AEHRC wins international competition in machine learning

If you’ve noticed a few of our people with bags under their eyes over the past weeks, it’s because our medical image analysis team has been working around the clock on their projects for ImageCLEF Medical 2022.

The good news is–their hard work has paid off and we have placed 1^st and 3^rd in two projects comprising the competition, which features participants from around the globe!

Our medical analysis team came 1^st in the “detect tuberculosis from 3D chest CT images” task and 3^rd in the “predict concepts from medical images/ generate captions from medical images”.

This year, the aim of the competition was to broaden the evaluation of technologies for annotation, indexing and retrieval of visual data.

It’s hoped this will provide information access to large collections of images in various usage scenarios and application domains: medical, nature, Internet and social media around the world.

Participants came from a large variety of fields including information retrieval (e.g., text, vision, audio, multimedia, social media, sensor data), machine learning, deep learning, data mining, natural language processing, image and video processing; with special emphasis on the challenges of multi-modality, multilinguality, and interactive search.

Our participation in this year’s competition was preceded last year by Aaron Nicholson, who entered solo. This year, Aaron brought an entire team together and co-ordinated efforts, no doubt with matchsticks holding up his eyelids! Congratulations to the entire team: Hollie Min, Ash Gillman, Aaron Nicolson, Leo Lebrat, Bevan Koopman, Bowen Xin and Jason Dowling.

The results of the campaign are published in the working notes proceedings, published by CEUR Workshop Proceedings (CEUR-WS.org). Selected contributions are invited for publication in next year’s Springer Lecture Notes in Computer Science (LNCS), together with the annual lab overviews.