Preoperative to Intraoperative Laparoscopy Fusion (P2ILF)

News 📌

  • Due to high demand of this challenge, upon committee meeting review we have decided to extend data download and team registration till 21st July 2022
  • Teams can now ask questions directly - please do not contact organisers but rather post your queries here (
  • Training data (patient) is now available upon request for challenge participants!!! (release on 13th June, 2022) - click to see access details
  • Training data download open till 30th June, 2022   21st July 2022 NO LONGER ACCEPTING ANY DATA REQUEST!!!
  • New team enrolment closes 30th June, 2022 - 21st July 2022 (due to high demand, upon committee meeting we have decided to extend this!!! - no extension post this date!)  NO LONGER ACCEPTING ANY DATA REQUEST!!!
  • (upcoming!!!)   Participants enrolled will be contacted for leaderboard participation briefing (online) - 4th July 2022 28th July 2022 
  • Urgent and mandatory for next stage: If you are ready to take the challenge then we require you to list your team here: (we will conduct an interactive session with participants!!! - We want to support only participants who are taking the challenge please!)
  • (upcoming!!!)   Phantom dataset is intended to be released on 25th July 2022  5th August (only invited participants!)
  • Final participating teams have been selected - No more team enrolment or request is valid. We will be not responding to any such requests!!!
  • (upcoming!!!) Metric code release
  • Winner prize worth up to $500 (sponsored by Intuitive Surgery)

Ongoing: Preliminary test phase started now!!! (3rd - 9th September)


Augmented reality (AR) in laparoscopic liver surgery needs key landmark detection in intraoperative 2D laparoscopic images and its registration with the preoperative 3D model from CT/MRI data. Such AR techniques are vital to surgeons as they enable precise tumor localisation for surgical removal. A full resection of targeted tumor minimises the risk of recurrence. However, the task of automatic anatomical curve segmentation (considered as landmarks), and its registration to 3D models is a non-trivial and complex task. Most developed methods in this domain are built around traditional methodologies in computer vision. This challenge is designed to challenge participants to deploy machine learning methods for two tasks - Task I: segmentation of five key anatomical curves from laparoscopic video images and 3D model, including ridge (L, R), ligament, silhouettes, liver boundary; Task 2: matching these segmented curves to the 3D liver model from volumetric data (CT/MR). 

Challenge tasks in detail:

Task 1: Segmentation of liver anatomical curves (landmarks)Participants are required to predict segmented curves of intra-operative laparoscopic video image frames and preoperative 3D model. These anatomical curves include silhouette, falciform ligament, ridge and liver boundary (see Figure below). Evaluation will be based on the F1-score and Hausdorff distance. A 2% tolerance for the predicted anatomical curves w.r.t ground truth will be used.

Task 2: 2D-3D fusion. Registration of anatomical curves in 2D laparoscopy to corresponding curves in 3D model is the main aim of this task. Participants are required to provide transformation matrixes for this task for which the re-projection error and target registration error (only for phantom data) will be computed.  

Data curation and annotation protocols are detailed here:

Organising team:

Sharib Ali, Yueming Jin,  Yamid Espinel López, and Adrien Bartoli

(Full committee details - Refer to committee page)

Follow us on twitter: p2ilf

Contact: @P2ILF


About Intuitive Surgical CompanyAt Intuitive, we are united behind our mission: we believe that minimally invasive care is life-enhancing care. Through ingenuity and intelligent technology, we expand the potential of physicians to heal without constraints.  As a pioneer and market leader in robotic-assisted surgery, we strive to foster an inclusive and diverse team, committed to making a difference. For more than 25 years, we have worked with hospitals and care teams around the world to help solve some of healthcare's hardest challenges and advance what is possible. 

This challenge is being sponsored by the Imaging and Intelligence team within Intuitive that focuses on exploring advanced computer vision and machine learning solutions that will enable next generation of da Vinci robots to be smarter and safer to improve patient outcome.  

Join a team committed to taking big leaps forward for a global community of healthcare professionals and their patients. Together, let's advance the world of robotic surgery.