ALL diagnosis: can efficiency and transparency coexist? An explainble deep learning approach
Abstract Acute Lymphoblastic Leukemia (ALL) is a life-threatening malignancy characterized by its aggressive progression and detrimental effects on the hematopoietic system. Early and accurate diagnosis is paramount to optimizing therapeutic interventions and improving clinical outcomes. This study...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-04-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-025-97297-5 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849699343450243072 |
|---|---|
| author | Dost Muhammad Muhammad Salman Ayse Keles Malika Bendechache |
| author_facet | Dost Muhammad Muhammad Salman Ayse Keles Malika Bendechache |
| author_sort | Dost Muhammad |
| collection | DOAJ |
| description | Abstract Acute Lymphoblastic Leukemia (ALL) is a life-threatening malignancy characterized by its aggressive progression and detrimental effects on the hematopoietic system. Early and accurate diagnosis is paramount to optimizing therapeutic interventions and improving clinical outcomes. This study introduces a novel diagnostic framework that synergizes the EfficientNet-B7 architecture with Explainable Artificial Intelligence (XAI) methodologies to address challenges in performance, computational efficiency, and explainability. The proposed model achieves improved diagnostic performance, with accuracies exceeding 96% on the Taleqani Hospital dataset and 95.50% on the C-NMC-19 and Multi-Cancer datasets. Rigorous evaluation across multiple metrics-including Area Under the Curve (AUC), mean Average Precision (mAP), Accuracy, Precision, Recall, and F1-score-demonstrates the model’s robustness and establishes its superiority over state-of-the-art architectures namely VGG-19, InceptionResNetV2, ResNet50, DenseNet50 and AlexNet . Furthermore, the framework significantly reduces computational overhead, achieving up to 40% faster inference times, thereby enhancing its clinical applicability. To address the opacity inherent in Deep learning (DL) models, the framework integrates advanced XAI techniques, including Gradient-weighted Class Activation Mapping (Grad-CAM), Class Activation Mapping (CAM), Local Interpretable Model-Agnostic Explanations (LIME), and Integrated Gradients (IG), providing transparent and explainable insights into model predictions. This fusion of high diagnostic precision, computational efficiency, and explainability positions the proposed framework as a transformative tool for ALL diagnosis, bridging the gap between cutting-edge AI technologies and practical clinical deployment. |
| format | Article |
| id | doaj-art-6d8d4471b4e84896a89df8392a290441 |
| institution | DOAJ |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-04-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-6d8d4471b4e84896a89df8392a2904412025-08-20T03:18:38ZengNature PortfolioScientific Reports2045-23222025-04-0115112010.1038/s41598-025-97297-5ALL diagnosis: can efficiency and transparency coexist? An explainble deep learning approachDost Muhammad0Muhammad Salman1Ayse Keles2Malika Bendechache3CRT-AI and ADAPT Research Centres, School of Computer Science, University of GalwayDepartment of Software Engineering, University of MalakandADAPT Research Centre, School of Computer Science, University of GalwayADAPT Research Centre, School of Computer Science, University of GalwayAbstract Acute Lymphoblastic Leukemia (ALL) is a life-threatening malignancy characterized by its aggressive progression and detrimental effects on the hematopoietic system. Early and accurate diagnosis is paramount to optimizing therapeutic interventions and improving clinical outcomes. This study introduces a novel diagnostic framework that synergizes the EfficientNet-B7 architecture with Explainable Artificial Intelligence (XAI) methodologies to address challenges in performance, computational efficiency, and explainability. The proposed model achieves improved diagnostic performance, with accuracies exceeding 96% on the Taleqani Hospital dataset and 95.50% on the C-NMC-19 and Multi-Cancer datasets. Rigorous evaluation across multiple metrics-including Area Under the Curve (AUC), mean Average Precision (mAP), Accuracy, Precision, Recall, and F1-score-demonstrates the model’s robustness and establishes its superiority over state-of-the-art architectures namely VGG-19, InceptionResNetV2, ResNet50, DenseNet50 and AlexNet . Furthermore, the framework significantly reduces computational overhead, achieving up to 40% faster inference times, thereby enhancing its clinical applicability. To address the opacity inherent in Deep learning (DL) models, the framework integrates advanced XAI techniques, including Gradient-weighted Class Activation Mapping (Grad-CAM), Class Activation Mapping (CAM), Local Interpretable Model-Agnostic Explanations (LIME), and Integrated Gradients (IG), providing transparent and explainable insights into model predictions. This fusion of high diagnostic precision, computational efficiency, and explainability positions the proposed framework as a transformative tool for ALL diagnosis, bridging the gap between cutting-edge AI technologies and practical clinical deployment.https://doi.org/10.1038/s41598-025-97297-5Explainable artificial intelligenceXAI for medical diagnosisEXplainble medical imagingALL detectionDecision support systemResponsible AI |
| spellingShingle | Dost Muhammad Muhammad Salman Ayse Keles Malika Bendechache ALL diagnosis: can efficiency and transparency coexist? An explainble deep learning approach Scientific Reports Explainable artificial intelligence XAI for medical diagnosis EXplainble medical imaging ALL detection Decision support system Responsible AI |
| title | ALL diagnosis: can efficiency and transparency coexist? An explainble deep learning approach |
| title_full | ALL diagnosis: can efficiency and transparency coexist? An explainble deep learning approach |
| title_fullStr | ALL diagnosis: can efficiency and transparency coexist? An explainble deep learning approach |
| title_full_unstemmed | ALL diagnosis: can efficiency and transparency coexist? An explainble deep learning approach |
| title_short | ALL diagnosis: can efficiency and transparency coexist? An explainble deep learning approach |
| title_sort | all diagnosis can efficiency and transparency coexist an explainble deep learning approach |
| topic | Explainable artificial intelligence XAI for medical diagnosis EXplainble medical imaging ALL detection Decision support system Responsible AI |
| url | https://doi.org/10.1038/s41598-025-97297-5 |
| work_keys_str_mv | AT dostmuhammad alldiagnosiscanefficiencyandtransparencycoexistanexplainbledeeplearningapproach AT muhammadsalman alldiagnosiscanefficiencyandtransparencycoexistanexplainbledeeplearningapproach AT aysekeles alldiagnosiscanefficiencyandtransparencycoexistanexplainbledeeplearningapproach AT malikabendechache alldiagnosiscanefficiencyandtransparencycoexistanexplainbledeeplearningapproach |