Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Comprehensive Robotic Cholecystectomy Dataset (CRCD): Integrating Kinematics, Pedal Signals, and Endoscopic Videos

Ki-Hwan Oh, Leonardo Borgioli, Alberto Mangano, Valentina Valle, Marco Di Pangrazio, Francesco Toti, Gioia Pozza, Luciano Ambrosini, Alvaro Ducas, Milos Zefran, Liaohai Chen, Pier Cristoforo Giulianotti

TL;DR

A novel dataset recorded during ex vivo pseudo-cholecystectomy procedures on pig livers, utilizing the da Vinci Research Kit, highlighting the dataset's potential for advancing automation in surgical robotics.

Abstract

In recent years, the potential applications of machine learning to Minimally Invasive Surgery (MIS) have spurred interest in data sets that can be used to develop data-driven tools. This paper introduces a novel dataset recorded during ex vivo pseudo-cholecystectomy procedures on pig livers, utilizing the da Vinci Research Kit (dVRK). Unlike current datasets, ours bridges a critical gap by offering not only full kinematic data but also capturing all pedal inputs used during the procedure and providing a time-stamped record of the endoscope's movements. Contributed by seven surgeons, this data set introduces a new dimension to surgical robotics research, allowing the creation of advanced models for automating console functionalities. Our work addresses the existing limitation of incomplete recordings and imprecise kinematic data, common in other datasets. By introducing two models, dedicated to predicting clutch usage and camera activation, we highlight the dataset's potential for advancing automation in surgical robotics. The comparison of methodologies and time windows provides insights into the models' boundaries and limitations.

Comprehensive Robotic Cholecystectomy Dataset (CRCD): Integrating Kinematics, Pedal Signals, and Endoscopic Videos

TL;DR

A novel dataset recorded during ex vivo pseudo-cholecystectomy procedures on pig livers, utilizing the da Vinci Research Kit, highlighting the dataset's potential for advancing automation in surgical robotics.

Abstract

In recent years, the potential applications of machine learning to Minimally Invasive Surgery (MIS) have spurred interest in data sets that can be used to develop data-driven tools. This paper introduces a novel dataset recorded during ex vivo pseudo-cholecystectomy procedures on pig livers, utilizing the da Vinci Research Kit (dVRK). Unlike current datasets, ours bridges a critical gap by offering not only full kinematic data but also capturing all pedal inputs used during the procedure and providing a time-stamped record of the endoscope's movements. Contributed by seven surgeons, this data set introduces a new dimension to surgical robotics research, allowing the creation of advanced models for automating console functionalities. Our work addresses the existing limitation of incomplete recordings and imprecise kinematic data, common in other datasets. By introducing two models, dedicated to predicting clutch usage and camera activation, we highlight the dataset's potential for advancing automation in surgical robotics. The comparison of methodologies and time windows provides insights into the models' boundaries and limitations.
Paper Structure (17 sections, 1 equation, 8 figures, 5 tables)

This paper contains 17 sections, 1 equation, 8 figures, 5 tables.

Table of Contents

  1. Introduction
  2. Dataset Composition
  3. Stereo Endoscopic Images
  4. Pedals
  5. Kinematic Dataset
  6. Surgical Task
  7. Setup and repetitions
  8. Description
  9. Preliminary Work
  10. Pedal Prediction
  11. Dataset Preprocessing
  12. Trained Classifiers
  13. Model Evaluations
  14. Training Tissue Segmentation Models
  15. Generating segmentation dataset
  16. ...and 2 more sections

Figures (8)

  • Figure 1: Sample of the stereo endoscopic images.
  • Figure 2: Circuit of the monopolar pedal and the electrosurgical generator connected to the Arduino Uno Device.
  • Figure 3: Circuit of Fig. \ref{['fig:mp_circuit']} when the monopolar output is deactivated (a), and when it is activated (b).
  • Figure 4: The setup showing how our custom-calibrated kinematics work. The transformations are shown based on the direction of the arrows and eventually, they are used to find the transformation between the ECM tip and PSM tip.
  • Figure 5: A sample from the recorded 3D trajectory of the (a) MTMR and (b) PSM1.
  • ...and 3 more figures