For women, cervical cancer is one of the most common cancers globally and is now screened using state-of-the-art technologies. In this field, artificial intelligence also offers a number of innovative opportunities for medicine to prevent the disease.
The most recent scientific advances in the field were presented in Houston at IEEE ISBI 2025 [International Symposium on Biomedical Imaging], one of the world's most prestigious scientific events on medical imaging. During the course of this conference, researchers from the University of Debrecen, Eötvös Loránd University and the University of Göttingen participated in the Pap Smear Cell Classification Challenge, where they submitted innovative solutions that could facilitate the systematic use of AI-based automated screening systems in everyday clinical trials.
Among the members of the board of the presentation section of the competition, there were several Hungarian researchers, including András Hajdu, Dean of the Faculty of Informatics at the University of Debrecen, and Balázs Harangi, Associate Professor of the Department of Data Science and Visualization at the Faculty of Informatics, UD.
During the various stages of the challenge, there were more than forty solutions produced in total. Sixteen of these were then sent to the final evaluation phase, with ten research teams remaining to submit their final papers for publication. In the end, seven of these papers were finally presented at the ISBI conference.
“Currently, the members of the organizing committee are working on a joint scientific publication summarizing the methods presented during the challenge. This summary paper will be published in a high-prestige international journal. While the competition and the conference produced important scientific results, it also laid the foundations for valuable international collaborations. Both the competition and the conference were an excellent opportunity for our faculty to have fruitful discussions with the challenge participants, to build professional relationships and to prepare future cooperation opportunities with them,” said Dean András Hajdu.
The papers presented at the Pap Smear Cell Classification Challenge were produced through support from prestigious international universities and research institutions. These included among others the University of Tokyo and its hospital, ShanghaiTech and Shanghai Jiao Tong University, the campuses of the University of Ghent in Belgium and South Korea, the Indian Institute of Technology in Kharagpur, Université Claude Bernard Lyon and INSA Lyon in addition to US universities such as the University of South Dakota or Massey University of New Zealand.
For further information and details of the competition, visit the official ISBI 2025 website (at https://biomedicalimaging.org/2025/challenges/).
Press Office, UD - BZ
New solutions for the automated classification of cervical cell images were developed by experts from some of the world's best universities at a scientific competition held in Houston by researchers from the University of Debrecen, Eötvös Loránd University and the University of Göttingen. By the end of what is called Pap-kenet Sejtosztályozási Kihívás [Pap Smear Cell Classification Challenge], as many as ten research teams completed their papers presenting their own solutions.
