Development and validation of a machine learning approach for screening new leprosy cases based on the leprosy suspicion questionnaire

Abstract Leprosy is a dermatoneurological disease and can cause irreversible nerve damage. In addition to being able to mimic different rheumatological, neurological and dermatological diseases, leprosy is underdiagnosed because several professionals present lack of training. The World Health Organi...

Full description

Saved in:
Bibliographic Details
Main Authors: Mateus Mendonça Ramos Simões, Filipe Rocha Lima, Helena Barbosa Lugão, Natália Aparecida de Paula, Cláudia Maria Lincoln Silva, Alexandre Ferreira Ramos, Marco Andrey Cipriani Frade
Format: Article
Language:English
Published: Nature Portfolio 2025-02-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-91462-6
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849766485690417152
author Mateus Mendonça Ramos Simões
Filipe Rocha Lima
Helena Barbosa Lugão
Natália Aparecida de Paula
Cláudia Maria Lincoln Silva
Alexandre Ferreira Ramos
Marco Andrey Cipriani Frade
author_facet Mateus Mendonça Ramos Simões
Filipe Rocha Lima
Helena Barbosa Lugão
Natália Aparecida de Paula
Cláudia Maria Lincoln Silva
Alexandre Ferreira Ramos
Marco Andrey Cipriani Frade
author_sort Mateus Mendonça Ramos Simões
collection DOAJ
description Abstract Leprosy is a dermatoneurological disease and can cause irreversible nerve damage. In addition to being able to mimic different rheumatological, neurological and dermatological diseases, leprosy is underdiagnosed because several professionals present lack of training. The World Health Organization instituted active search for new leprosy cases as one of the four pillars of the zero-leprosy strategy. The Leprosy Suspicion Questionnaire (LSQ) was created aiming to be a screening tool to actively detect new cases; it is composed of 14 simple yes/no questions that can be answered with the help of a health professional or by the very patient themselves. During its development, it was noticed that the combination of marked questions was related to new case detections. To better encapsulate and being able to expand its use, we developed MaLeSQs, a Machine Learning tool whose output may be LSQ Positive when the subject is indicated for being further clinically evaluated or LSQ Negative when the subject does not present any evidence that justify being further evaluated for leprosy. To achieve a reasonable product, we trained four classifiers with different learning paradigms, Support Vectors Machine, Logistic Regression, Random Forest and XGBoost. We compared them based on sensitivity, specificity, positive predicted value, negative predicted value, and area under the ROC curve. After the training process, the Support Vectors Machine was the classifier with the most balanced metrics of 85.7% sensitivity, 69.2% specificity, 18.6% precision, 98.3% negative predicted values and an area under the ROC curve of 0.775, and it was chosen as the MaLeSQs. With Shapley values, we were able to evaluate variable importance and nerve symptoms were considered important to differentiate between subjects that potentially had leprosy from those who did not.
format Article
id doaj-art-b8f192b331d248ac9aa578d5f5e6aa1b
institution DOAJ
issn 2045-2322
language English
publishDate 2025-02-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-b8f192b331d248ac9aa578d5f5e6aa1b2025-08-20T03:04:34ZengNature PortfolioScientific Reports2045-23222025-02-0115111610.1038/s41598-025-91462-6Development and validation of a machine learning approach for screening new leprosy cases based on the leprosy suspicion questionnaireMateus Mendonça Ramos Simões0Filipe Rocha Lima1Helena Barbosa Lugão2Natália Aparecida de Paula3Cláudia Maria Lincoln Silva4Alexandre Ferreira Ramos5Marco Andrey Cipriani Frade6Dermatology Division, Department of Internal Medicine, National Referral Center for Sanitary Dermatology and Hansen’s Disease, Clinical Hospital of the Ribeirão Preto Medical School, University of São PauloDermatology Division, Department of Internal Medicine, National Referral Center for Sanitary Dermatology and Hansen’s Disease, Clinical Hospital of the Ribeirão Preto Medical School, University of São PauloDermatology Division, Department of Internal Medicine, National Referral Center for Sanitary Dermatology and Hansen’s Disease, Clinical Hospital of the Ribeirão Preto Medical School, University of São PauloDermatology Division, Department of Internal Medicine, National Referral Center for Sanitary Dermatology and Hansen’s Disease, Clinical Hospital of the Ribeirão Preto Medical School, University of São PauloDermatology Division, Department of Internal Medicine, National Referral Center for Sanitary Dermatology and Hansen’s Disease, Clinical Hospital of the Ribeirão Preto Medical School, University of São PauloArts, Science and Humanities School, University of São PauloDermatology Division, Department of Internal Medicine, National Referral Center for Sanitary Dermatology and Hansen’s Disease, Clinical Hospital of the Ribeirão Preto Medical School, University of São PauloAbstract Leprosy is a dermatoneurological disease and can cause irreversible nerve damage. In addition to being able to mimic different rheumatological, neurological and dermatological diseases, leprosy is underdiagnosed because several professionals present lack of training. The World Health Organization instituted active search for new leprosy cases as one of the four pillars of the zero-leprosy strategy. The Leprosy Suspicion Questionnaire (LSQ) was created aiming to be a screening tool to actively detect new cases; it is composed of 14 simple yes/no questions that can be answered with the help of a health professional or by the very patient themselves. During its development, it was noticed that the combination of marked questions was related to new case detections. To better encapsulate and being able to expand its use, we developed MaLeSQs, a Machine Learning tool whose output may be LSQ Positive when the subject is indicated for being further clinically evaluated or LSQ Negative when the subject does not present any evidence that justify being further evaluated for leprosy. To achieve a reasonable product, we trained four classifiers with different learning paradigms, Support Vectors Machine, Logistic Regression, Random Forest and XGBoost. We compared them based on sensitivity, specificity, positive predicted value, negative predicted value, and area under the ROC curve. After the training process, the Support Vectors Machine was the classifier with the most balanced metrics of 85.7% sensitivity, 69.2% specificity, 18.6% precision, 98.3% negative predicted values and an area under the ROC curve of 0.775, and it was chosen as the MaLeSQs. With Shapley values, we were able to evaluate variable importance and nerve symptoms were considered important to differentiate between subjects that potentially had leprosy from those who did not.https://doi.org/10.1038/s41598-025-91462-6LeprosyLeprosy suspicion questionnaireMachine learningScreeningActive search
spellingShingle Mateus Mendonça Ramos Simões
Filipe Rocha Lima
Helena Barbosa Lugão
Natália Aparecida de Paula
Cláudia Maria Lincoln Silva
Alexandre Ferreira Ramos
Marco Andrey Cipriani Frade
Development and validation of a machine learning approach for screening new leprosy cases based on the leprosy suspicion questionnaire
Scientific Reports
Leprosy
Leprosy suspicion questionnaire
Machine learning
Screening
Active search
title Development and validation of a machine learning approach for screening new leprosy cases based on the leprosy suspicion questionnaire
title_full Development and validation of a machine learning approach for screening new leprosy cases based on the leprosy suspicion questionnaire
title_fullStr Development and validation of a machine learning approach for screening new leprosy cases based on the leprosy suspicion questionnaire
title_full_unstemmed Development and validation of a machine learning approach for screening new leprosy cases based on the leprosy suspicion questionnaire
title_short Development and validation of a machine learning approach for screening new leprosy cases based on the leprosy suspicion questionnaire
title_sort development and validation of a machine learning approach for screening new leprosy cases based on the leprosy suspicion questionnaire
topic Leprosy
Leprosy suspicion questionnaire
Machine learning
Screening
Active search
url https://doi.org/10.1038/s41598-025-91462-6
work_keys_str_mv AT mateusmendoncaramossimoes developmentandvalidationofamachinelearningapproachforscreeningnewleprosycasesbasedontheleprosysuspicionquestionnaire
AT filiperochalima developmentandvalidationofamachinelearningapproachforscreeningnewleprosycasesbasedontheleprosysuspicionquestionnaire
AT helenabarbosalugao developmentandvalidationofamachinelearningapproachforscreeningnewleprosycasesbasedontheleprosysuspicionquestionnaire
AT nataliaaparecidadepaula developmentandvalidationofamachinelearningapproachforscreeningnewleprosycasesbasedontheleprosysuspicionquestionnaire
AT claudiamarialincolnsilva developmentandvalidationofamachinelearningapproachforscreeningnewleprosycasesbasedontheleprosysuspicionquestionnaire
AT alexandreferreiraramos developmentandvalidationofamachinelearningapproachforscreeningnewleprosycasesbasedontheleprosysuspicionquestionnaire
AT marcoandreyciprianifrade developmentandvalidationofamachinelearningapproachforscreeningnewleprosycasesbasedontheleprosysuspicionquestionnaire