External validation of risk prediction platforms for pancreatic fistula after pancreatoduodenectomy using nomograms and artificial intelligence
Annals of Surgical Treatment and Research 2022³â 102±Ç 3È£ p.147 ~ p.152
À±¼ÒÁ¤(Yoon So-Jeong) - Sungkyunkwan University School of Medicine Samsung Medical Center Department of Surgery
±Ç¿ìÀÏ(Kwon Woo-Il) - Seoul National University College of Medicine Seoul National University Hospital Department of Surgery
ÀÌ¿ÁÁÖ(Lee Ok-Joo) - Sungkyunkwan University School of Medicine Samsung Medical Center Department of Surgery
Á¤ÁöÇý(Jung Ji-Hye) - Sungkyunkwan University School of Medicine Samsung Medical Center Department of Surgery
½Å¿ëÂù(Shin Yong-Chan) - Inje University College of Medicine Ilsan Paik Hospital Department of Surgery
ÀÓâ¼·(Lim Chang-Sup) - Seoul Metropolitan Government-Seoul National University Boramae Medical Center Department of Surgery
±èÈ«¹ü(Kim Hong-Beom) - Seoul National University College of Medicine Seoul National University Hospital Department of Surgery
ÀåÁø¿µ(Jang Jin-Young) - Seoul National University College of Medicine Seoul National University Hospital Department of Surgery
½Å»óÇö(Shin Sang-Hyun) - Sungkyunkwan University School of Medicine Samsung Medical Center Department of Surgery
ÇãÁø¼®(Heo Jin-Seok) - Sungkyunkwan University School of Medicine Samsung Medical Center Department of Surgery
ÇÑÀοõ(Han In-Woong) - Sungkyunkwan University School of Medicine Samsung Medical Center Department of Surgery
Abstract
Purpose: Postoperative pancreatic fistula (POPF) is a life-threatening complication following pancreatoduodenectomy (PD). We previously developed nomogram- and artificial intelligence (AI)-based risk prediction platforms for POPF after PD. This study aims to externally validate these platforms.
Methods: Between January 2007 and December 2016, a total of 1,576 patients who underwent PD in Seoul National University Hospital, Ilsan Paik Hospital, and Boramae Medical Center were retrospectively reviewed. The individual risk scores for POPF were calculated using each platform by Samsung Medical Center. The predictive ability was evaluated using a receiver operating characteristic curve and the area under the curve (AUC). The optimal predictive value was obtained via backward elimination in accordance with the results from the AI development process.
Results: The AUC of the nomogram after external validation was 0.679 (P < 0.001). The values of AUC after backward elimination in the AI model varied from 0.585 to 0.672. A total of 13 risk factors represented the maximal AUC of 0.672 (P < 0.001).
Conclusion: We performed external validation of previously developed platforms for predicting POPF. Further research is needed to investigate other potential risk factors and thereby improve the predictability of the platform.
Å°¿öµå
Artificial intelligence, Nomograms, Pancreatic fistula, Pancreatoduodenectomy, Postoperative complications
¿ø¹® ¹× ¸µÅ©¾Æ¿ô Á¤º¸
µîÀçÀú³Î Á¤º¸
À¯È¿¼º°á°ú(Recomendation)
The AUC of the nomogram after external validation was 0.679 (P < 0.001). The values of AUC after backward elimination in the AI model varied from 0.585 to 0.672. A total of 13 risk factors represented the maximal AUC of 0.672 (P < 0.001).