Erschienen in:
23.08.2023 | ASO Author Reflections
ASO Author Reflections: Machine Learning-Based Preoperative Prediction of Pancreatic Fistula after Pancreaticoduodenectomy
verfasst von:
Amir Ashraf Ganjouei, MD, MPH, Jaeyun Jane Wang, MD, Fernanda Romero-Hernandez, MD, Adnan Alseidi, MD, EdM, Mohamed A. Adam, MD
Erschienen in:
Annals of Surgical Oncology
|
Ausgabe 12/2023
Einloggen, um Zugang zu erhalten
Excerpt
Pancreatoduodenectomy (PD) is the primary curative treatment for periampullary cancers. However, postoperative pancreatic fistula (POPF) remains a significant complication that can lead to considerable morbidity, increased mortality, and delay or omission of adjuvant therapy.
1 Existing calculators predicting POPF have limitations, such as small sample sizes from single institutions and reliance on intraoperative and postoperative variables.
2,3 As a result, there is a need to develop innovative models that can better predict outcomes after PD by using preoperatively known variables. Machine learning (ML) algorithms have the potential of predicting outcomes from highly dimensional data with enhanced accuracy and using a smaller number of variables. …