Cardiac arrhythmia is a major cause of cardiovascular mortality, therefore accurate and timely analysis of ECG is essential for early diagnosis. In this study, a lightweight One-dimensional Convolutional Neural Network (1D CNN) with Convolutional Block Attention Module (CBAM) is proposed for explainable ECG classification according to the AAMI standard. The CBAM-ECGnet model performs convolutional feature extraction with channel and spatial attention with only 4,088 total parameters (≈16 KB). The proposed model is evaluated on the MIT-BIH Arrhythmia Database and achieves an overall accuracy of 98.53%, and a ROC-AUC of 99% . In this study we use Grad-CAM to visualize the regions of the ECG that contribute to each prediction, improving the explainability of the model. This model achieves high accuracy classification with low computational complexity and thus is suitable for real-time and low-resource ECG monitoring application.

This paper introduces CalForget, a four-stage pipeline that detects conformal prediction calibration drift under distribution shift, attributes miscalibration to specific training points using influence functions, selectively removes their influence via a soft-miscoverage objective, and restores finite-sample coverage guarantees through reconformalization. Unlike full retraining, CalForget recovers most of the calibration benefit while modifying only a small fraction (≈2.5%) of the training data, demonstrated across four UCI regression benchmarks and three shift severities.

Glaucoma is a major cause of irreversible blindness,
and early screening from fundus photographs is essential because
structural damage may progress before noticeable visual symp-
toms occur. This paper proposes HMSGA-Net, a Hybrid Multi-
Scale Gated Attention Network for automated glaucoma classi-
fication. The framework combines a ConvNeXt-Small primary
branch and a ConvNeXt-Tiny auxiliary branch, each equipped
with an HMSGA module comprising parallel 1 × 1, 3 × 3, and
5 × 5 depth-wise convolutions, Efficient Channel Attention, dual
sigmoid gating, and residual feature refinement. The two branch
probabilities are fused using a validation-selected ensemble
weight and decision threshold. Integrated Gradients and Occlu-
sion Sensitivity are employed to assess whether predictions focus
on clinically relevant optic-disc and neuroretinal-rim regions. On
the ACRIMA dataset, the best seed-42 configuration achieved
98.84% accuracy, 98.97% F1-score, 99.72% AUC-ROC, and
100% glaucoma recall. Across three independent splits, HMSGA-
Net obtained mean accuracy and AUC-ROC values of 94.14%
and 97.28%, respectively, demonstrating strong discrimination
while also revealing sensitivity to data partitioning.

Accurate brain tumor identification from magnetic
resonance imaging (MRI) is essential for early clinical decision
support, yet reliable multi-class classification remains challenging
because tumor appearance varies across shape, texture, and
intensity patterns. This study proposes an explainable deep
ensemble framework for four-class brain tumor MRI classification, covering glioma, meningioma, pituitary tumor, and
no-tumor cases. The pipeline applies image preprocessing and
augmentation, evaluates multiple pretrained convolutional base-
lines, and combines complementary transfer-learning models
through weighted decision fusion to improve generalization.
Model performance is assessed using accuracy, precision, recall,
F1-score, bootstrap confidence estimation, and train-validation
stability analysis. Experimental results on the Kaggle brain tumor
MRI dataset show that the proposed framework achieves a
verified accuracy of 99.44%, with precision, recall, and F1-score
above 0.994. Explainability analysis further supports the clinical
relevance of the learned decision regions. The results indicate
that ensemble-based transfer learning can provide robust and
interpretable MRI-based brain tumor screening support

Colorectal cancer remains one of the leading causes
of cancer-related mortality, making accurate polyp segmentation
important for early diagnosis and treatment. This study proposes
a PVT-v2-B2 + FPN framework for automatic colorectal polyp
segmentation. The model combines a hierarchical transformer
encoder with feature pyramid decoding to integrate global
context and multi-scale spatial features for accurate boundary
localization. Geometric and photometric augmentations were applied to improve robustness under diverse colonoscopy conditions.
Experiments were conducted on Kvasir-SEG using a 700/150/150
train-validation-test split with multi-seed and multi-resolution
evaluation. The selected proposed configuration achieved a test
Dice score of 91.05% and IoU of 83.71%, while the best
individual run reached 91.08% Dice. Compared with CNN-,
transformer-, and hybrid-based baselines, the proposed model
showed competitive segmentation performance with 29.8M parameters. These results indicate that hierarchical transformer
features with FPN-based fusion can provide an effective balance
between segmentation accuracy and computational practicality
for colonoscopy-assisted polyp analysis.

Over 70 million deaf people worldwide frequently use sign language. However, the challenges of real-time implementation and interpretation have limited automated sign language recognition systems. CNN-based older techniques frequently overlook subtle features that are crucial for differentiating between similar motions. This research examines the recognition of 37 distinct sign language classes, which include A to Z, 0 to 9, and a space. We utilized a dataset of 55,500 photos for this. Our recommended method, ViT-Sign, uses a Tiny version of Vision Transformer. It has only 5.5 million parameters, and we evaluate its performance using accuracy, precision, recall, and F1-score. Our model achieves 99.42\% accuracy, 99.44\% precision, 99.42\% recall, and a 99.42\% F1-score. We also employ Grad-CAM visuals to make it clear that our model concentrates on the correct regions of the hands rather than random background information. Experiments show that Vision Transformers can efficiently record spatial relationships and acquire discriminative representations of gestures. The proposed method can be utilized to create successful communication systems that allow hearing-impaired people to communicate more easily.

While banana is an important crop in the agricultural sector of South Asia, fungal and bacterial leaf diseases continue to cause substantial losses and pose a threat to the livelihoods of farmers in the region. The current method of field diagnosis is mostly manually, which is subjective and slow, inconvenient in the field. This paper proposes a rigorous and explainable deep learning framework for the automated banana leaf disease classification that overcomes three major drawbacks of previous works: 1) the lack of comparison of multiple models in similar experimental setups, 2) the lack of statistical validation, and 3) the absence of a connection between the model prediction and the farmer’s guidance. We test the three architectures: a lightweight Custom CNN, ResNet- 50 and EfficientNet-B0, on the publicly available BananaLSD dataset, based on a standardized preprocessing and augmentation pipeline. Stratified 5-fold cross validation is used for assessing the performance and paired t test is used to check the statistical significance of the difference observed. The best classification accuracy of EfficientNet-B0 is 99.69%, the mean cross-validation accuracy is 99.36%, and the standard deviation is the lowest (0.10), showing very high stability. For transparency, we integrate four explainability methods: Grad-CAM, Grad-CAM++, Score- CAM and Layer-CAM, which result in uniform lesion localized heatmaps that validate biologically meaningful feature focus. Finally, an intelligent advisory module translates the predictions in actionable management recommendations directly deployable by agricultural extension services. The proposed framework is built upon predictive excellence, interpretable decision making, and real-world usability in one seamless pipeline, which is critical for advancing precision agriculture.

• A MOGA-based framework is proposed for healthcare
fraud detection, which simultaneously optimizes classifi
cation performance and feature subset size.
• The framework integrates LIME to enable instance-level
interpretability, facilitating a deeper understanding of
model predictions.
• A comprehensive comparative analysis is conducted to
evaluate model performance and interpretability before
and after feature selection.