Copyright
©The Author(s) 2020.
World J Gastroenterol. Aug 14, 2020; 26(30): 4453-4464
Published online Aug 14, 2020. doi: 10.3748/wjg.v26.i30.4453
Published online Aug 14, 2020. doi: 10.3748/wjg.v26.i30.4453
Variables | Values | Variables | Values |
Age (yr) | 67.7 ± 10.1 | Preoperative ERBD/ENBD (n, %) | 470 (26.6) |
Sex (male/female) | 1079: 690 | Preoperative PTBD (n, %) | 254 (14.4) |
Body mass index (kg/m2) | 22.5 ± 3.8 | Preoperative ERPD (n, %) | 19 (1.1) |
Heart disease including | 735 (41.5) | Neoadjuvant therapy (RT/chemotherapy/CCRT/No) | 2: 6: 16: 1745 |
Diabetes mellitus (n, %) | 472 (26.7) | Operative time (min) | 443.2 ± 90.1 |
Pulmonary disease (n, %) | 153 (8.6) | †Intraoperative fluid infusion (mL) | 3129 ± 3495 |
Liver disease (n, %) | 99 (5.6) | Intraoperative transfusion (n, %) | 171 (9.7) |
Cerebrovascular disease (n, %) | 73 (4.1) | Estimated blood loss (ml) | 962.4 ± 665.1 |
Chronic kidney disease (n, %) | 17 (1.0) | Soft pancreas (n, %) | 750 (43.2) |
ASA score (1-4) | 372:1267: 128: 2 | Pancreatic duct diameter (mm) | 4.2 ± 2.8 |
White blood cell count (x10³/μL) | 6.6 ± 2.3 | Type of surgery (PPPD/PRPD/PD) | 1254: 244: 271 |
Hemoglobin (g/dL) | 10.9 ± 1.5 | Combined organ resection (n, %) | 67 (3.8) |
Platelet count (x10³/μL) | 280.2 ± 52.7 | Combined vascular resection(n, %) | 188 (10.6) |
Albumin (d/dL) | 3.5 ± 0.4 | C- reactive protein (mg/dL) | 1.9 ± 10.9 |
Total bilirubin level (mg/dL) | 3.5 ± 4.3 | CA 19-9 (U/mL) | 1786.5 ± 7141.5 |
Combined portal vein resection (n, %) | 175 (9.9) | P-duct stent (Internal/external/none) | 1051: 134: 584 |
Amylase (U/L) | 94.5 ± 5586.9 | Co-existing pancreatitis (n, %) | 370 (20.9) |
Lipase (U/L) | 160.5 ± 277.0 | ‡Location of tumor (Pancreas/others) | 856: 913 |
Anastomotic methods (1) (Duct-to-mucosa/Dunkin) | 1756: 13 | Anastomotic methods (2) (P-J/P-G/Others) | 1761: 8: 0 |
Dataset and algorithm | Number of original variables | Number of encoded variables | Number of samples | Area under curve |
Random forest with complete cases | 38 | 44 | 889 | 0.670.02 |
Neural network with complete cases | 0.740.02 | |||
Random forest with complete variables | 34 | 40 | 1769 | 0.670.01 |
Neural network with complete variables | 0.720.02 | |||
Random forest with missing data treatment | 38 | 44 | 1769 | 0.680.02 |
Neural network with missing data treatment | 0.710.02 |
- Citation: Han IW, Cho K, Ryu Y, Shin SH, Heo JS, Choi DW, Chung MJ, Kwon OC, Cho BH. Risk prediction platform for pancreatic fistula after pancreatoduodenectomy using artificial intelligence. World J Gastroenterol 2020; 26(30): 4453-4464
- URL: https://www.wjgnet.com/1007-9327/full/v26/i30/4453.htm
- DOI: https://dx.doi.org/10.3748/wjg.v26.i30.4453