Home Research and Innovation Projects iPATH: Smart Network Center for Digital Pathology
iPATH: Smart Network Center for Digital Pathology
Date: 2021-2023
Sectors

eHealth and medical care

Services

Research and Development under Contract

Departments

Computer Vision, Interaction and Graphics

Competences

Artificial intelligence, machine learning, and decision support systems
Computer vision and image processing

 

FRAMEWORK

The use of digital imaging in pathological analysis is increasingly crucial, and it is urgent and timely to develop solutions to support diagnosis and distributed review of cases. 

 

 

The area of ​​pathology is beginning to take the first steps towards digitalization, particularly in the use of specialty standards, this is due to the reduced number of pathologists per capita and the increase in requests for clinical services.

PROPOSED SOLUTIONS

The iPATH project aims to create a cloud service platform for the archiving, sharing, visualization and intelligent analysis of Pathological Anatomy image studies, a medical specialty dedicated to diagnosing pathologies based on the microscopic analysis of cells and fabrics.​​​​

 

 

It is essential to develop solutions that enable distributed scanning of samples with a centralized archive and that allow for review using intelligent decision support tools, validated in a clinical context.

CCG/ZGDV CONTRIBUTION

CCG/ZGDV participates in the iPATH project, through the R&I department, CVIG, with the development of advanced decision support modules, through a mitosis detection system, applying artificial intelligence.

 

 

 

The PathoBox platform provides a centralized archive for digital pathology images, 100% web-based. The doctor can use the intelligent tools available in the system to draw up a complete clinical report, from the image acquisition stage to remote clinical review.

The annotation of the different regions that make up these images, identifying the type of cells present and the type of diseases involved, will feed into an artificial intelligence network. This network allows for the automatic counting of mitoses, a time-consuming process for pathologists and a determining factor in different oncological areas.

This is a successful partnership between CCG/ZGDV, BMD Software, and FMUC (Faculdade de Medicina da Universidade de Coimbra), which will continue to work together to address the significant challenges inherent to digital histopathology in the future.