The four grand challenges chosen for ISBI 2017 are listed below.
1) LiTS – Liver Tumor Segmentation Challenge
The liver is a common site of primary (i.e. originating in the liver like hepatocellular carcinoma, HCC) or secondary (i.e. spreading to the liver like colorectal cancer) tumor development. Due to their heterogeneous and diffusive shape, automatic segmentation of tumor lesions is very challenging. Until now, only interactive methods achieve acceptable results on segmenting liver lesions. With our challenge we encourage researcher to develop automatic segmentation algorithms to segment liver lesions in contrast-enhanced abdomen CT scans. The data and segmentations are provided by serveral clinical sites, scanners and protocols. The challenge is organized by the image-based biomedical modeling Group (IBBM) and the Institute of Radiology of Technical University of Munich, Ludwigs Maximilian University, Radboudumc, Polytechnique Montréal, Tel Aviv University and IRCAD.
Point of contact: Patrick Christ (firstname.lastname@example.org)
2) A Grand Challenge for Tissue Microarray Analysis in Thyroid Cancer Diagnosis
The tissue microarray (TMA) represents a high throughput technology for the assessment of new and potentially valuable tissue biomarkers, markers that could be of value in diagnosis, predicting clinical outcome and response to therapy. By generating an array of several hundred small tissue samples on a single slide, biomarker expression can be evaluated across samples in a much more rapid and cost-effective fashion. Protein-related biomarkers are commonly evaluated using IHC. We have built a TMA database for thyroid cancer with clinical diagnosis information of 154 patients and 28 tissue microarrays, including 14 H&E tissue microarrays and 14 IHC tissue microarrays with BRAF; for each patient, there are four tissue cores, including three tumour ones and one normal tissue core. The goal of this challenge is to build predictions models that given the H&E morphological patterns, BRAF protein expression and patients’ background information, such as sex, age and the thyroid cancer type, produces similar results as in clinical diagnosis in five parameters, including Size, Extension, N, Stage and BRAF. The challenge participating teams will be invited to contribute to a joint journal paper describing and summarizing the challenge outcome, which will be submitted to a high-impact SCI journal in the field.
Point of contact: Prof. Ching-Wei Wang, Ph.D. (email@example.com)
3) Skin Lesion Analysis Towards Melanoma Detection
Skin cancer is a major public health problem, with over 5 million newly diagnosed cases in the United States each year. Melanoma is the deadliest form of skin cancer, responsible for over 9,000 deaths each year. As pigmented lesions occurring on the surface of the skin, melanoma is amenable to early detection by expert visual inspection. It is also amenable to automated detection with image analysis.
The International Skin Imaging Collaboration (ISIC) is an international effort to improve melanoma diagnosis. (See description at http://www.isdis.net/index.php/isic-project). The current database contains over 10,000 dermoscopic images. Additional images are being added to the archive in an ongoing fashion. A subset of the images have undergone annotation and markup by recognized skin cancer experts. These markups include dermoscopic features (i.e., global and focal morphologic elements in the image known to discriminate between types of skin lesions).
There will be 3 reference standard annotations for the 3 sub-challenges:
1) Segmentation: Lesions have been segmented against background normal skin and miscellaneous structures by expert dermatologists. 2) Dermoscopic Features: Lesions have been locally annotated for clinical dermoscopic features by expert dermatologists. 3) Disease State: The gold standard for diagnosis of skin lesions is pathology. Images in the ISIC archive have been derived from centers with expert pathology that can be deemed the gold standard. Benign lesions included in the archive without benefit of pathology diagnosis are reviewed by multiple experts and only included in the event of unanimous clinical diagnosis.
Point of contact: Aadi Kalloo (firstname.lastname@example.org)
Digital pathology is a new, rapidly expanding field in medical imaging. In digital pathology whole-slide scanners are used to digitize glass slides containing tissue specimens at high resolution. The availability of digital images has garnered the interest of the medical image analysis community, resulting in increasing numbers of publications on histopathologic image analysis. Last year at ISBI, we organized the highly successful CAMELYON16 grand challenge (http://camelyon16.grand-challenge.org), in which as many as 23 research groups participated by submitting their results. The task they addressed is of high clinical significance: detection of cancer metastases in lymph nodes in hematoxylin and eosin (H&E) stained whole-slide images. Solving this challenging task will help pathologists in efficiently and accurately staging breast cancers. This was the first challenge ever using whole-slide images in histopathology, having participants download over 500GByte of data. This year, CAMELYON17 (http://camelyon17.grand-challenge.org), will build on the success of its predecessor and will invigorate the challenge by moving from slide level analysis to patient level analysis (i.e. combining multiple slides into one outcome). This will bring the efforts closer to direct usefulness in a clinical setting.
Point of contact: Oscar Geessink (email@example.com)
We look forward to your participation in the ISBI 2017 challenges! When registration for ISBI 2017 opens you will be able to register your interest in attending the challenges via the registration form.
- Stephen Aylward, Kitware , USA
- Bram Van Ginneken, Radboud University Medical Center, NL
- Adrienne Mendrik, Netherlands eScience Center, NL
The deadline to submit proposals for organising challenges was September 29th 2016.