A Meta-Learning-Based Ensemble Model for Explainable Alzheimer’s Disease Diagnosis
<b>Background/Objectives:</b> Artificial intelligence (AI) models for Alzheimer’s disease (AD) diagnosis often face the challenge of limited explainability, hindering their clinical adoption. Previous studies have relied on full-scale MRI, which increases unnecessary features, creating a...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-06-01
|
| Series: | Diagnostics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-4418/15/13/1642 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849428718428094464 |
|---|---|
| author | Fatima Hasan Al-bakri Wan Mohd Yaakob Wan Bejuri Mohamed Nasser Al-Andoli Raja Rina Raja Ikram Hui Min Khor Zulkifli Tahir The Alzheimer’s Disease Neuroimaging Initiative |
| author_facet | Fatima Hasan Al-bakri Wan Mohd Yaakob Wan Bejuri Mohamed Nasser Al-Andoli Raja Rina Raja Ikram Hui Min Khor Zulkifli Tahir The Alzheimer’s Disease Neuroimaging Initiative |
| author_sort | Fatima Hasan Al-bakri |
| collection | DOAJ |
| description | <b>Background/Objectives:</b> Artificial intelligence (AI) models for Alzheimer’s disease (AD) diagnosis often face the challenge of limited explainability, hindering their clinical adoption. Previous studies have relied on full-scale MRI, which increases unnecessary features, creating a “black-box” problem in current XAI models. <b>Methods:</b> This study proposes an explainable ensemble-based diagnostic framework trained on both clinical data and mid-slice axial MRI from the ADNI and OASIS datasets. The methodology involves training an ensemble model that integrates Random Forest, Support Vector Machine, XGBoost, and Gradient Boosting classifiers, with meta-logistic regression used for the final decision. The core contribution lies in the exclusive use of mid-slice MRI images, which highlight the lateral ventricles, thus improving the transparency and clinical relevance of the decision-making process. Our mid-slice approach minimizes unnecessary features and enhances model explainability by design. <b>Results:</b> We achieved state-of-the-art diagnostic accuracy: 99% on OASIS and 97.61% on ADNI using clinical data alone; 99.38% on OASIS and 98.62% on ADNI using only mid-slice MRI; and 99% accuracy when combining both modalities. The findings demonstrated significant progress in diagnostic transparency, as the algorithm consistently linked predictions to observed structural changes in the dilated lateral ventricles of the brain, which serve as a clinically reliable biomarker for AD and can be easily verified by medical professionals. <b>Conclusions:</b> This research presents a step toward more transparent AI-driven diagnostics, bridging the gap between accuracy and explainability in XAI. |
| format | Article |
| id | doaj-art-231fb05dbb9442489fad598cc6f06249 |
| institution | Kabale University |
| issn | 2075-4418 |
| language | English |
| publishDate | 2025-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Diagnostics |
| spelling | doaj-art-231fb05dbb9442489fad598cc6f062492025-08-20T03:28:37ZengMDPI AGDiagnostics2075-44182025-06-011513164210.3390/diagnostics15131642A Meta-Learning-Based Ensemble Model for Explainable Alzheimer’s Disease DiagnosisFatima Hasan Al-bakri0Wan Mohd Yaakob Wan Bejuri1Mohamed Nasser Al-Andoli2Raja Rina Raja Ikram3Hui Min Khor4Zulkifli Tahir5The Alzheimer’s Disease Neuroimaging InitiativeFaculty of Information and Communication Technology, Universiti Teknikal Malaysia Melaka, Melaka 76100, MalaysiaFaculty of Information and Communication Technology, Universiti Teknikal Malaysia Melaka, Melaka 76100, MalaysiaFaculty of Computing Informatics, Multimedia University, Cyberjaya 63100, MalaysiaFaculty of Information and Communication Technology, Universiti Teknikal Malaysia Melaka, Melaka 76100, MalaysiaDepartment of Medicine, Faculty of Medicine, University of Malaya, Kuala Lumpur 50603, MalaysiaFaculty of Engineering, Universitas Hasanuddin, Gowa 92171, Indonesia<b>Background/Objectives:</b> Artificial intelligence (AI) models for Alzheimer’s disease (AD) diagnosis often face the challenge of limited explainability, hindering their clinical adoption. Previous studies have relied on full-scale MRI, which increases unnecessary features, creating a “black-box” problem in current XAI models. <b>Methods:</b> This study proposes an explainable ensemble-based diagnostic framework trained on both clinical data and mid-slice axial MRI from the ADNI and OASIS datasets. The methodology involves training an ensemble model that integrates Random Forest, Support Vector Machine, XGBoost, and Gradient Boosting classifiers, with meta-logistic regression used for the final decision. The core contribution lies in the exclusive use of mid-slice MRI images, which highlight the lateral ventricles, thus improving the transparency and clinical relevance of the decision-making process. Our mid-slice approach minimizes unnecessary features and enhances model explainability by design. <b>Results:</b> We achieved state-of-the-art diagnostic accuracy: 99% on OASIS and 97.61% on ADNI using clinical data alone; 99.38% on OASIS and 98.62% on ADNI using only mid-slice MRI; and 99% accuracy when combining both modalities. The findings demonstrated significant progress in diagnostic transparency, as the algorithm consistently linked predictions to observed structural changes in the dilated lateral ventricles of the brain, which serve as a clinically reliable biomarker for AD and can be easily verified by medical professionals. <b>Conclusions:</b> This research presents a step toward more transparent AI-driven diagnostics, bridging the gap between accuracy and explainability in XAI.https://www.mdpi.com/2075-4418/15/13/1642Alzheimer’s disease diagnosisexplainable artificial intelligence (XAI)ensemble learningmid-slice MRIlateral ventriclesclinical data |
| spellingShingle | Fatima Hasan Al-bakri Wan Mohd Yaakob Wan Bejuri Mohamed Nasser Al-Andoli Raja Rina Raja Ikram Hui Min Khor Zulkifli Tahir The Alzheimer’s Disease Neuroimaging Initiative A Meta-Learning-Based Ensemble Model for Explainable Alzheimer’s Disease Diagnosis Diagnostics Alzheimer’s disease diagnosis explainable artificial intelligence (XAI) ensemble learning mid-slice MRI lateral ventricles clinical data |
| title | A Meta-Learning-Based Ensemble Model for Explainable Alzheimer’s Disease Diagnosis |
| title_full | A Meta-Learning-Based Ensemble Model for Explainable Alzheimer’s Disease Diagnosis |
| title_fullStr | A Meta-Learning-Based Ensemble Model for Explainable Alzheimer’s Disease Diagnosis |
| title_full_unstemmed | A Meta-Learning-Based Ensemble Model for Explainable Alzheimer’s Disease Diagnosis |
| title_short | A Meta-Learning-Based Ensemble Model for Explainable Alzheimer’s Disease Diagnosis |
| title_sort | meta learning based ensemble model for explainable alzheimer s disease diagnosis |
| topic | Alzheimer’s disease diagnosis explainable artificial intelligence (XAI) ensemble learning mid-slice MRI lateral ventricles clinical data |
| url | https://www.mdpi.com/2075-4418/15/13/1642 |
| work_keys_str_mv | AT fatimahasanalbakri ametalearningbasedensemblemodelforexplainablealzheimersdiseasediagnosis AT wanmohdyaakobwanbejuri ametalearningbasedensemblemodelforexplainablealzheimersdiseasediagnosis AT mohamednasseralandoli ametalearningbasedensemblemodelforexplainablealzheimersdiseasediagnosis AT rajarinarajaikram ametalearningbasedensemblemodelforexplainablealzheimersdiseasediagnosis AT huiminkhor ametalearningbasedensemblemodelforexplainablealzheimersdiseasediagnosis AT zulkiflitahir ametalearningbasedensemblemodelforexplainablealzheimersdiseasediagnosis AT thealzheimersdiseaseneuroimaginginitiative ametalearningbasedensemblemodelforexplainablealzheimersdiseasediagnosis AT fatimahasanalbakri metalearningbasedensemblemodelforexplainablealzheimersdiseasediagnosis AT wanmohdyaakobwanbejuri metalearningbasedensemblemodelforexplainablealzheimersdiseasediagnosis AT mohamednasseralandoli metalearningbasedensemblemodelforexplainablealzheimersdiseasediagnosis AT rajarinarajaikram metalearningbasedensemblemodelforexplainablealzheimersdiseasediagnosis AT huiminkhor metalearningbasedensemblemodelforexplainablealzheimersdiseasediagnosis AT zulkiflitahir metalearningbasedensemblemodelforexplainablealzheimersdiseasediagnosis AT thealzheimersdiseaseneuroimaginginitiative metalearningbasedensemblemodelforexplainablealzheimersdiseasediagnosis |