PathAI project supported by the Innovation Fund of Montenegro
THE PROBLEM / CHALLENGE
Histopathological analysis — the microscopic examination of tissue samples — remains the gold standard for cancer diagnosis. In the case of adenocarcinoma, one of the most common forms of malignant tumours, accurate and timely interpretation of tissue samples is critical for determining the appropriate course of treatment. However, this process places significant demands on the expertise and time of specialist pathologists, whose capacity is finite. With growing volumes of samples and increasing diagnostic complexity, there is a clear need for tools that can support pathologists in their work — helping to improve consistency, reduce turnaround time, and enhance the reliability of the diagnostic process.
Developing an AI model capable of assisting in the analysis of histopathological images is computationally intensive. High-resolution whole-slide images generate massive datasets, and training deep learning models on such data requires substantial GPU compute resources that go far beyond what is available locally to an SME. The Montenegrin company IT Advanced Services (ITAS), already in the implementation phase of the PathAI project — co-financed by the Innovation Fund of Montenegro in partnership with the Institute of Pathology of the Clinical Centre of Montenegro (KCCG) — identified this computational bottleneck as the critical barrier to advancing the project at the scale and speed required.
SOLUTION
To address the GPU resource gap, ITAS turned to NCC Montenegro for guidance. Following a dedicated technical consultation with Prof. Enis Kočan of the NCC Montenegro team, ITAS prepared and submitted an application to the EuroHPC JU Development Access Call, securing access to high-performance GPU computing infrastructure at the European level. The NCC Montenegro team provided full support throughout the application process — from explaining the call requirements and available systems to advising on the technical specification of resource needs. During the same consultation, ITAS was also informed about the FFPlus open call for Business Experiments as a potential avenue for further EU-level support.
The PathAI project itself is conducted under the Innovation Fund of Montenegro’s Collaborative Grants Programme, with ITAS and KCCG as co-applicants. The research focuses exclusively on existing, archived, and fully anonymised pathohistological samples collected in the course of KCCG’s regular diagnostic practice — no new sampling is carried out and no patient-identifiable data is used at any stage. The AI model under development is designed to assist the pathologist in sample analysis, supporting — not replacing — expert clinical judgement. The project runs for 18 months.
BENEFITS
- Pathologist support, not replacement: The AI model is designed as a decision-support tool that assists specialist pathologists in reviewing histopathological samples, with the potential to improve consistency and help manage growing workloads in clinical pathology.
- EuroHPC access unlocked for a Montenegrin SME: With the hands-on support of NCC Montenegro, ITAS successfully navigated the EuroHPC JU Development Call process — gaining access to GPU compute resources that would otherwise have been entirely out of reach for a company of its size.
- Privacy-by-design from the ground up: The project uses only pre-existing, archived, and fully anonymised samples from KCCG’s Institute of Pathology. No patient-identifiable data is processed or transferred at any point, ensuring full GDPR compliance and alignment with applicable medical data protection standards.
- Layered national and European funding: The combination of Innovation Fund of Montenegro co-financing and EuroHPC JU compute access demonstrates how Montenegrin SMEs can stack national and European support instruments to tackle research challenges that would be unachievable through either source alone.
- Public-private collaboration in healthcare innovation: The partnership between ITAS (a private SME) and KCCG (a public clinical institution) creates a direct bridge between AI development capability and real-world clinical data and expertise — a model for responsible medical AI development in Montenegro.
- A gateway to further EU opportunities: NCC Montenegro’s introduction of the FFPlus call during the consultation illustrates how NCC support extends beyond a single access application, opening doors to a broader ecosystem of European funding and collaboration for innovative Montenegrin companies.
NCC Montenegro provided direct technical support to ITAS in preparing the EuroHPC JU Development Call application, enabling this Montenegrin SME to access European supercomputing infrastructure in support of a nationally co-funded medical AI project.