Real-time Injection Therapy Monitoring Dataset

The current healthcare industry faces challenges like insufficient monitoring and low efficiency during injection therapy processes, especially in real-time monitoring of patients’ injections, lacking effective data support. Existing monitoring technologies often rely on manual observation, prone to omissions and misjudgments, making it hard to ensure the safety and effectiveness of the therapy. This dataset aims to provide high-quality real-time injection therapy monitoring data to support the development of object detection technology and enhance patient management levels. Data collection uses high-resolution medical camera equipment in a standardized medical injection room, ensuring that the captured images reflect real clinical scenarios. Quality control measures include multiple rounds of annotation, consistency checks among annotators, and expert reviews to ensure annotation accuracy and consistency. The data is stored in JPEG format, organized by patient group. The core advantage of the dataset lies in its high annotation accuracy (over 95%), introducing a new annotation method through deep learning technology that enhances data completeness and consistency. Additionally, this dataset effectively addresses insufficient monitoring, improving the safety of injection therapy and is expected to reduce error rates by up to 30%.