Challenges

Challenge 1: Classification of Normal versus Malignant Cells in B-ALL White Blood Cancer Microscopic Images

Shiv Kumar Gehlot <shivg@iiitd.ac.in>
Anubha Gupta <anubha@iiitd.ac.in>

Challenge website: https://competitions.codalab.org/competitions/20429

Computer assisted tools can provide cost effective and easily deployable solutions for cancer diagnostics. The aim of this challenge is to build a classifier for the identification of leukemic versus normal immature cells for while blood cancer, namely, B-ALL diagnostics. A dataset of cells with class labels, marked by the expert based on the domain knowledge, will be provided at the subject-level to train the classifier. This problem is interesting because the two cell types appear similar under the microscope and subject-level variability plays a key role. Hence, it is challenging to build a classifier that can yield good results on prospective data.

Challenge 2 Time-Lapse Cell Segmentation Benchmark

Michal Kozubek <kozubek@fi.muni.cz>

Challenge website: http://celltrackingchallenge.net

In 2012, Cell Tracking Challenge (CTC) was launched to objectively compare and evaluate state-of-the-art whole-cell and nucleus segmentation and tracking methods using both real (2D and 3D) time-lapse microscopy videos of cells and nuclei, along with computer generated (2D and 3D) video sequences simulating nuclei moving in realistic environments. To address numerous requests for benchmarking only cell segmentation methods (without tracking), we are launching now a new time-lapse cell segmentation benchmark on the same datasets (plus one new dataset).

Challenge 3: Multi-class artefact detection in video endoscopy

Sharib Ali <sharib.ali@eng.ox.ac.uk>
Felix Zhou <felix.zhou@ludwig.ox.ac.uk>

Challenge website: https://ead2019.grand-challenge.org/

Endoscopic Artefact Detection (EAD) is a core challenge in facilitating diagnosis and treatment of diseases in hollow organs. Precise detection of specific artefacts like pixel saturations, motion blur, specular reflections, bubbles and debris is essential for high-quality frame restoration and is crucial for realising reliable computer-assisted tools for improved patient care. The challenge is sub-divided into three tasks:

  1. Multi-class artefact detection: Localization of bounding boxes and class labels for 6 artefact classes for given frames.
  2. Region segmentation: Precise boundary delineation of detected artefacts.
  3. Detection generalization: Detection performance independent of specific data type and source.

Challenge 4: SegTHOR: Segmentation of THoracic Organs at Risk in CT images

Caroline Petitjean <caroline.petitjean@univ-rouen.fr>

Challenge website: https://segthor.grand-challenge.org/

The SegTHOR challenge addresses the problem of organs at risk segmentation in Computed Tomography (CT) images. In lung and esophageal cancer, radiation therapy planning begins with the delineation of the target tumor and healthy organs located near the target tumor, called Organs at Risk (OAR) on CT images. Routinely, the delineation is largely manual which is tedious and source of anatomical errors. In this challenge, the goal is to automatically segment 4 OAR: heart, aorta, trachea, esophagus. Participants will be provided with a training set 40 CT scans with manual segmentation. The test set will include 20 CT scans.

Challenge 5: Automatic Non-rigid Histological Image Registration (ANHIR)

Jiri Borovec (jiri.borovec@fel.cvut.cz)
Jan Kybic (kybic@fel.cvut.cz)
Arrate Munoz Barrutia (mamunozb@ing.uc3m.es)  

Challenge website: https://anhir.grand-challenge.org

In digital pathology, it is often useful to align spatially close but differently stained tissue sections in order to obtain the combined information. The images are large, in general, their appearance and their local structure are different, and they are related through a nonlinear transformation. The proposed challenge focuses on comparing the accuracy and approximative speed of automatic non-linear registration methods for this task. Registration accuracy will be evaluated using manually annotated landmarks. All methods are supposed to run fully automatically, with no image specific parameters.

Challenge 6: MRI White Matter Reconstruction Challenge (MEMENTO)

Kurt Gregory Schilling <kurt.g.schilling.1@vanderbilt.edu>
Bennett A Landman <bennett.landman@vanderbilt.edu>

Challenge website: https://my.vanderbilt.edu/memento/

Diffusion MRI has emerged as a key modality for imaging brain tissue microstructural features, yet, validation is necessary for accurate and useful biomarkers. Towards this end, we present the two-year ISBI 2019/2020 diffusion Mri whitE Matter rEcoNstrucTiOn (MEMENTO) challenge. The first year is dedicated to designing the challenge, building the appropriate dataset(s), and making it available to the community. The challenge and participant submissions will take place in the second year, with the aim to evaluate and advance the state of the microstructural modeling field.

Challenge 7: Automatic Lung Cancer Detection and Classification in Whole-slide Histopathology

Zhang Li <zhangli_nudt@163.com>
Tao Tan <tao.tan911@gmail.com>

Challenge website: TBD

Digital pathology has been gradually introduced in clinical practice. Although the digital pathology scanner could give very high resolution whole-slide images (WSI) (up to 160nm per pixel), the manual analysis of WSI is still a time-consuming task for the pathologists. Automatic analysis algorithms offer a way to reduce the burden for pathologists. Our proposed challenge will focus on automatic detection and classification of lung cancer using Whole-slide Histopathology. This subject is highly clinical relevant because lung cancer is the top cause of cancer-related death in the world.

Challenge 8: CHAOS : Combined (CT-MR) Healthy Abdominal Organ Segmentation

Alper Selver <alper.selver@deu.edu.tr>

Challenge website: https://chaos.grand-challenge.org/

CHAOS has two separate but related aims:

  1. Segmentation of liver from computed tomography (CT) data sets, which are acquired at portal phase after contrast agent injection for pre-evaluation of living donated liver transplantation donors (15 training + 15 test sets).
  2. Segmentation of four abdominal organs (i.e. liver, spleen, right and left kidneys) from magnetic resonance imaging (MRI) data sets acquired with two different sequences (T1-DUAL and T2-SPIR) (15 training + 15 test sets).

There will be five competition categories as Liver Segmentation (CT only, MR only, CT-MRI combined) and abdominal organ segmentation (MRI only, CT-MRI combined).

Challenge 9: PALM: PathologicAL Myopia detection from retinal images

Yanwu (Frank) Xu <xuyanwu@baidu.com>
Hrvoje Bogunovic <hrvoje.bogunovic@meduniwien.ac.at>

Challenge website: https://palm.grand-challenge.org/

The PALM challenge focuses on investigation and development of algorithms associated with diagnosis of Pathologic Myopia (PM) and segmentation of lesions in fundus photos from PM patients. Myopia is currently the ocular disease with the highest morbidity. About 2 billion people have myopia in the world, 35% of which are high myopia. High myopia leads to elongation of axial length and thinning of retinal structures. With progression of the disease into PM, macular retinoschisis, retinal atrophy and even retinal detachment may occur, causing irreversible impairment to visual acuity. There are typical signs of PM in fundus photos, including atrophy, lacquer crack, etc. Neural networks can be trained to help screening and grading PM in large populations.