This study investigates the integration of attention mechanisms within deep learning architectures, specifically targeting the segmentation of bacilli microorganisms in microscopy images. We introduce novel modifications to attention blocks, tailored to address challenges inherent to microscopy imaging, such as blurred boundaries and indistinct microorganism features. Our experimental results demonstrate significant performance improvements compared to baseline models, with the mean Intersection over Union (mIoU) increasing from 0.844 to 0.966 and the Dice coefficient improving from 0.887 to 0.978 on the Microscopy Bacilli Images Dataset (MBID). These advancements underscore the potential of specialized attention-based architectures to enhance diagnostic accuracy and operational efficiency in biomedical image analysis.