Bridging the AI and HPC Competence Gap in Healthcare: New Framework for Digital Innovation

As artificial intelligence (AI) and high-performance computing (HPC) create new opportunities for healthcare innovation, developing the right digital competencies is becoming just as important as access to advanced technologies and computing infrastructure.

A newly published study by researchers from Montenegro addresses this challenge by proposing a competence framework for AI and HPC adoption in healthcare, highlighting an important gap in existing research and professional skills development.

Based on a systematic literature review, the study identifies 22 competence categories across four dimensions: technical and data-centric skills, ethical and legal considerations, human-centred and interpersonal competencies, and organizational and leadership capabilities. The framework also considers different proficiency levels and professional roles.

A key finding is the lack of explicit HPC competencies in the reviewed literature. None of the 17 studies included in the review specifically addressed HPC skills, despite the growing importance of advanced computing for AI development, large-scale health data analysis, and digital healthcare innovation.

Beyond healthcare, the findings offer a valuable starting point for technology companies, training providers, and organizations developing AI-powered services. The proposed framework could support specialized training programmes, digital skills assessments, and the identification of competencies needed to adopt advanced AI and HPC solutions.

These findings closely align with the mission of NCC Montenegro within the EuroCC3 project, which works to strengthen national HPC and AI capacities through professional training, industry consultations, knowledge transfer, and access to European supercomputing infrastructure.

By connecting academic research with the practical needs of healthcare professionals, technology developers, and industry, the study provides useful evidence for future capacity-building activities and collaboration between academia and industry.

The research was published in the international journal Technology and Health Care, under the title “Competences for Artificial Intelligence in Healthcare and the High-Performance Computing Competence Gap: A Systematic Literature Review and Proposed Framework”, authored by Milica Milošević, Sandra Tinaj, Tomo Popović, Milica Vukotić, and Bojana Mališić.

The publication represents another contribution to Montenegro’s growing AI and HPC knowledge base, supporting the broader goal of translating scientific expertise into skills, innovation opportunities, and sustainable digital transformation.

Read the full publication: https://doi.org/10.1177/09287329261504912

3rd HPC & AI Forum: Business Innovation & PoC Demonstrations

HPC Montenegro – National Competence Centre for High-Performance Computing is pleased to announce the 3rd HPC & AI Forum on Business Innovation & PoC Demonstrators, taking place on 11–12 December 2026 at the University of Donja Gorica (UDG) in Podgorica.

Building on the previous editions, the Forum will bring together students, researchers, industry representatives, and other stakeholders to explore how High-Performance Computing (HPC) and Artificial Intelligence (AI) can support innovation, research, and real-world applications.

The event will highlight practical experiences, national and European project activities, and Proof-of-Concept (PoC) demonstrations, with particular emphasis on solutions developed by students and young researchers, as well as opportunities for academia–industry collaboration.

Organised within the activities of EuroCC 3, the Forum aims to strengthen Montenegro’s HPC/AI ecosystem and encourage further cooperation between academia and industry.

Industry representatives, students and other interested participants are invited to register their interest through the preliminary registration form: [Registration Form – Google Forms link]

Further information and the detailed programme will be announced soon.

MSc Thesis: Evaluating LLM-Generated Synthetic Mental Health Datasets

On 30th October 2026, MSc candidate Sara Đurović defended her thesis entitled “Evaluation of Synthetic Mental Health Datasets Generated by Large Language Models” within the Information Technologies Master’s programme (Software Engineering module) at the Faculty for Information Systems and Technologies, University of Donja Gorica, under the mentorship of Prof. Dr. Tomo Popović and co-mentorship of Dr. Stevan Čakić.

The research and experimental work were partially supported by NCC Montenegro through HPC infrastructure and expertise within the EuroCC3 project. Through its support for academic research and the development of AI and HPC competencies, NCC Montenegro contributes to advancing the application of advanced computing technologies in natural language processing and responsible AI development.

Research workflow and evaluation framework of the thesis (Source: Sara Đurović, MSc thesis, 2026).

The thesis investigates how the quality and research usability of synthetic text generated by large language models can be assessed, using mental health diary entries as a sensitive domain where access to real-world data is limited, and focusing on a multidimensional evaluation framework rather than a single quality metric.

ABSTRACT – This thesis presents a methodological evaluation of the quality and research usability of three synthetic mental health diary datasets produced through workflows based on GPT-5.5, Claude Sonnet 4.6 and Gemini 3.1 Pro. Each dataset contains 1,000 entries across 100 journals, sharing a common data schema and co-generated labels, with the journal treated as the unit of statistical independence. The evaluation combined cluster-aware permutation inference with Benjamini-Hochberg correction, a classifier two-sample test against a human reference corpus (SemEval-2026 Task 2), cross-dataset transfer experiments, a TF-IDF versus RoBERTa comparison, and zero-shot mmBERT transfer between paired English and Montenegrin text. The datasets differed significantly in seven of nine labels, and synthetic text remained clearly separable from human-written text even after controlling for text length (balanced accuracy 0.9855–0.9942). All 30 cross-dataset comparisons showed performance drops (26.04%–84.99%), while RoBERTa outperformed the TF-IDF baseline (macro-F1 0.7872 vs. 0.7118). The findings show that the datasets support controlled experiments but are not distributionally interchangeable, and that synthetic data quality requires multidimensional evaluation rather than a single metric. The results do not establish clinical validity of the labels or rank the generators by quality.

HPC Montenegro at CIGRE NGN 2026: Advancing AI and HPC Applications in the Energy Sector

Representatives of the National Competence Centre for High-Performance Computing (HPC Montenegro) participated in the CIGRE Next Generation Network (NGN) 2026 workshop for students and young engineers, held on November 4, 2026, at the Science and Technology Park of Montenegro in Podgorica. The event attracted 74 registrations, including representatives of major energy companies such as EPCG, CEDIS, and CGES, other industry stakeholders, and academia.

Prof. Tomo Popović (University of Donja Gorica) delivered a lecture titled “Artificial Intelligence in Energy – Opportunities and Limitations”, prepared jointly with Dr. Luka Filipović (UDG) and Prof. Božo Krstajić (University of Montenegro). The presentation explored practical applications of AI and machine learning in electrical power systems, including energy demand and generation forecasting, predictive maintenance, grid management, and energy system optimization.

Special attention was given to the role of High-Performance Computing (HPC) in supporting advanced AI models and large-scale simulations. Participants were introduced to the EuroCC3 project, HPC Montenegro’s activities and services, and opportunities for expert consultations, training, and access to European supercomputing infrastructure. The presentation also highlighted selected HPC and AI success stories relevant to the energy sector from Montenegro and across Europe, demonstrating the benefits of collaboration with National Competence Centres.

Following the lecture, Prof. Popović moderated the panel discussion “Opportunities for AI Applications in Electrical Power Systems”, featuring Mihailo Micev (UCG), Krsto Biskupović (CEDIS), Ivan Vujović (TERING), and Stevan Šandi (BIXBIT/INNO4GRID).

Participation in the workshop provided a valuable opportunity for HPC Montenegro to engage directly with energy sector stakeholders and young engineers, exchange experiences, and establish new industry contacts. Such activities strengthen academia–industry collaboration, raise awareness of EuroCC3 opportunities, and help identify potential areas for future cooperation in applying HPC and AI technologies to the energy sector [link].

Cyber Co awarded Fast Lane Access for the third time and Joins NVIDIA Inception Program for Startups

GetMoved, an AI platform for the moving industry developed by the Cyber Co Startup, has been awarded free access to GPU resources on the Leonardo supercomputer at CINECA in Italy — its third EuroHPC JU allocation through the Fast Lane access. The reason for changing from the LUMI to Leonardo supercomputer is in that Cyber CO has joined NVIDIA Inception, NVIDIA’s programme for technology startups and should now use NVIDIA graphical casrs, like the ones at Leonardo AI factory. Together the three awards have carried GetMoved’s computer vision work from the first experiments to production training runs.

GetMoved turns a short smartphone walkthrough of an apartment into a structured household inventory, a volume and weight estimate, and a moving cost estimate. The platform combines a customer marketplace, an ERP for moving companies, and the inventory engine in a single product. The detection models cover 78 object classes and are supported by a complete training pipeline — dataset preparation, annotation, training, evaluation, and continuous retraining.

Leonardo provides compute continuity as that pipeline moves toward production. NVIDIA Inception adds access to NVIDIA’s model catalogue and development tooling, along with compute credits; the company has opened an organisation on NVIDIA Build and NGC. On the technical roadmap, the current detector is being evaluated against a vision-language model, NVIDIA Cosmos, in phases with a decision gate before each replacement step.

The platform launched in the New York and New Jersey market in June 2026 and works with licensed moving companies on a non-exclusive basis.

HPC Montenegro and HashiMotto Explore HPC and AI Opportunities in Healthcare

The National Competence Centre for High-Performance Computing (HPC Montenegro) held an initial consultation meeting at the University of Donja Gorica (UDG) with HashiMotto, an innovative startup from Podgorica, to explore opportunities for applying high-performance computing (HPC) and artificial intelligence (AI) in healthcare. The meeting was attended by Omar Radončić, Director of HashiMotto, and HPC Montenegro representatives Tomo Popović and Luka Filipović.

Representatives of HashiMotto and HPC Montenegro during initial consultations at UDG

HashiMotto is developing an AI-powered application for individuals with Hashimoto’s thyroiditis, offering personalized support, health monitoring, and medical data analysis. Discussions focused on the potential development and training of predictive AI/ML models using large-scale medical datasets, particularly for applications in endocrinology and personalized treatment monitoring.

HPC Montenegro presented opportunities for accessing local HPC computing nodes, technical assistance with configuring computational environments and running AI/ML experiments, and support for accessing European supercomputing resources through the EuroHPC Benchmark Access and Development Access calls.

This consultation reflects HPC Montenegro’s continued commitment to supporting local startups and innovative companies in developing advanced AI solutions and adopting HPC technologies.

MSc Thesis: Computer Vision Models for Liver Tumor Segmentation

On October 21, 2026, MSc candidate Ida Mušović completed her thesis entitled “Development of Computer Vision-Based Predictive Models for Early Diagnosis of Liver Tumors” within the Artificial Intelligence Master’s programme at the Faculty for Information Systems and Technologies, University of Donja Gorica, under the mentorship of Prof. Dr. Tomo Popović.

The research and experimental work were partially supported by NCC Montenegro through HPC infrastructure and expertise within the EuroCC3 project. Through its support for academic research and the development of AI and HPC competencies, NCC Montenegro contributes to advancing the application of advanced computing technologies in healthcare.

Examples of CT images and anatomical segmentation masks generated using the TransUNet model (Source: Ida Mušović, MSc thesis, 2026).

The thesis investigates the application of deep learning and computer vision to automated liver tumor segmentation in CT images, focusing on the comparison of convolutional, hybrid, and transformer-based architectures for computer-aided medical image analysis.

ABSTRACT – This thesis presents the development and comparative evaluation of three deep learning architectures—U-Net, TransUNet, and SegFormer—for automated liver tumor segmentation in CT images. Using the publicly available Liver Tumor Segmentation (LiTS) dataset and a 2.5D image-processing approach, the models were evaluated using Dice similarity and Intersection over Union metrics, with additional analysis of performance across different lesion sizes. SegFormer achieved the highest overall Dice score (82.86%) for tumor segmentation, although differences between the architectures were not statistically significant at the patient level. The study also identified lesion size as an important factor influencing segmentation accuracy, with smaller tumors presenting greater challenges. The findings demonstrate the potential of computer vision to support medical image analysis while emphasizing the need for further validation before clinical application.