A federated attention-based stacked LSTM framework for interpretable malaria diagnosis under simulated non-IID federated conditions
7 affiliations
- 1Department of Computer Science and Engineering, St. Peter’s Institute of Higher Education and Research
- 2Department of Artificial Intelligence and Data Science, Saveetha Engineering College
- 3School of Mechanical and Chemical Engineering, WOLLO University, Kombolcha Institute of Technology
- 4Department of Electronics and Communication Engineering, S. A. Engineering College
- 5Department of Electrical and Electronics Engineering, Easwari Engineering College
- 6Department of Computer Science & Engineering (Data Science), Madanapalle Institute of Technology & Science (MITS), Deemed to be University
- 7Department of Electronics and Communication Engineering, St Xavier’s Catholic College of Engineering
Abstract
Malaria remains a major global health burden, particularly in low-resource regions where microscopic diagnosis relies heavily on expert interpretation and is prone to variability. Although deep learning models have demonstrated strong classification performance for automated malaria detection, most existing approaches operate as black-box systems and assume centralized, independent and identically distributed (IID) data. These limitations restrict clinical trust and motivate the exploration of privacy-preserving learning paradigms for distributed healthcare environments. To address these challenges, this study presents FASL-Net, a Federated Attention-Based Stacked Long Short-Term Memory framework that integrates quantifiable interpretability with evaluation under simulated non-IID federated learning conditions. Microscopic blood smear images are transformed into patch-level sequential representations (100 × 75), enabling spatial dependency modeling through stacked LSTM layers. A temperature-controlled attention mechanism identifies diagnostically relevant regions, and interpretability is quantitatively assessed using attention entropy and causal deletion–insertion analysis. The proposed model achieved 94.39% accuracy in centralized training and 94.95% accuracy in a simulated three-client federated learning setting with heterogeneous data partitions. Attention entropy analysis yielded an average entropy of approximately 2.46, substantially lower than the theoretical maximum (≈ 4.60), indicating focused decision patterns. Deletion experiments resulted in an accuracy reduction of approximately 44% when highly attended patches were removed, whereas insertion experiments preserved approximately 95% performance using only the top 10% salient patches. McNemar’s test between centralized and federated models produced a test statistic of 0.03 (p = 0.861), indicating no statistically significant difference under the current simulated setting. These results demonstrate the feasibility of combining federated learning, sequential attention modeling, and quantifiable interpretability for malaria diagnosis under controlled heterogeneous conditions. The study provides a foundational proof-of-concept for trustworthy federated malaria diagnosis and establishes a basis for future validation using real multi-center clinical datasets.
Related
- Interpretable attention-deficit hyperactivity disorder diagnosis from functional magnetic resonance imaging via hybrid convolutional neural network–transformer learning and multi-level graph-based comprehensibility
- AXIAL: Attention-based eXplainability for Interpretable Alzheimer’s Localized diagnosis using 2D CNNs on 3D MRI brain scans
- HAFI-DTA: an interpretable multimodal framework for drug-target affinity prediction based on hierarchical attention guidance and feature iterative interaction
- RMSD: an interpretable framework for streaming-compatible multi-source data fusion and skill-gap diagnosis in vocational education
- A reproducible open-data framework for interpretable breast cancer diagnosis and cytopathology education
BibTeX
@article{murugan2026federated,
title = {{A federated attention-based stacked LSTM framework for interpretable malaria diagnosis under simulated non-IID federated conditions}},
author = {Sangeetha Murugan and K. Thirunadanasikamani and Komal Kumar Napa and Haiter Lenin Allasi and S. Sathya and G. Geethamahalakshmi and Rajkumar Govindarajan and Mary Vasanthi Soosaimariyan},
journal = {Scientific Reports},
year = {2026},
month = jul,
doi = {10.1038/s41598-026-62791-x},
url = {https://doi.org/10.1038/s41598-026-62791-x},
note = {CC BY},
}
RIS
TY - JOUR
TI - A federated attention-based stacked LSTM framework for interpretable malaria diagnosis under simulated non-IID federated conditions
AU - Sangeetha Murugan
AU - K. Thirunadanasikamani
AU - Komal Kumar Napa
AU - Haiter Lenin Allasi
AU - S. Sathya
AU - G. Geethamahalakshmi
AU - Rajkumar Govindarajan
AU - Mary Vasanthi Soosaimariyan
PY - 2026
DA - 2026/07/01
JO - Scientific Reports
DO - 10.1038/s41598-026-62791-x
UR - https://doi.org/10.1038/s41598-026-62791-x
AB - Malaria remains a major global health burden, particularly in low-resource regions where microscopic diagnosis relies heavily on expert interpretation and is prone to variability. Although deep learning models have demonstrated strong classification performance for automated malaria detection, most existing approaches operate as black-box systems and assume centralized, independent and identically distributed (IID) data. These limitations restrict clinical trust and motivate the exploration of privacy-preserving learning paradigms for distributed healthcare environments. To address these challenges, this study presents FASL-Net, a Federated Attention-Based Stacked Long Short-Term Memory framework that integrates quantifiable interpretability with evaluation under simulated non-IID federated learning conditions. Microscopic blood smear images are transformed into patch-level sequential representations (100 × 75), enabling spatial dependency modeling through stacked LSTM layers. A temperature-controlled attention mechanism identifies diagnostically relevant regions, and interpretability is quantitatively assessed using attention entropy and causal deletion–insertion analysis. The proposed model achieved 94.39% accuracy in centralized training and 94.95% accuracy in a simulated three-client federated learning setting with heterogeneous data partitions. Attention entropy analysis yielded an average entropy of approximately 2.46, substantially lower than the theoretical maximum (≈ 4.60), indicating focused decision patterns. Deletion experiments resulted in an accuracy reduction of approximately 44% when highly attended patches were removed, whereas insertion experiments preserved approximately 95% performance using only the top 10% salient patches. McNemar’s test between centralized and federated models produced a test statistic of 0.03 (p = 0.861), indicating no statistically significant difference under the current simulated setting. These results demonstrate the feasibility of combining federated learning, sequential attention modeling, and quantifiable interpretability for malaria diagnosis under controlled heterogeneous conditions. The study provides a foundational proof-of-concept for trustworthy federated malaria diagnosis and establishes a basis for future validation using real multi-center clinical datasets.
SN - 2045-2322
ER -