The construction of HMME-PDT efficacy prediction model for port-wine stain based on machine learning algorithms

Abstract Hematoporphyrin monomethyl ether-photodynamic therapy (HMME-PDT) is a safe and effective treatment for port-wine stain (PWS). Comprehensive methods for predicting HMME-PDT efficacy based on clinical factors are lacking. This study aims to develop and validate two machine learning models to...

Full description

Saved in:
Bibliographic Details
Main Authors: Hongxia Yan, Yixin Tan, Fan Qiao, Zhuotong Zeng, Yaqian Shi, Xueqin Zhang, Lu Li, Ting Zeng, Yi Zhan, Ruixuan You, Xinglan He, Rong Xiao, Xiangning Qiu
Format: Article
Language:English
Published: Nature Portfolio 2025-07-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-06589-3
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849238656806551552
author Hongxia Yan
Yixin Tan
Fan Qiao
Zhuotong Zeng
Yaqian Shi
Xueqin Zhang
Lu Li
Ting Zeng
Yi Zhan
Ruixuan You
Xinglan He
Rong Xiao
Xiangning Qiu
author_facet Hongxia Yan
Yixin Tan
Fan Qiao
Zhuotong Zeng
Yaqian Shi
Xueqin Zhang
Lu Li
Ting Zeng
Yi Zhan
Ruixuan You
Xinglan He
Rong Xiao
Xiangning Qiu
author_sort Hongxia Yan
collection DOAJ
description Abstract Hematoporphyrin monomethyl ether-photodynamic therapy (HMME-PDT) is a safe and effective treatment for port-wine stain (PWS). Comprehensive methods for predicting HMME-PDT efficacy based on clinical factors are lacking. This study aims to develop and validate two machine learning models to predict the therapeutic effect of HMME-PDT for PWS. We conducted a retrospective study of 131 facial PWS patients treated with single HMME-PDT at the Second Xiangya Hospital from May 2022 to January 2025. The patients were divided into the training cohort and the validation cohort based on the order of their enrollment. Key clinical features were selected using recursive feature elimination (RFE). We developed and validated prediction models with Extreme Gradient Boosting (XGBoost) and Random Forest (RF) algorithms. Model performance was assessed using confusion matrix and evaluation metrics. RFE identified the top predictive factors: dermoscopy vascular pattern, immediate fluorescence intensity (IFI) after HMME-PDT, the facial port-wine stain area and severity index score, and age. In the training cohort, both models demonstrated strong predictive performance, with accuracies, F1 scores, and AUC values exceeding 0.8. The XGBoost model outperformed with an accuracy of 0.8750, F1 score of 0.8750, and AUC of 0.8636. In the validation cohort, XGBoost model achieved an accuracy and F1 score both greater than 0.73, with an AUC value of 0.7672. It had the better comprehensive performance. Our findings suggest these models are promising for predicting HMME-PDT efficacy in PWS. This is the first study to explore IFI after HMME-PDT in efficacy assessment.
format Article
id doaj-art-e865646f077b49289018dbca3f0b5850
institution Kabale University
issn 2045-2322
language English
publishDate 2025-07-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-e865646f077b49289018dbca3f0b58502025-08-20T04:01:26ZengNature PortfolioScientific Reports2045-23222025-07-0115111310.1038/s41598-025-06589-3The construction of HMME-PDT efficacy prediction model for port-wine stain based on machine learning algorithmsHongxia Yan0Yixin Tan1Fan Qiao2Zhuotong Zeng3Yaqian Shi4Xueqin Zhang5Lu Li6Ting Zeng7Yi Zhan8Ruixuan You9Xinglan He10Rong Xiao11Xiangning Qiu12Department of Dermatology, The Second Xiangya Hospital, Central South UniversityDepartment of Dermatology, The Second Xiangya Hospital, Central South UniversityDepartment of Dermatology, The Second Xiangya Hospital, Central South UniversityDepartment of Dermatology, The Second Xiangya Hospital, Central South UniversityDepartment of Dermatology, The Second Xiangya Hospital, Central South UniversityDepartment of Dermatology, The Second Xiangya Hospital, Central South UniversityDepartment of Dermatology, The Second Xiangya Hospital, Central South UniversityDepartment of Dermatology, The Second Xiangya Hospital, Central South UniversityDepartment of Dermatology, The Second Xiangya Hospital, Central South UniversityDepartment of Dermatology, The Second Xiangya Hospital, Central South UniversityDepartment of Dermatology, The Second Xiangya Hospital, Central South UniversityDepartment of Dermatology, The Second Xiangya Hospital, Central South UniversityDepartment of Dermatology, The Second Xiangya Hospital, Central South UniversityAbstract Hematoporphyrin monomethyl ether-photodynamic therapy (HMME-PDT) is a safe and effective treatment for port-wine stain (PWS). Comprehensive methods for predicting HMME-PDT efficacy based on clinical factors are lacking. This study aims to develop and validate two machine learning models to predict the therapeutic effect of HMME-PDT for PWS. We conducted a retrospective study of 131 facial PWS patients treated with single HMME-PDT at the Second Xiangya Hospital from May 2022 to January 2025. The patients were divided into the training cohort and the validation cohort based on the order of their enrollment. Key clinical features were selected using recursive feature elimination (RFE). We developed and validated prediction models with Extreme Gradient Boosting (XGBoost) and Random Forest (RF) algorithms. Model performance was assessed using confusion matrix and evaluation metrics. RFE identified the top predictive factors: dermoscopy vascular pattern, immediate fluorescence intensity (IFI) after HMME-PDT, the facial port-wine stain area and severity index score, and age. In the training cohort, both models demonstrated strong predictive performance, with accuracies, F1 scores, and AUC values exceeding 0.8. The XGBoost model outperformed with an accuracy of 0.8750, F1 score of 0.8750, and AUC of 0.8636. In the validation cohort, XGBoost model achieved an accuracy and F1 score both greater than 0.73, with an AUC value of 0.7672. It had the better comprehensive performance. Our findings suggest these models are promising for predicting HMME-PDT efficacy in PWS. This is the first study to explore IFI after HMME-PDT in efficacy assessment.https://doi.org/10.1038/s41598-025-06589-3Port-wine stainHMME-PDTMachine learning algorithmsEfficacy prediction modelImmediate fluorescence intensity at the lesion site after HMME-PDT
spellingShingle Hongxia Yan
Yixin Tan
Fan Qiao
Zhuotong Zeng
Yaqian Shi
Xueqin Zhang
Lu Li
Ting Zeng
Yi Zhan
Ruixuan You
Xinglan He
Rong Xiao
Xiangning Qiu
The construction of HMME-PDT efficacy prediction model for port-wine stain based on machine learning algorithms
Scientific Reports
Port-wine stain
HMME-PDT
Machine learning algorithms
Efficacy prediction model
Immediate fluorescence intensity at the lesion site after HMME-PDT
title The construction of HMME-PDT efficacy prediction model for port-wine stain based on machine learning algorithms
title_full The construction of HMME-PDT efficacy prediction model for port-wine stain based on machine learning algorithms
title_fullStr The construction of HMME-PDT efficacy prediction model for port-wine stain based on machine learning algorithms
title_full_unstemmed The construction of HMME-PDT efficacy prediction model for port-wine stain based on machine learning algorithms
title_short The construction of HMME-PDT efficacy prediction model for port-wine stain based on machine learning algorithms
title_sort construction of hmme pdt efficacy prediction model for port wine stain based on machine learning algorithms
topic Port-wine stain
HMME-PDT
Machine learning algorithms
Efficacy prediction model
Immediate fluorescence intensity at the lesion site after HMME-PDT
url https://doi.org/10.1038/s41598-025-06589-3
work_keys_str_mv AT hongxiayan theconstructionofhmmepdtefficacypredictionmodelforportwinestainbasedonmachinelearningalgorithms
AT yixintan theconstructionofhmmepdtefficacypredictionmodelforportwinestainbasedonmachinelearningalgorithms
AT fanqiao theconstructionofhmmepdtefficacypredictionmodelforportwinestainbasedonmachinelearningalgorithms
AT zhuotongzeng theconstructionofhmmepdtefficacypredictionmodelforportwinestainbasedonmachinelearningalgorithms
AT yaqianshi theconstructionofhmmepdtefficacypredictionmodelforportwinestainbasedonmachinelearningalgorithms
AT xueqinzhang theconstructionofhmmepdtefficacypredictionmodelforportwinestainbasedonmachinelearningalgorithms
AT luli theconstructionofhmmepdtefficacypredictionmodelforportwinestainbasedonmachinelearningalgorithms
AT tingzeng theconstructionofhmmepdtefficacypredictionmodelforportwinestainbasedonmachinelearningalgorithms
AT yizhan theconstructionofhmmepdtefficacypredictionmodelforportwinestainbasedonmachinelearningalgorithms
AT ruixuanyou theconstructionofhmmepdtefficacypredictionmodelforportwinestainbasedonmachinelearningalgorithms
AT xinglanhe theconstructionofhmmepdtefficacypredictionmodelforportwinestainbasedonmachinelearningalgorithms
AT rongxiao theconstructionofhmmepdtefficacypredictionmodelforportwinestainbasedonmachinelearningalgorithms
AT xiangningqiu theconstructionofhmmepdtefficacypredictionmodelforportwinestainbasedonmachinelearningalgorithms
AT hongxiayan constructionofhmmepdtefficacypredictionmodelforportwinestainbasedonmachinelearningalgorithms
AT yixintan constructionofhmmepdtefficacypredictionmodelforportwinestainbasedonmachinelearningalgorithms
AT fanqiao constructionofhmmepdtefficacypredictionmodelforportwinestainbasedonmachinelearningalgorithms
AT zhuotongzeng constructionofhmmepdtefficacypredictionmodelforportwinestainbasedonmachinelearningalgorithms
AT yaqianshi constructionofhmmepdtefficacypredictionmodelforportwinestainbasedonmachinelearningalgorithms
AT xueqinzhang constructionofhmmepdtefficacypredictionmodelforportwinestainbasedonmachinelearningalgorithms
AT luli constructionofhmmepdtefficacypredictionmodelforportwinestainbasedonmachinelearningalgorithms
AT tingzeng constructionofhmmepdtefficacypredictionmodelforportwinestainbasedonmachinelearningalgorithms
AT yizhan constructionofhmmepdtefficacypredictionmodelforportwinestainbasedonmachinelearningalgorithms
AT ruixuanyou constructionofhmmepdtefficacypredictionmodelforportwinestainbasedonmachinelearningalgorithms
AT xinglanhe constructionofhmmepdtefficacypredictionmodelforportwinestainbasedonmachinelearningalgorithms
AT rongxiao constructionofhmmepdtefficacypredictionmodelforportwinestainbasedonmachinelearningalgorithms
AT xiangningqiu constructionofhmmepdtefficacypredictionmodelforportwinestainbasedonmachinelearningalgorithms