Grad-CAM Enhanced Explainability in Multi-Class Chest X-Ray Classification
Bibliographic record
Abstract
Rapid and accurate diagnosis of respiratory illnesses is essential for improving patient outcomes and reducing the burden on healthcare systems, especially during global health crises. This paper presents a comparative study of three deep learning models— a custom lightweight Convolutional Neural Network (CNN), VGG16, and DenseNet121—for the multi-class classification of chest X-rays into COVID-19, pneumonia, and healthy cases. A balanced dataset of 12,750 images was curated from public sources, with equal distribution across all classes. Preprocessing included resizing, normalization, and extensive augmentation to mitigate overfitting and improve generalization. The custom CNN architecture, consisting of four convolutional blocks, global average pooling, and dense layers with dropout regularization, was optimized for efficiency and interpretability. All models were trained with the Adam optimizer, categorical cross-entropy loss, and early stopping. Evaluation metrics included accuracy, F1-score, AUC, and confusion matrices, complemented by Grad-CAM visualizations for interpretability. Experimental results showed the custom CNN outperforming transfer learning models, achieving 94.1% test accuracy, macro F1-score of 0.94, and AUC of 0.99, compared to VGG16 (92.3%) and DenseNet121 (91.6%). The proposed CNN also demonstrated faster convergence and superior recall for COVID-19 detection. These findings highlight that lightweight, well-regularized CNNs can match or exceed deeper architectures, offering scalable and explainable solutions for real-world clinical deployment in resource-constrained environments.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.007 |
| Meta-epidemiology (narrow) | 0.002 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.003 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".