Comparison of Resampling Methods and Radiomic Machine Learning Classifiers for Predicting Bone Quality Using Dual-Energy X-Ray Absorptiometry
<b>Background/Objectives</b>: This study presents a novel approach, based on a combination of radiomic feature extraction, data resampling techniques, and machine learning algorithms, for the detection of degraded bone structures in Dual X-ray Absorptiometry (DXA) images. This comprehens...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Diagnostics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-4418/15/2/175 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832588722651004928 |
|---|---|
| author | Mailen Gonzalez José Manuel Fuertes García María Belén Zanchetta Rubén Abdala José María Massa |
| author_facet | Mailen Gonzalez José Manuel Fuertes García María Belén Zanchetta Rubén Abdala José María Massa |
| author_sort | Mailen Gonzalez |
| collection | DOAJ |
| description | <b>Background/Objectives</b>: This study presents a novel approach, based on a combination of radiomic feature extraction, data resampling techniques, and machine learning algorithms, for the detection of degraded bone structures in Dual X-ray Absorptiometry (DXA) images. This comprehensive approach, which addresses the critical aspects of the problem, distinguishes this work from previous studies, improving the performance achieved by the most similar studies. The primary aim is to provide clinicians with an accessible tool for quality bone assessment, which is currently limited. <b>Methods</b>: A dataset of 1531 spine DXA images was automatically segmented and labelled based on Trabecular Bone Score (TBS) values. Radiomic features were extracted using Pyradiomics, and various resampling techniques were employed to address class imbalance. Three machine learning classifiers (Logistic Regression, Support Vector Machine (SVM), and XGBoost) were trained and evaluated using standard performance metrics. <b>Results</b>: The SVM classifier outperformed the other classifiers. The highest F-score of 97.5% was achieved using the Grey Level Dependence Matrix and Grey Level Run Length Matrix feature combination with SMOTEENN resampling, which proved to be the most effective resampling technique, while the undersampling method yielded the lowest performance. <b>Conclusions</b>: This research demonstrates the potential of radiomic texture features, resampling techniques, and machine learning methods for classifying DXA images into healthy or degraded bone structures, which potentially leads to improved clinical diagnosis and treatment. |
| format | Article |
| id | doaj-art-c01387d72e7a4bce92f712c25d709b38 |
| institution | Kabale University |
| issn | 2075-4418 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Diagnostics |
| spelling | doaj-art-c01387d72e7a4bce92f712c25d709b382025-01-24T13:29:00ZengMDPI AGDiagnostics2075-44182025-01-0115217510.3390/diagnostics15020175Comparison of Resampling Methods and Radiomic Machine Learning Classifiers for Predicting Bone Quality Using Dual-Energy X-Ray AbsorptiometryMailen Gonzalez0José Manuel Fuertes García1María Belén Zanchetta2Rubén Abdala3José María Massa4Instituto de Investigación en Tecnología Informática Avanzada, Universidad Nacional del Centro de la Provincia de Buenos Aires, Tandil 7000, ArgentinaDepartamento de Informática, Escuela Politécnica Superior, Universidad de Jaén, 23071 Jaén, SpainInstituto de Diagnóstico e Investigaciones Metabólicas, Buenos Aires 1012, ArgentinaInstituto de Diagnóstico e Investigaciones Metabólicas, Buenos Aires 1012, ArgentinaInstituto de Investigación en Tecnología Informática Avanzada, Universidad Nacional del Centro de la Provincia de Buenos Aires, Tandil 7000, Argentina<b>Background/Objectives</b>: This study presents a novel approach, based on a combination of radiomic feature extraction, data resampling techniques, and machine learning algorithms, for the detection of degraded bone structures in Dual X-ray Absorptiometry (DXA) images. This comprehensive approach, which addresses the critical aspects of the problem, distinguishes this work from previous studies, improving the performance achieved by the most similar studies. The primary aim is to provide clinicians with an accessible tool for quality bone assessment, which is currently limited. <b>Methods</b>: A dataset of 1531 spine DXA images was automatically segmented and labelled based on Trabecular Bone Score (TBS) values. Radiomic features were extracted using Pyradiomics, and various resampling techniques were employed to address class imbalance. Three machine learning classifiers (Logistic Regression, Support Vector Machine (SVM), and XGBoost) were trained and evaluated using standard performance metrics. <b>Results</b>: The SVM classifier outperformed the other classifiers. The highest F-score of 97.5% was achieved using the Grey Level Dependence Matrix and Grey Level Run Length Matrix feature combination with SMOTEENN resampling, which proved to be the most effective resampling technique, while the undersampling method yielded the lowest performance. <b>Conclusions</b>: This research demonstrates the potential of radiomic texture features, resampling techniques, and machine learning methods for classifying DXA images into healthy or degraded bone structures, which potentially leads to improved clinical diagnosis and treatment.https://www.mdpi.com/2075-4418/15/2/175dual energy X-ray absorptiometrytrabecular bone scoreradiomicsdata resamplingmachine learning |
| spellingShingle | Mailen Gonzalez José Manuel Fuertes García María Belén Zanchetta Rubén Abdala José María Massa Comparison of Resampling Methods and Radiomic Machine Learning Classifiers for Predicting Bone Quality Using Dual-Energy X-Ray Absorptiometry Diagnostics dual energy X-ray absorptiometry trabecular bone score radiomics data resampling machine learning |
| title | Comparison of Resampling Methods and Radiomic Machine Learning Classifiers for Predicting Bone Quality Using Dual-Energy X-Ray Absorptiometry |
| title_full | Comparison of Resampling Methods and Radiomic Machine Learning Classifiers for Predicting Bone Quality Using Dual-Energy X-Ray Absorptiometry |
| title_fullStr | Comparison of Resampling Methods and Radiomic Machine Learning Classifiers for Predicting Bone Quality Using Dual-Energy X-Ray Absorptiometry |
| title_full_unstemmed | Comparison of Resampling Methods and Radiomic Machine Learning Classifiers for Predicting Bone Quality Using Dual-Energy X-Ray Absorptiometry |
| title_short | Comparison of Resampling Methods and Radiomic Machine Learning Classifiers for Predicting Bone Quality Using Dual-Energy X-Ray Absorptiometry |
| title_sort | comparison of resampling methods and radiomic machine learning classifiers for predicting bone quality using dual energy x ray absorptiometry |
| topic | dual energy X-ray absorptiometry trabecular bone score radiomics data resampling machine learning |
| url | https://www.mdpi.com/2075-4418/15/2/175 |
| work_keys_str_mv | AT mailengonzalez comparisonofresamplingmethodsandradiomicmachinelearningclassifiersforpredictingbonequalityusingdualenergyxrayabsorptiometry AT josemanuelfuertesgarcia comparisonofresamplingmethodsandradiomicmachinelearningclassifiersforpredictingbonequalityusingdualenergyxrayabsorptiometry AT mariabelenzanchetta comparisonofresamplingmethodsandradiomicmachinelearningclassifiersforpredictingbonequalityusingdualenergyxrayabsorptiometry AT rubenabdala comparisonofresamplingmethodsandradiomicmachinelearningclassifiersforpredictingbonequalityusingdualenergyxrayabsorptiometry AT josemariamassa comparisonofresamplingmethodsandradiomicmachinelearningclassifiersforpredictingbonequalityusingdualenergyxrayabsorptiometry |