Medical Sleep Quality Monitoring Instrument Image Recognition Dataset

Currently, the medical health field is highly concerned with monitoring sleep quality. However, many traditional monitoring systems face challenges in accurately recognizing images of sleep states, primarily due to the lack of datasets and limitations in existing algorithms. Most solutions rely on shallow image processing techniques, making it difficult to ensure detection accuracy and precision. This dataset aims to enhance the precision and reliability of image recognition in medical sleep monitoring instruments, meeting specific needs of the medical health field. Data collection is carried out using high-precision medical monitoring instruments in simulated sleep environments to ensure data diversity and authenticity. Quality control includes multiple rounds of expert annotation, data consistency checks, and regular reviews. The annotation team consists of medical imaging experts and data scientists, involving a total of 50 participants. Data preprocessing steps include image denoising, contrast enhancement, and annotation correction. Finally, the data is stored in a hierarchical folder structure, with each image accompanied by detailed metadata.