Obstructive sleep apnea-hypopnea syndrome (OSAHS) is a significant public health concern. Wearable devices, such as smart rings and smart watches, provide a more convenient and less invasive alternative to traditional polysomnography (PSG) testing for detecting OSAHS. Smart rings, which collect signals from the finger base, offer superior signal quality and robustness compared to wrist signals from smart watches, making them a more convenient option for OSAHS detection. However, there is currently no research validating the performance of smart ring-based OSAHS detection. To our best knowledge, we are the first to investigate the feasibility and effectiveness of detecting OSAHS using smart rings. We analyzed 58 subjects, each wearing both a PSG device and RingConn smart rings for overnight OSAHS monitoring. Additionally, we propose a novel deep learning model to address the challenges posed by the distribution gap between photoplethysmography and blood oxygen saturation in OSAHS detection modeling, while efficiently capturing local and global features. Experimental results have demonstrated that smart rings detected OSAHS in good agreement with PSG, with a correlation coefficient of r =0.93. Our model also outperforms state-of-the-art methods.