Development and validation of a machine learning-based predictive model to improve the prediction of inguinal status of anal cancer patients

A preliminary report

Berardino De Bari, Mauro Vallati, Roberto Gatta, Laëtitia Lestrade, Stefania Manfrida, Christian Carrie, Vincenzo Valentini

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Introduction: The role of prophylactic inguinal irradiation (PII) in the treatment of anal cancer patients is controversial. We developped an innovative algorithm based on the Machine Learning (ML) allowing the tailoring of the prescription of PII. Results: Once verified on the independent testing set, J48 showed the better performances, with specificity, sensitivity, and accuracy rates in predicting relapsing patients of 86.4%, 50.0% and 83.1% respectively (vs 36.5%, 90.4% and 80.25%, respectively, for LR). Methods: We classified 194 anal cancer patients with Logistic Regression (LR) and other 3 ML techniques based on decision trees (J48, Random Tree and Random Forest), using a large set of clinical and therapeutic variables. We tested obtained ML algorithms on an independent testing set of 65 anal cancer patients. TRIPOD (Transparent Reporting of a multivariable prediction model for Individual Prognosis or Diagnosis) methodology was used for the development, the Quality Assurance and the description of the experimental procedures. Conclusion: In an internationally approved quality assurance framework, ML seems promising in predicting the outcome of patients that would benefit or not of the PII. Once confirmed in larger and/or multi-centric databases, ML could support the physician in tailoring the treatment and in deciding if deliver or not the PII.

Original languageEnglish
Pages (from-to)108509-108521
Number of pages13
JournalOncotarget
Volume8
Issue number65
DOIs
Publication statusPublished - 21 Jul 2017

Fingerprint

Anus Neoplasms
Groin
Logistic Models
Decision Trees
Prescriptions
Therapeutics
Machine Learning
Databases
Physicians
Sensitivity and Specificity

Cite this

De Bari, Berardino ; Vallati, Mauro ; Gatta, Roberto ; Lestrade, Laëtitia ; Manfrida, Stefania ; Carrie, Christian ; Valentini, Vincenzo. / Development and validation of a machine learning-based predictive model to improve the prediction of inguinal status of anal cancer patients : A preliminary report. In: Oncotarget. 2017 ; Vol. 8, No. 65. pp. 108509-108521.
@article{9d8d8306da9347f1ba0ced7d102782db,
title = "Development and validation of a machine learning-based predictive model to improve the prediction of inguinal status of anal cancer patients: A preliminary report",
abstract = "Introduction: The role of prophylactic inguinal irradiation (PII) in the treatment of anal cancer patients is controversial. We developped an innovative algorithm based on the Machine Learning (ML) allowing the tailoring of the prescription of PII. Results: Once verified on the independent testing set, J48 showed the better performances, with specificity, sensitivity, and accuracy rates in predicting relapsing patients of 86.4{\%}, 50.0{\%} and 83.1{\%} respectively (vs 36.5{\%}, 90.4{\%} and 80.25{\%}, respectively, for LR). Methods: We classified 194 anal cancer patients with Logistic Regression (LR) and other 3 ML techniques based on decision trees (J48, Random Tree and Random Forest), using a large set of clinical and therapeutic variables. We tested obtained ML algorithms on an independent testing set of 65 anal cancer patients. TRIPOD (Transparent Reporting of a multivariable prediction model for Individual Prognosis or Diagnosis) methodology was used for the development, the Quality Assurance and the description of the experimental procedures. Conclusion: In an internationally approved quality assurance framework, ML seems promising in predicting the outcome of patients that would benefit or not of the PII. Once confirmed in larger and/or multi-centric databases, ML could support the physician in tailoring the treatment and in deciding if deliver or not the PII.",
keywords = "Anal canal cancer, Machine learning, Predicitive models, Prophylactic inguinal irradiation, Radiochemotherapy",
author = "{De Bari}, Berardino and Mauro Vallati and Roberto Gatta and La{\"e}titia Lestrade and Stefania Manfrida and Christian Carrie and Vincenzo Valentini",
year = "2017",
month = "7",
day = "21",
doi = "10.18632/oncotarget.10749",
language = "English",
volume = "8",
pages = "108509--108521",
journal = "Oncotarget",
issn = "1949-2553",
publisher = "Impact Journals",
number = "65",

}

Development and validation of a machine learning-based predictive model to improve the prediction of inguinal status of anal cancer patients : A preliminary report. / De Bari, Berardino; Vallati, Mauro; Gatta, Roberto; Lestrade, Laëtitia; Manfrida, Stefania; Carrie, Christian; Valentini, Vincenzo.

In: Oncotarget, Vol. 8, No. 65, 21.07.2017, p. 108509-108521.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Development and validation of a machine learning-based predictive model to improve the prediction of inguinal status of anal cancer patients

T2 - A preliminary report

AU - De Bari, Berardino

AU - Vallati, Mauro

AU - Gatta, Roberto

AU - Lestrade, Laëtitia

AU - Manfrida, Stefania

AU - Carrie, Christian

AU - Valentini, Vincenzo

PY - 2017/7/21

Y1 - 2017/7/21

N2 - Introduction: The role of prophylactic inguinal irradiation (PII) in the treatment of anal cancer patients is controversial. We developped an innovative algorithm based on the Machine Learning (ML) allowing the tailoring of the prescription of PII. Results: Once verified on the independent testing set, J48 showed the better performances, with specificity, sensitivity, and accuracy rates in predicting relapsing patients of 86.4%, 50.0% and 83.1% respectively (vs 36.5%, 90.4% and 80.25%, respectively, for LR). Methods: We classified 194 anal cancer patients with Logistic Regression (LR) and other 3 ML techniques based on decision trees (J48, Random Tree and Random Forest), using a large set of clinical and therapeutic variables. We tested obtained ML algorithms on an independent testing set of 65 anal cancer patients. TRIPOD (Transparent Reporting of a multivariable prediction model for Individual Prognosis or Diagnosis) methodology was used for the development, the Quality Assurance and the description of the experimental procedures. Conclusion: In an internationally approved quality assurance framework, ML seems promising in predicting the outcome of patients that would benefit or not of the PII. Once confirmed in larger and/or multi-centric databases, ML could support the physician in tailoring the treatment and in deciding if deliver or not the PII.

AB - Introduction: The role of prophylactic inguinal irradiation (PII) in the treatment of anal cancer patients is controversial. We developped an innovative algorithm based on the Machine Learning (ML) allowing the tailoring of the prescription of PII. Results: Once verified on the independent testing set, J48 showed the better performances, with specificity, sensitivity, and accuracy rates in predicting relapsing patients of 86.4%, 50.0% and 83.1% respectively (vs 36.5%, 90.4% and 80.25%, respectively, for LR). Methods: We classified 194 anal cancer patients with Logistic Regression (LR) and other 3 ML techniques based on decision trees (J48, Random Tree and Random Forest), using a large set of clinical and therapeutic variables. We tested obtained ML algorithms on an independent testing set of 65 anal cancer patients. TRIPOD (Transparent Reporting of a multivariable prediction model for Individual Prognosis or Diagnosis) methodology was used for the development, the Quality Assurance and the description of the experimental procedures. Conclusion: In an internationally approved quality assurance framework, ML seems promising in predicting the outcome of patients that would benefit or not of the PII. Once confirmed in larger and/or multi-centric databases, ML could support the physician in tailoring the treatment and in deciding if deliver or not the PII.

KW - Anal canal cancer

KW - Machine learning

KW - Predicitive models

KW - Prophylactic inguinal irradiation

KW - Radiochemotherapy

UR - http://www.scopus.com/inward/record.url?scp=85037687323&partnerID=8YFLogxK

U2 - 10.18632/oncotarget.10749

DO - 10.18632/oncotarget.10749

M3 - Article

VL - 8

SP - 108509

EP - 108521

JO - Oncotarget

JF - Oncotarget

SN - 1949-2553

IS - 65

ER -