International Journal of Basic and Applied Science

Visual Classification of Spirulina Microalgae Maturity Stages Based on Color Using Convolutional Neural Networks

2026-04-20T14:55:03+00:00

Manual assessment of microalgae maturity is subjective and does not scale well. This study evaluates whether a convolutional neural network can classify Spirulina microalgae into five visual maturity stages (dead, unripe, under-ripe, medium-ripe, ripe) using color alone. From 5,005 images collected over three months at a single facility, byte-exact duplicates accounted for 41.9% of the dataset; after removal, 2,909 unique images remained. To prevent near-duplicate images from the same session leaking across folds, evaluation used a group-stratified 5-fold cross-validation scheme, with groups defined via perceptual-hash clustering. An ablation across four preprocessing configurations found that combining CLAHE with augmentation, the original proposed method, was numerically weakest (94.74% ± 2.15% accuracy versus 97.43% ± 0.89% with no preprocessing), differing significantly from CLAHE alone (p = 0.021) though not from the plain baseline at this sample size (p = 0.085). A separate comparison, matching fine-tuning depth (last 30 layers) and preprocessing across a custom CNN, MobileNetV2, and ResNet50, found ResNet50 highest at 99.52% ± 0.18% accuracy, significantly ahead of the custom CNN (97.06% ± 1.39%; p = 0.024), while MobileNetV2 showed severe fold-to-fold instability. The custom CNN's advantage lay in efficiency: 54.8 times fewer parameters and roughly 3.5 times faster inference than ResNet50, for only a small accuracy trade-off. A brief, unlabeled field test on six images outside the lab setting showed frequent disagreement among the three models, suggesting limited transfer from controlled to field conditions. Grad-CAM maps are presented as an indication of model attention, not evidence of underlying chlorophyll degradation biochemistry

Spatio-Temporal Disinformation Detection Using GNN–LSTM and Cross-Attention for South China Sea Information Analysis

2026-06-06T09:17:53+00:00

The rapid propagation of disinformation through social media presents a significant challenge because information diffusion is characterized by complex interactions among users and continuously evolving temporal patterns. Conventional deep-learning and transformer-based approaches primarily focus on textual or sequential representations and may not fully capture the structural and temporal characteristics of information propagation. This study proposes a spatio-temporal framework for disinformation detection that integrates Graph Neural Network (GNN)-based spatial representation, Long Short-Term Memory (LSTM)-based temporal modeling, and an adaptive cross-attention mechanism for spatial–temporal fusion. The GNN is employed to model relational dependencies within social-media interaction networks, while LSTM captures the temporal evolution of graph-based representations. Cross-attention is subsequently introduced to adaptively model the interaction between spatial and temporal representations rather than relying solely on fixed feature concatenation. The proposed framework is evaluated using a social-media dataset associated with the South China Sea and compared with CNN, LSTM, GNN, BERT, and GAT-BERT baselines. Experimental results show that the proposed model achieves 95.84% accuracy, 95.30% precision, 95.60% recall, and 95.45% F1-score, outperforming the strongest baseline, GAT-BERT, by 2.63 percentage points in accuracy and 2.55 percentage points in F1-score. The ablation analysis further demonstrates that integrating LSTM with GNN improves accuracy by 5.32 percentage points, while the subsequent incorporation of cross-attention provides an additional improvement of 3.74 percentage points. These findings indicate that adaptive spatial–temporal fusion contributes beyond the independent use of graph and sequential encoders and provides a potential decision-support approach for analyzing dynamic disinformation propagation in geopolitical social-media environments.

Slice-Group-Level Brain Tumor Classification Using CRNN-GRU

2026-07-31T16:22:09+00:00

Interpreting brain MRI for tumor diagnosis is time-consuming and varies between raters, and single-slice decisions can be fragile. This paper compares two classification pipelines on the Brain Tumor Classification (MRI) Dataset from Kaggle (glioma, meningioma, pituitary, no-tumor) of T1-contrast-enhanced images. The first is a per-slice CNN. The second is a per-group CRNN-GRU that reads an ordered sequence of ten slices with a bidirectional GRU and inter-slice attention. Because the dataset provides no patient identifiers, the analysis unit is a surrogate slice group, and the splits are group-disjoint by construction; all models are compared at the group-level. The per-slice CNN reached 80.0% slice-level accuracy. Aggregated to the group-level, the same CNN reached 92.68% by majority voting and 95.12% by mean-probability pooling. The CRNN-GRU reached 90.2% in a single pass, and 87.8% with test-time augmentation. No pairwise difference is statistically significant (exact McNemar, all p >= 0.25). The results do not show that sequential modelling and inter-slice attention improve on simple probability aggregation on this dataset, and the simplest configuration was the most accurate. A content-level audit found a small number of images duplicated across partitions, so all figures should be read as upper bounds. These results describe internal performance on one dataset under group-disjoint evaluation with a content-level leakage audit and do not establish clinical generalization.