AutoLaparo: A New Dataset of Integrated Multi-tasks for 
Image-guided Surgical Automation​ in Laparoscopic Hysterectomy

Computer-assisted minimally invasive surgery has great potential in benefiting modern operating theatres. The video data streamed from the endoscope provide rich information to support context-awareness for next-generation intelligent surgical systems. To enhance the surgical scene understanding towards image-guided automation, one promising solution is to rely on learning-based methods, which highly relies on large-scale, high-quality and multi-task labelled data. To address this issue, we present and release the AutoLaparo dataset.

AutoLaparo is a new dataset of integrated multi-tasks for image-guided surgical automation in laparoscopic hysterectomy. The dataset is developed based on full-length videos of entire hysterectomy procedures. Specifically, three different yet highly correlated tasks are formulated in the dataset, including surgical workflow recognition, laparoscope motion prediction, and instrument and key anatomy segmentation. In addition, experimental results with state-of-the-art models are provided as reference benchmarks for further model developments and evaluations on this dataset.

[paper] [code]