Overview

Stroke represents the second leading cause of mortality worldwide. The key component for immediate diagnosis is the localization (over CT scans) and delineation of lesions ( over MRI studies). The lesions are nonetheless poorly delineated, only visible at advanced stages, and analysis uses manual delineation.

APIS Challenge 2023 pretends to be a workshop that introduces a paired dataset of CT and ADC studies. The researchers are invited to propose computational strategies that approach paired data, during training, and deal with lesion segmentation over CT onset sequences. During training will be available annotated paired sequences (from one expert), and resultant segmentation for testing will be compared regarding two experts.

About

The APIS Challenge

The APIS Challenge is a workshop dedicated to stroke segmentation from paired (CT-ADC) dataset. For training, 60 paired studies with CT and ADC image modalities are released, with an associated delineation from an expert radiologist. Regarding the testing stage, a total of 40 studies will be used as validation.

We encourage participants to learn projections from ADC and CT sequences and to carry out predicted delineations over CT studies. The challenge claims to capture the generalization capability of proposed approaches, by measuring resultant segmentations from two independent expert radiologists with more than five years of experience. Optionally, authors may generate synthetic ADC images and submit them to the challenge to receive quantitative results from reconstruction metrics.

How to participate in APIS challenge?

If you are interested in participating, you are invited to download the training set in data section, which includes CT and ADC modalities of 60 stroke patients.

The website will perform the evaluation of the proposed methods via docker methodology. To complete a successful participation, participants are encouraged to submit an abstract, describing the employed method. All results will be publicly available in results section, only the top 10 results will be displayed in the website.

Each participant team are encouraged to submit their method as a poster or journal in any other conference after ISBI 2023 ends. Eventually, submissions will be included in the workshops post-proceedings and potentially compiled for a high-impact journal paper to summarise and present the findings.

Prizes

• The three top-performing teams are eligible to participate in a joint publication with the committee sumitting to an Open Access Journal. There is a fixed maximum of two authors per team. The committee will also invite teams that submit particularly novel solutions to join as co-authors.
• The 10 top-performing teams will have the opportunity to present their results at ISBI 2023

Data

The proposed challenge was elaborated by BIVL²ab (Biomedical, Imaging and Vision Learning Laboratory) from Universidad Industrial de Santander, in collaboration with FOSCAL (Fundacion Oftalmologica de Santander) and FOSUNAB (FOSCAL and Universidad Autonoma de Bucaramanga) clinics. The studies were captured using: two CT scanners, namely i) a Toshiba Aquilion 64 TSX-101 and ii) a Toshiba Aquilion one TSX-301/ac scanner, with 64 and 320 detectors, respectively. Also, one Toshiba Vantage Titan MRI scanner was employed for acquisition of different MRI modalities. Such scanners are located at FOSCAL and FOSUNAB clinics in Bucaramanga, Colombia. Spatial resolution for NCCT images was 512x512 px, and 256x256 for most of diffusion sequences.

The annotation of ischemic lesions was carried out by two neurointerventional radiologists with more than five years of experience. The annotation protocol involved the examination of DWI B-1000 and ADC images to delineate the damaged tissue in the brain. To achieve a more accurate diagnosis, experts had access to clinical variables previous to the manual delineation of each case. The lesion annotation process was performed on the ADC images by each radiologist individually using the MRICroGL software. Furthermore, the associated NCCT studies were skull-stripped and co-registered w.r.t. labeled ADC sequences. The following figure shows a preview of the image modalities and the respective segmentation mask. The delineated lesions show high variability in shape, size and number of consecutive slices where the lesion occurs.

Submission

How to submit?

Click the following link to enter in submission website. The credentials to login are send to the team leader email at the time of acceptance.

After you login with your credentials, you will be able to upload your model contained in a zip file following the next steps:

1. Select the "APIS ISBI 2023" option in the "Challenge to submit" selector.
2. Select the zip file that contains your Dockerfile and model weights (Remember to follow the docker template).
3. Press the "Submit" button and wait for your docker to be correctly uploaded (The status is displayed on the right window).
4. For a visual explanation, you can see the following video.

Docker template guidelines

In order to make a successful submission you can download and follow the comments instructions in the docker template zip which contains a Dockerfile and evaluate.py files with their respective instructions. The following are general rules to submit your solution in the challenges:

• Your zip must contain the Dockerfile and evaluate.py which have defined elements that mustn't be modified in order to run.
• You can attach inside the zip your model weights and model structure in any ".py", ".h5", ".pth", etc. There are no rules for the files contained in the zip file, except for Dockerfile and evaluate.py files.
• The image used in Dockerfile must be available in Docker Hub Repository.
• In case of any error during the execution of the uploaded zip, it will be shown in the status window on the right.
• You can upload as many solutions you want but only the best will be stored in the server. All the others solutions will have their results only.