Pubic Symphysis-Fetal Head Segmentation and Angle of Progression

  • ⏲ Timeline


    ✅3 Jan 2022: Our paper describing the FH-PS-AOP dataset is published (Open Acess)📖

    ✅ 5 Jan 2023: Release of Public Training set (also called Set 1) on Zenodo🗂

    ✅ 30 April 2023: Accepting submissions to the Preliminary Test Phase 🤸‍♀️

    ✅  30 August 2023: Closing submissions to the Preliminary Test Phase 🛑

    ✅ 1 Sep 2023: Accepting submissions to the Final Test Phase (Each individual or team can submit a maximum of 2 times)🤸‍♂️

    ✅ 20 Sep 2023: Closing submissions to the Final Test Phase🛑

    ✅ 8 Oct 2023: 🎉Winners of the FH-PS-AOP challenge are publicly announced!🎉

    🏆 Prizes 🏆


    🥇1st place (Gregor Köhler and Raphael Stock): 3,000 RMB

    🥈2nd place (Marawan Elbatel): 2,000 RMB

    🥉3rd place (Yaoyang Qiu ): 1,000 RMB

    🥉4th-7th (Gongping Chen, Lei Zhao, Fangyijie Wang, Hongkun Sun and Pengzhou Cai ): 500 RMB


  • Read this paper to find out more about the FH-PS-AOP dataset.

  • This is a semantic segmentation challenge. The task is to segment fetal head (FH)-pubic symphysis (PS) and then compute angle of progression (AOP) based on the segmented FH-PS.

  • Training data consisting of 4000 manually segmented cases is publically available on Zenodo. 

  • Our challenge has two phases (check the timeline and instructions on how to prepare your algorithm submission):

    1. Preliminary Test Phase (401 test cases): Participants develop their methods and test them on a subset of the test set. 

    2. Final Test Phase (700 test cases): Participants submit their final methods and test them on a full test set. 

  • NOTE: provided US images are not yet registered. We decided to leave this fundamental step to the participants as this can be an important methodological contribution that we did not want to bias in any way.

    🎯 Problem statement 🎯


    FH-PS segmentation could further determine fetal head descent.

    Current obstetric practice strives to avoid difficult vaginal deliveries and the clinician's skill resides mainly in precisely identifying the fetal head station at which forceps or ventouse should be applied. Digital transvaginal examination remains the ‘gold standard’ for obtaining this information, but it is a subjective evaluation with several limitations.

    Recent studies have shown that transperineal ultrasound imaging in the mid-sagittal plane might allow objective quantification of the level of fetal head descent in the birth canal by measurement of measurements of angle of progression (AOP). AOP is the angle between a straight line drawn along the longitudinal axis of the pubic symphysis (PS) and a line drawn from the inferior edge of the PS to the leading edge of the fetal head (FH). FH-PS segmentation is a prerequisite for automatic estimation of parameters

    Figure 1. Example of reference organ-at-risk (OAR) segmentations, displayed as color-coded three-dimensional binary masks.

    👩‍🎓 The FH-PS-AOP Challenge👨‍🎓


    The task of the FH-PS-AOP grand challenge is to automatically segment 700 FH-PSs from transperineal ultrasound images in the devised Set 2 (test set), given the availability of Set 1, consisting of 401 images.

    Set 2 is held private and therefore not released to the potential participants to prevent algorithm tuning, but instead the algorithms have to be submitted in the form of Docker containers that will be run by organizers on Set 2. The challenge is organized by taking into account the current guidelines for biomedical image analysis competitions, in particular the recommendations of the Biomedical Image Analysis Challenges (BIAS) initiative for transparent challenge reporting.

    🎇 Motivation of the FH-PS-AOP Challenge:🎇


    • To promote the development of new and application of existing state-of-the-art fully automated techniques for FH-PS segmentation from US images. 

    • To serve as a benchmark dataset for objective comparison of new methods for FH-PS segmentation.

    • To encourage the development of novel general-purpose methods for the AOP estimation.