Objective: Establishing a prognostic model for endometrial cancer (EC), that individualizes risk and management plan per patient and disease characteristics. Methods: this is multicentre retrospective study conducted in 9 European gynaecologic cancer centres. Women with confirmed EC between January 2008 to December 2015 were included. Demographics, disease characteristics, management, and follow-up information were collected. Cancer-specific survival (CSS) and disease-free survival (DFS) at 3 and 5 years comprise the primary outcomes of the study. Machine learning algorithms were applied to patient and disease characteristics. Model I: pre-treatment model. Calculated probability was added to management variables (model II: treatment model), and the second calculated probability was added to perioperative and postoperative variables (model III). Results: Out of 1,150 women, 1,144 were eligible for 3-year survival analysis and 860 for 5-year survival analysis. Model I, II, and III accuracies of prediction of 5-year CSS were 84.88%/85.47% (in train and test sets), 85.47%/84.88% and 87.35%/86.05%, respectively. Model I predicted 3-year CSS at an accuracy of 91.34%/87.02%. Accuracy of model I, II and III in predicting 5-year DFS were 74.63%/76.72%, 77.03%/76.72%, and 80.61%/77.78%, respectively. Conclusion: Endometrial Cancer Individualised Scoring System (ECISS) is a novel machine learning tool assessing patient-specific survival probability with high accuracy.

Endometrial Cancer Individualised Scoring System (ECISS): A machine learning-based prediction model of endometrial cancer prognosis / Shazly, Sherif A; Coronado, Pluvio J; Yılmaz, Ercan; Melekoglu, Rauf; Sahin, Hanifi; Giannella, Luca; Ciavattini, Andrea; Delli Carpini, Giovanni; Di Giuseppe, Jacopo; Yordanov, Angel; Karakadieva, Konstantina; Nedelcheva, Nevena Milenova; Vasileva-Slaveva, Mariela; Alcazar, Juan Luis; Chacon, Enrique; Manzour, Nabil; Vara, Julio; Karaman, Erbil; Karaaslan, Onur; Hacıoğlu, Latif; Korkmaz, Duygu; Onal, Cem; Knez, Jure; Ferrari, Federico; Hosni, Esraa M; Mahmoud, Mohamed E; Elassall, Gena M; Abdo, Mohamed S; Mohamed, Yasmin I; Abdelbadie, Amr S. - In: INTERNATIONAL JOURNAL OF GYNECOLOGY & OBSTETRICS. - ISSN 0020-7292. - (2022). [10.1002/ijgo.14639]

Endometrial Cancer Individualised Scoring System (ECISS): A machine learning-based prediction model of endometrial cancer prognosis

Giannella, Luca;Ciavattini, Andrea;Delli Carpini, Giovanni;Di Giuseppe, Jacopo;
2022-01-01

Abstract

Objective: Establishing a prognostic model for endometrial cancer (EC), that individualizes risk and management plan per patient and disease characteristics. Methods: this is multicentre retrospective study conducted in 9 European gynaecologic cancer centres. Women with confirmed EC between January 2008 to December 2015 were included. Demographics, disease characteristics, management, and follow-up information were collected. Cancer-specific survival (CSS) and disease-free survival (DFS) at 3 and 5 years comprise the primary outcomes of the study. Machine learning algorithms were applied to patient and disease characteristics. Model I: pre-treatment model. Calculated probability was added to management variables (model II: treatment model), and the second calculated probability was added to perioperative and postoperative variables (model III). Results: Out of 1,150 women, 1,144 were eligible for 3-year survival analysis and 860 for 5-year survival analysis. Model I, II, and III accuracies of prediction of 5-year CSS were 84.88%/85.47% (in train and test sets), 85.47%/84.88% and 87.35%/86.05%, respectively. Model I predicted 3-year CSS at an accuracy of 91.34%/87.02%. Accuracy of model I, II and III in predicting 5-year DFS were 74.63%/76.72%, 77.03%/76.72%, and 80.61%/77.78%, respectively. Conclusion: Endometrial Cancer Individualised Scoring System (ECISS) is a novel machine learning tool assessing patient-specific survival probability with high accuracy.
2022
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11566/309707
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? 0
  • Scopus 5
  • ???jsp.display-item.citation.isi??? 5
social impact