AI-powered software represents a significant shift in the examination of pathological processes

The Cima University of Navarra (Spain) has unveiled the outcomes of a decade-long international competition aimed at enhancing the investigation of cell migration. The Cell Tracking Challenge findings validate that artificial intelligence-driven software heralds a transformative shift in the exploration of both regular cell biology and abnormal processes like cancer or tissue regeneration.

Cellular movement is a fundamental aspect of various biological processes, such as embryonic development, where cells exhibit remarkable mobility. Understanding the intricate journey of cells from initial stages to the complete formation of an embryo, as well as identifying the cellular origins of different organs based on their migration within the embryo, is crucial. Moreover, cell migration plays a significant role in pathological scenarios like cancer development and metastasis, as well as the healing process of wounds.

Traditionally, researchers have relied on labor-intensive and often impractical visual methods to study cell movement. However, recognizing these limitations, a decade ago, we initiated a challenge to foster the development of computer programs capable of automatically tracking cells. Over the years, we have built an extensive and diverse video database specifically designed for testing these programs. Additionally, we have established objective criteria and shared metrics to evaluate the accuracy of the various programs received.

Dr. Carlos Ortiz de Solórzano, the director of the Biomedical Engineering Program at the Cima University of Navarra and coordinator of this international competition, explains that the results of this challenge have been published in the latest issue of the scientific journal Nature Methods.

Artificial Intelligence

A total of 80 programs from 50 research groups around the world were assessed by the challenge organizers, representing collaborative teams from Spain, Germany, the Czech Republic, the United States, and Australia.

''Among the data that the participants have analyzed automatically, there are large videos (more than 300GB), such as the one containing the embryonic development of the red flour beetle (Tribolium castaneum) or other similar embryonic models widely used in biomedical research. Due to their size and complexity, these videos test the ability of programmers to manipulate and analyze large amounts of data,"

Among the noteworthy findings by the challenge organizers is the superior performance of artificial intelligence-based methods in cell recognition and tracking compared to traditional techniques. Recognizing the data-intensive nature of these methods, the scientific community has been provided with a valuable resource known as the "Silver Ground Truth." This comprehensive dataset has been automatically annotated using a combination of the most effective solutions submitted by participants, as explained by the project coordinator.

Building upon the achieved results, the challenge will persist in gathering new programs to explore the accurate detection of cell divisions and the reconstruction of complete cell trajectories.

Source : Cima University of Navarra