Accurate detection of cochlear outer and inner hair cells is essential for diagnosing auditory impairments. This study presents a deep learning approach that integrates the object detection capabilities of You Only Look Once (YOLO) v11 with the segmentation proficiency of the Segment Anything Model 2 (SAM2) to identify missing cochlear hair cells in high-resolution microscopic images. YOLO v11 features an enhanced backbone and neck architecture that improves feature extraction for precise object detection. Its optimized design achieves a higher mean Average Precision (mAP) on the COCO dataset while utilizing 22% fewer parameters than its predecessor, YOLO v8m, thereby enhancing computational efficiency without compromising accuracy. In this study, YOLO v11 is employed to detect and localize existing cochlear outer and inner hair cells within microscopic images, providing bounding boxes that indicate the presence and positions of these cells. To identify missing hair cells, SAM2 is applied directly to detect regions where outer and inner hair cells are absent. SAM2 extends the capabilities of its predecessor by incorporating a per-session memory module that captures information about the expected morphology of hair cells, enabling it to identify gaps or abnormalities in the cellular arrangement. This architecture allows SAM2 to handle complex visual data through a unified, promptable model that supports real-time processing and zero-shot generalization. By generating precise segmentation masks of missing areas, SAM2 facilitates the direct identification of regions affected by hair cell loss. Experimental results demonstrate that the integration of YOLO v11 and SAM2 significantly enhances the accuracy and efficiency of detecting missing cochlear hair cells. The combined approach leverages YOLO v11's real-time object detection capabilities and SAM2's proficiency in identifying absent structures, providing a comprehensive tool for auditory research.