CFP: 𝗠𝗘𝗗𝗜𝗤𝗔-𝗖𝗢𝗥𝗘 @ 𝗖𝗟𝗘𝗙 𝟮𝟬𝟮𝟲 | 𝗠𝘂𝗹𝘁𝗶𝗺𝗼𝗱𝗮𝗹 𝗥𝗲𝗮𝘀𝗼𝗻𝗶𝗻𝗴 & 𝗥𝗲𝗰𝗼𝗻𝗰𝗶𝗹𝗶𝗮𝘁𝗶𝗼𝗻 𝗶𝗻 𝗥𝗲𝗮𝗹-𝗪𝗼𝗿𝗹𝗱 𝗖𝗹𝗶𝗻𝗶𝗰𝗮𝗹 𝗪𝗼𝗿𝗸𝗳𝗹𝗼𝘄𝘀

[Asma Ben Abacha]

Dear all, 

We're excited to introduce the MEDIQA-CORE 2026 shared tasks, focused on advancing clinically grounded multimodal AI models built for real-world clinical workflows and evaluated on real clinical data.

 𝐓𝐚𝐬𝐤 𝟏: 𝐁𝐫𝐚𝐢𝐧 𝐓𝐮𝐦𝐨𝐫 𝐒𝐮𝐛𝐭𝐲𝐩𝐞 𝐂𝐥𝐚𝐬𝐬𝐢𝐟𝐢𝐜𝐚𝐭𝐢𝐨𝐧
Develop models that combine brain MRI images, H&E pathology slides, and pathology reports to predict tumor subtypes while being robust to missing modalities, reflecting how data actually becomes available in clinical practice.

 𝗧𝗮𝘀𝗸 𝟮: 𝗥𝗮𝗱𝗶𝗼𝗹𝗼𝗴𝘆 𝗥𝗲𝗽𝗼𝗿𝘁 𝗗𝗶𝘀𝗰𝗿𝗲𝗽𝗮𝗻𝗰𝘆 𝗔𝘀𝘀𝗲𝘀𝘀𝗺𝗲𝗻𝘁
Build multimodal systems that analyze 3D abdominal CT scans, preliminary radiology reports, and candidate report edits to assess image-text agreement, clinical severity, and discrepancy type, simulating real radiology review workflows.

We're looking forward to seeing the community's ideas and participation!

 𝗪𝗲𝗯𝘀𝗶𝘁𝗲: https://www.imageclef.org/2026/medical/mediqa-core ​

​

​​​

 𝗖𝗼𝗻𝘁𝗮𝗰𝘁: mediqa.o...@gmail.com
𝗠𝗘𝗗𝗜𝗤𝗔 𝗠𝗮𝗶𝗹𝗶𝗻𝗴 𝗟𝗶𝘀𝘁: https://lnkd.in/eGdPBwDf

 𝗞𝗲𝘆 𝗗𝗮𝘁𝗲𝘀:
 Registration closes: 𝐌𝐚𝐫𝐜𝐡 𝟐𝟏, 𝟐𝟎𝟐𝟔
 Test set release: April 23, 2026
 Run submission deadline: May 7, 2026
 Paper submission deadline: May 28, 2026
 CLEF 2026 Conference: September 21-24, 2026 (Jena, Germany)

𝗧𝗮𝘀𝗸 𝟭 𝗢𝗿𝗴𝗮𝗻𝗶𝘇𝗲𝗿𝘀:
• Asma Ben Abacha, Microsoft
• Wen-wai Yim, Microsoft
• Daniel K Low, University of Washington, Kurtlab
• Juampablo Heras Rivera, University of Washington, Kurtlab
• Mehmet Kurt, University of Washington, Kurtlab
• Jake Ruzevick, UW Medicine, Neurosurgery
• Daniel D. Child, UW Medicine, Pathology

𝗧𝗮𝘀𝗸 𝟮 𝗢𝗿𝗴𝗮𝗻𝗶𝘇𝗲𝗿𝘀:
• Asma Ben Abacha, Microsoft
• Wen-wai Yim, Microsoft
• Zhaoyi Sun, University of Washington
• Meliha Yetisgen, University of Washington
• Fei Xia, University of Washington