RIS Competition 2027

Competition on the development of new analytical methods in medicine

ORGANIZER

DMFA Slovenia
PS Medical Physics, FMF UL

INTRODUCTORY LECTURE

TBA

Zoom link

WHERE?

Online

Start of applications

TBA

End of applications

TBA

Publication of results

Level 1 – TBA
Level 2 – TBA

News

Date	Event
	Start of applications
	End of applications
	Introductory lecture
	SLING workshop
	Deadline for submission of first round tasks
	First round results published
	Deadline for appeals on first round results
	Announcement of official first round results
	Second round execution
	Second round results published
	Deadline for appeals on second round results
	Announcement of official competition results

Challenge

The theme for this year’s RIS competition is segmentation and staging of tumors for radiotherapy.

Vision and motivation

Radiotherapy is among the most widely used cancer treatment modalities, in which medical imaging plays a central role in patient management. Studies have shown promise for non-invasive tumor characterization using imaging data such as PET and CT. These modalities capture complementary metabolic and anatomical information about tumor biology. However, turning this information into treatment decisions is far from straightforward. Before radiation treatment, the tumor must be segmented. Different clinicians working on the same images often draw different boundaries, driven mainly by their judgment of how far the primary tumor has invaded the surrounding tissue and has spread to nearby lymph nodes. These judgments rest on several pieces of information from different imaging and clinical data. Combining such information consistently and efficiently is a hard problem, one that sits squarely at the intersection of physics, data science, and medicine.

This leads to the question behind this year’s competition: How closely can we approach these clinical judgments using imaging data alone? A model that approximates the conclusions could support clinical decision-making, speed up routine workflows, and bring greater consistency to areas that currently rely heavily on individual expertise.

Competition format

The competition is carried out in two rounds with an additional introductory seminar. At the introductory seminar, participants become familiar with the field and the clinical background of the problem. They learn fundamental concepts that serve as a basis for developing algorithms applicable to a realistic medical problem.

The competition continues with the first, pre-selection round, where participants prepare a proposal of their solution supported by simplified examples. Based on the submitted results, innovation, suitability, and efficiency of the idea are evaluated. The best participants of the first round advance to the second round, which serves as the national competition.

In the second round, participants prepare their solution to the problem using modern analytical technologies, depending on the selected challenge. The fulfillment of the defined objectives is evaluated by testing solutions on real data that participants did not have access to. The solution is evaluated by a committee selected from Slovenian experts.

The format of the competition is a so-called hackathon, where participants are faced with an open, realistic problem and compete in developing the best solution. They work in teams appropriate to the defined problem. They use their own equipment; in case of issues, each case is handled individually by the organizing committee, which can provide limited material resources at the Faculty of Mathematics and Physics, University of Ljubljana.

Funding

The competition is co-financed from public funds (ARIS P1-0389).

Flyer