"Manduca et al. 2026 TMRB"

"Understanding the neuro-motor effects of sub-lethal exposure to bioactive compounds is crucial for advancing both toxicological assessment and bio-inspired sensing systems. In this study, an artificial intelligence (AI)-driven behavioral phenotyping framework was designed to investigate sex-specific neuro-motor responses in the Mediterranean fruit fly Ceratitis capitata following ingestion of an LC30 concentration of carlina oxide. High-resolution behavioral data were acquired through video recordings and processed using deep learning for markerless pose estimation. From the pose-derived kinematics, 44 features describing locomotion, wing activity, and rotational dynamics were extracted and analyzed. Statistical analysis revealed pronounced sex- and treatment-dependent alterations, with the largest behavioral divergence observed between treated males and females (25\/44 features significantly different). Machine learning classifiers further confirmed these findings, with Random Forest achieving the highest performance in the treated sex comparison (accuracy = 0.79, F1-score = 0.80, ROC-AUC = 0.87). The effects of treatment were more evident in males (AUC = 0.71). These results suggest that sub-lethal exposure induces measurable neuro-motor alterations that can be effectively captured through AI-based behavioral analysis. The proposed framework provides a scalable and non-invasive tool for biomedical-inspired toxicological screening and supports the development of bio-hybrid and bio-inspired sensing platforms for environmental and health monitoring."