Microfluidic-Based Analysis of Trichomonas Vaginalis Attachment and Drug Response Using AI-Based Imaging Algorithm

T. vaginalis is a protozoan parasite responsible for the most common non-viral sexually transmitted infection worldwide. Effective therapeutic evaluation is essential, particularly with rising concerns over drug resistance. In this study, we developed a microfluidic culture platform integrated with an AI-based imaging algorithm to quantify parasite attachment dynamics and assess drug efficacy. The platform enabled controlled exposure of T. vaginalis to human cancer cell lines (HeLa and BFTC905) under single and combined drug gradients of metronidazole and paromomycin. Real-time imaging and spatial color-coded map analysis revealed that metronidazole significantly reduced parasite attachment in a dose-dependent manner. Furthermore, combination therapy exhibited superior inhibitory effects compared to single-drug treatments, suggesting potential synergistic interactions. The platform demonstrated robust and reproducible results across different cell types, highlighting its value for high-throughput anti-parasitic screening. These findings support the application of combination regimens in managing T. vaginalis infections and establish the microfluidic-AI system as a promising tool for drug evaluation and therapeutic development.