÑ °£ Á¾¾ç Á¶¿µÁõ° CT ¿µ»óÀÇ Ç¥ÁØ µ¥ÀÌÅͼÂÀº ÀÓ»óÀÇÇÐ °áÁ¤ Áö¿ø ½Ã½ºÅÛÀ» À§ÇÑ ±â°èÇнÀ ±â¹Ý ÀΰøÁö´É ¾Ë°í¸®ÁòÀ» Æò°¡ÇÏ´Â µ¥¿¡ È°¿ëµÉ ¼ö ÀÖ´Ù.
Purpose: To construct a standard dataset of contrast-enhanced CT images of liver tumors to test the performance and safety of artificial intelligence (AI)-based algorithms for clinical decision support systems (CDSSs).
Materials and Methods: A consensus group of medical experts in gastrointestinal radiology from four national tertiary institutions discussed the conditions to be included in a standard dataset. Seventy-five cases of hepatocellular carcinoma, 75 cases of metastasis, and 30-50 cases of benign lesions were retrieved from each institution, and the final dataset consisted of 300 cases of hepatocellular carcinoma, 300 cases of metastasis, and 183 cases of benign lesions. Only pathologically confirmed cases of hepatocellular carcinomas and metastases were enrolled. The medical experts retrieved the medical records of the patients and manually labeled the CT images. The CT images were saved as Digital Imaging and Communications in Medicine (DICOM) files.
Results: The medical experts in gastrointestinal radiology constructed the standard dataset of contrast-enhanced CT images for 783 cases of liver tumors. The performance and safety of the AI algorithm can be evaluated by calculating the sensitivity and specificity for detecting and characterizing the lesions.
Conclusion: The constructed standard dataset can be utilized for evaluating the machine-learningbased AI algorithm for CDSS.
Å°¿öµå
Liver Neoplasms, Aritificial Intelligence, Machine Learning, Deep Learning, Datasets as Topic
¿ø¹® ¹× ¸µÅ©¾Æ¿ô Á¤º¸
µîÀçÀú³Î Á¤º¸
À¯È¿¼º°á°ú(Recomendation)
The medical experts in gastrointestinal radiology constructed the standard dataset of contrast-enhanced CT images for 783 cases of liver tumors.