Crop pathogens threaten global agriculture by causing severe yield and economic losses. Conventional detection methods are often slow and inaccurate, limiting timely intervention. This study introduces a pentagon-shaped terahertz photonic crystal fiber (THz PCF) biosensor, optimized with the decision cascaded 3D return dilated secretary-bird aligned convolutional transformer network (DC3D-SBA-CTN). The biosensor is designed to detect a broad spectrum of pathogens, including fungi (e.g., Fusarium spp.) and bacteria (e.g., Xanthomonas spp.), by identifying their unique refractive index signatures. Integrating advanced neural networks and optimization algorithms, the biosensor achieves a detection accuracy of 99.87%, precision of 99.65%, sensitivity of 99.77%, and specificity of 99.83%, as validated by a 5-fold cross-validation protocol. It offers high sensitivity (up to 7340 RIU⁻¹), low signal loss, and robust performance against morphological variations, making it adaptable for diverse agricultural settings. This innovation enables rapid, precise monitoring of crop pathogens, revolutionizing plant disease management.

Diabetes mellitus is a pervasive global health challenge, necessitating early and accurate diagnostic methods. Current techniques often lack the sensitivity for early-stage detection and fail to capture complex biomarker interactions. This paper proposes a novel biosensing platform integrating an octagon-shaped terahertz (THz) photonic crystal fiber (PCF) biosensor with a custom deep learning model for high-sensitivity diabetes detection. The innovative sensor geometry enhances light-matter interaction, significantly reducing signal loss. The extracted THz spectral data is processed by a Dilated Causal Convolution with Botox Multi-Head Self-Attention Network (Dil-2CBM-SAN), which optimizes feature extraction and classification. Our synergistic approach demonstrates exceptional performance, achieving a peak accuracy of 99.98% and a wavelength sensitivity of 48,000 RIU for biomarker concentrations as low as 0.1 nM, while maintaining minimal confinement loss. This work presents a groundbreaking and robust framework for early, precise diabetes diagnosis.