Digital tools with AI for the improvement of the pathology imaging-based workflows

The main objective of PATHFLOW is to demonstrate how the use of artificial intelligence techniques and digital tools can lead to greater efficiency when the biotechnology and health industry apply them to their processes related to the management of samples and tissues. Although advanced technology is very present in this sector, there are still some processes within the usual workflow that require manual and tedious interpretation of samples by laboratory personnel or external personnel.

Vicomtech participates in this project by contributing its knowledge and researching artificial intelligence models for lung pathology image analysis and demonstrating how this analysis, together with a digital workflow, helps improve efficiency in the laboratory.

PATHFLOW researches the way in which image processing techniques based on artificial intelligence can be applied to different processes and pathologies - such as lung cancer - in order to evaluate the capacity of technology to support those tasks that are currently require a high dedication of highly qualified personnel.

To this end, research is carried out on artificial intelligence models for the analysis of digitized images of lung tissue samples from biopsies and the evaluation of end-to-end workflows is carried out from the taking of samples to the stages of elaboration of molecular tests oriented to personalized medicine. In addition, the evaluation of the tool that incorporates the models for the decision or initial screening of lung tissue samples is carried out in order to determine its viability for use in subsequent molecular diagnostic tests.

Therefore, the project will be able to add progress by:

• The identification of stages that can be improved by digitalization in an end-to-end evaluation of the workflows of the processes related to pathological anatomy, considering the flows related to inter and intra-laboratory.
• The development and application of artificial intelligence models that can support the screening to be carried out in the stages prior to carrying out molecular tests based on the digital image obtained from the sample
• The evaluation and preliminary tests of potential in saving time and improving the process, applied to the use case of tests for prognosis and treatment of lung cancer.

This project is coordinated by Arahealth, and has the participation of Vicomtech, Inycom, Fenomatch and Pangaea Oncology This project has been supported by the Ministry of Industry, Commerce and Tourism as well as the European Union through the Recovery, Transformation and Resilience Plan and has received funding from the aforementioned Ministry within the AEI support program to contribute to the improvement of the competitiveness of the Spanish industry.