Machine learning approach to predict medication overuse in migraine patients
Machine learning (ML) is largely used to develop automatic predictors in migraine classification but automatic predictors for medication overuse (MO) in migraine are still in their infancy. Thus, to understand the benefits of ML in MO prediction, we explored an automated predictor to estimate MO ris...
Main Authors: | , , , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Elsevier
2020-01-01
|
Series: | Computational and Structural Biotechnology Journal |
Subjects: | |
Online Access: | http://www.sciencedirect.com/science/article/pii/S2001037020302956 |
id |
doaj-e54009756a794769988ad947af78bbd4 |
---|---|
record_format |
Article |
spelling |
doaj-e54009756a794769988ad947af78bbd42021-01-02T05:08:41ZengElsevierComputational and Structural Biotechnology Journal2001-03702020-01-011814871496Machine learning approach to predict medication overuse in migraine patientsPatrizia Ferroni0Fabio M. Zanzotto1Noemi Scarpato2Antonella Spila3Luisa Fofi4Gabriella Egeo5Alessandro Rullo6Raffaele Palmirotta7Piero Barbanti8Fiorella Guadagni9BioBIM (InterInstitutional Multidisciplinary Biobank), IRCCS San Raffaele Pisana, Via di Val Cannuta 247, 00166 Rome, Italy; Dept. of Human Sciences & Quality of Life Promotion, San Raffaele Roma Open University, Via di Val Cannuta 247, 00166 Rome, ItalyDepartment of Enterprise Engineering, University of Rome “Tor Vergata”, Viale Oxford 81, 00133 Rome, ItalyDept. of Human Sciences & Quality of Life Promotion, San Raffaele Roma Open University, Via di Val Cannuta 247, 00166 Rome, ItalyBioBIM (InterInstitutional Multidisciplinary Biobank), IRCCS San Raffaele Pisana, Via di Val Cannuta 247, 00166 Rome, ItalyHeadache and Pain Unit, Dept. of Neurological, Motor and Sensorial Sciences, IRCCS San Raffaele Pisana, Via di Val Cannuta 247, 00166 Rome, ItalyHeadache and Pain Unit, Dept. of Neurological, Motor and Sensorial Sciences, IRCCS San Raffaele Pisana, Via di Val Cannuta 247, 00166 Rome, ItalyNeatec S.p.A., Via Campi Flegrei, 34, 80078 Pozzuoli, Naples, ItalyDepartment of Biomedical Sciences & Human Oncology, University of Bari ‘Aldo Moro’, Bari, ItalyDept. of Human Sciences & Quality of Life Promotion, San Raffaele Roma Open University, Via di Val Cannuta 247, 00166 Rome, Italy; Headache and Pain Unit, Dept. of Neurological, Motor and Sensorial Sciences, IRCCS San Raffaele Pisana, Via di Val Cannuta 247, 00166 Rome, ItalyBioBIM (InterInstitutional Multidisciplinary Biobank), IRCCS San Raffaele Pisana, Via di Val Cannuta 247, 00166 Rome, Italy; Dept. of Human Sciences & Quality of Life Promotion, San Raffaele Roma Open University, Via di Val Cannuta 247, 00166 Rome, Italy; Corresponding author at: San Raffaele Roma Open University and IRCCS San Raffaele Pisana, Via di Val Cannuta, 247, 00166 Rome, Italy.Machine learning (ML) is largely used to develop automatic predictors in migraine classification but automatic predictors for medication overuse (MO) in migraine are still in their infancy. Thus, to understand the benefits of ML in MO prediction, we explored an automated predictor to estimate MO risk in migraine. To achieve this objective, a study was designed to analyze the performance of a customized ML-based decision support system that combines support vector machines and Random Optimization (RO-MO). We used RO-MO to extract prognostic information from demographic, clinical and biochemical data. Using a dataset of 777 consecutive migraine patients we derived a set of predictors with discriminatory power for MO higher than that observed for baseline SVM. The best four were incorporated into the final RO-MO decision support system and risk evaluation on a five-level stratification was performed. ROC analysis resulted in a c-statistic of 0.83 with a sensitivity and specificity of 0.69 and 0.87, respectively, and an accuracy of 0.87 when MO was predicted by at least three RO-MO models. Logistic regression analysis confirmed that the derived RO-MO system could effectively predict MO with ORs of 5.7 and 21.0 for patients classified as probably (3 predictors positive), or definitely at risk of MO (4 predictors positive), respectively. In conclusion, a combination of ML and RO – taking into consideration clinical/biochemical features, drug exposure and lifestyle – might represent a valuable approach to MO prediction in migraine and holds the potential for improving model precision through weighting the relative importance of attributes.http://www.sciencedirect.com/science/article/pii/S2001037020302956MigraineMedication overuseArtificial intelligenceMachine learningDecision support systems |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Patrizia Ferroni Fabio M. Zanzotto Noemi Scarpato Antonella Spila Luisa Fofi Gabriella Egeo Alessandro Rullo Raffaele Palmirotta Piero Barbanti Fiorella Guadagni |
spellingShingle |
Patrizia Ferroni Fabio M. Zanzotto Noemi Scarpato Antonella Spila Luisa Fofi Gabriella Egeo Alessandro Rullo Raffaele Palmirotta Piero Barbanti Fiorella Guadagni Machine learning approach to predict medication overuse in migraine patients Computational and Structural Biotechnology Journal Migraine Medication overuse Artificial intelligence Machine learning Decision support systems |
author_facet |
Patrizia Ferroni Fabio M. Zanzotto Noemi Scarpato Antonella Spila Luisa Fofi Gabriella Egeo Alessandro Rullo Raffaele Palmirotta Piero Barbanti Fiorella Guadagni |
author_sort |
Patrizia Ferroni |
title |
Machine learning approach to predict medication overuse in migraine patients |
title_short |
Machine learning approach to predict medication overuse in migraine patients |
title_full |
Machine learning approach to predict medication overuse in migraine patients |
title_fullStr |
Machine learning approach to predict medication overuse in migraine patients |
title_full_unstemmed |
Machine learning approach to predict medication overuse in migraine patients |
title_sort |
machine learning approach to predict medication overuse in migraine patients |
publisher |
Elsevier |
series |
Computational and Structural Biotechnology Journal |
issn |
2001-0370 |
publishDate |
2020-01-01 |
description |
Machine learning (ML) is largely used to develop automatic predictors in migraine classification but automatic predictors for medication overuse (MO) in migraine are still in their infancy. Thus, to understand the benefits of ML in MO prediction, we explored an automated predictor to estimate MO risk in migraine. To achieve this objective, a study was designed to analyze the performance of a customized ML-based decision support system that combines support vector machines and Random Optimization (RO-MO). We used RO-MO to extract prognostic information from demographic, clinical and biochemical data. Using a dataset of 777 consecutive migraine patients we derived a set of predictors with discriminatory power for MO higher than that observed for baseline SVM. The best four were incorporated into the final RO-MO decision support system and risk evaluation on a five-level stratification was performed. ROC analysis resulted in a c-statistic of 0.83 with a sensitivity and specificity of 0.69 and 0.87, respectively, and an accuracy of 0.87 when MO was predicted by at least three RO-MO models. Logistic regression analysis confirmed that the derived RO-MO system could effectively predict MO with ORs of 5.7 and 21.0 for patients classified as probably (3 predictors positive), or definitely at risk of MO (4 predictors positive), respectively. In conclusion, a combination of ML and RO – taking into consideration clinical/biochemical features, drug exposure and lifestyle – might represent a valuable approach to MO prediction in migraine and holds the potential for improving model precision through weighting the relative importance of attributes. |
topic |
Migraine Medication overuse Artificial intelligence Machine learning Decision support systems |
url |
http://www.sciencedirect.com/science/article/pii/S2001037020302956 |
work_keys_str_mv |
AT patriziaferroni machinelearningapproachtopredictmedicationoveruseinmigrainepatients AT fabiomzanzotto machinelearningapproachtopredictmedicationoveruseinmigrainepatients AT noemiscarpato machinelearningapproachtopredictmedicationoveruseinmigrainepatients AT antonellaspila machinelearningapproachtopredictmedicationoveruseinmigrainepatients AT luisafofi machinelearningapproachtopredictmedicationoveruseinmigrainepatients AT gabriellaegeo machinelearningapproachtopredictmedicationoveruseinmigrainepatients AT alessandrorullo machinelearningapproachtopredictmedicationoveruseinmigrainepatients AT raffaelepalmirotta machinelearningapproachtopredictmedicationoveruseinmigrainepatients AT pierobarbanti machinelearningapproachtopredictmedicationoveruseinmigrainepatients AT fiorellaguadagni machinelearningapproachtopredictmedicationoveruseinmigrainepatients |
_version_ |
1724359713999552512 |