2026
Conference article
Open Access
Advancements in Artificial Intelligence and Computer Vision for Biomedical Applications @SI-Lab
Marco Cafiso, Andrea Carboni, Claudia Caudai, Sara Colantonio, Francesco Conti, Mario D’acunto, Said Daoudagh, Giulio Del Corso, Riccardo Del Gratta, Danila Germanese, Giacomo Ignesti, Gianmarco Lazzini, Giuseppe Riccardo Leone, Barbara Leporini, Massimo Magrini, Massimo Martinelli, Davide Moroni, Ali Reza Omrani, Francesca Pardini, Maria Antonietta Pascali, Paolo Paradisi, Laura Sebastiaini, Marco Tampucci, Valeria Tateo, Federico Volpini
This paper summarizes recent research at the Signals and Images Lab (ISTI-CNR) leveraging artificial intelligence, machine learning, and computer vision to address complex biomedical challenges. We highlight advancements in trustworthy, explainable algorithmic foundations alongside their practical clinical applications across diagnostic imaging, physiological signal analysis and neuromotor rehabilitation.
Marco Cafiso, Andrea Carboni, Claudia Caudai, Sara Colantonio, Francesco Conti, Mario D’acunto, Said Daoudagh, Giulio Del Corso, Riccardo Del Gratta, Danila Germanese, Giacomo Ignesti, Gianmarco Lazzini, Giuseppe Riccardo Leone, Barbara Leporini, Massimo Magrini, Massimo Martinelli, Davide Moroni, Ali Reza Omrani, Francesca Pardini, Maria Antonietta Pascali, Paolo Paradisi, Laura Sebastiaini, Marco Tampucci, Valeria Tateo, Federico Volpini
This paper summarizes recent research at the Signals and Images Lab (ISTI-CNR) leveraging artificial intelligence, machine learning, and computer vision to address complex biomedical challenges. We highlight advancements in trustworthy, explainable algorithmic foundations alongside their practical clinical applications across diagnostic imaging, physiological signal analysis and neuromotor rehabilitation.
2026
Contribution to book
Open Access
A survey on Good AI: user-centric AI design in healthcare
Berti Andrea, Giannini Valentina, Mazzetti Simone, Pascali Maria Antonietta, Regge Daniele, Colantonio Sara
The integration of Artificial Intelligence (AI) in healthcare has the potential to revolutionize patient care by enhancing diagnostic processes, treatment protocols, and overall healthcare delivery. However, the adoption of AI-powered tools and services is contingent upon establishing a robust foundation of trust among healthcare professionals. The ProCAncer-I project, informed by the FUTURE-AI framework, is at the forefront of this effort, promoting a user-centric design philosophy that prioritizes the needs and expectations of end-users, primarily clinicians and radiologists. This paper delves into the co-design methodology adopted by an interdisciplinary team, elucidating the collaborative efforts that underpin the customization of the FUTURE-AI principles to align with the clinical requirements of the project’s partners. The introduction sets the stage for a comprehensive discussion on the significance of stakeholder engagement in the design and implementation of trustworthy AI systems within clinical settings.Source: SMART INNOVATION, SYSTEMS AND TECHNOLOGIES, vol. 460, pp. 113-128
DOI: 10.1007/978-981-95-4076-1_10
DOI: 10.5281/zenodo.13918943
DOI: 10.5281/zenodo.13918942
Project(s): An AI Platform integrating imaging data and models, supporting precision care through prostate cancer’s continuum, NAVIGATOR
Metrics:
Berti Andrea, Giannini Valentina, Mazzetti Simone, Pascali Maria Antonietta, Regge Daniele, Colantonio Sara
The integration of Artificial Intelligence (AI) in healthcare has the potential to revolutionize patient care by enhancing diagnostic processes, treatment protocols, and overall healthcare delivery. However, the adoption of AI-powered tools and services is contingent upon establishing a robust foundation of trust among healthcare professionals. The ProCAncer-I project, informed by the FUTURE-AI framework, is at the forefront of this effort, promoting a user-centric design philosophy that prioritizes the needs and expectations of end-users, primarily clinicians and radiologists. This paper delves into the co-design methodology adopted by an interdisciplinary team, elucidating the collaborative efforts that underpin the customization of the FUTURE-AI principles to align with the clinical requirements of the project’s partners. The introduction sets the stage for a comprehensive discussion on the significance of stakeholder engagement in the design and implementation of trustworthy AI systems within clinical settings.Source: SMART INNOVATION, SYSTEMS AND TECHNOLOGIES, vol. 460, pp. 113-128
DOI: 10.1007/978-981-95-4076-1_10
DOI: 10.5281/zenodo.13918943
DOI: 10.5281/zenodo.13918942
Project(s): An AI Platform integrating imaging data and models, supporting precision care through prostate cancer’s continuum, NAVIGATOR
Metrics:
See at:
CNR IRIS 
| link.springer.com | ZENODO | doi.org 
| ZENODO | ZENODO | CNR IRIS
2026
Conference article
Open Access
Remind me of something? Zero-Shot learning for trustworthy image comparison in rolling stock
Papini Oscar, Del Corso Giulio, Bulotta Davide, Carboni Andrea, Gravili Silvia, Leone Giuseppe Riccardo, Pascali Maria Antonietta, Moroni Davide, Colantonio Sara
This paper discusses the need for trustworthy AI in urban mobility, focusing on high-stakes security applications such as anomaly detection in public transportation. Because the accuracy required to identify potentially dangerous objects often surpasses the capabilities of current models, there is an unavoidable incidence of false positives. We suggest a "learning to defer" approach as a solution. Our technique uses the deep features and label relative importance of a pre-trained classifier (DenseNet/ImageNET-1k) to create a unique item "fingerprint". We then employ a zero-shot meta-learning approach to calibrate the system, enabling it to distinguish between normal background items and genuine anomalies by assigning a similarity score. This method significantly reduces the false "new object" alarms that would otherwise overwhelm human operators. Our proof-of-concept demonstrates that the system is computationally light and can be easily adapted to specific environments and integrated into existing classification modules.Source: LECTURE NOTES IN COMPUTER SCIENCE, vol. 16170, pp. 323-334. Roma, Italy, 15-19/09/2025
DOI: 10.1007/978-3-032-11381-8_28
Project(s): FAITH
Metrics:
Papini Oscar, Del Corso Giulio, Bulotta Davide, Carboni Andrea, Gravili Silvia, Leone Giuseppe Riccardo, Pascali Maria Antonietta, Moroni Davide, Colantonio Sara
This paper discusses the need for trustworthy AI in urban mobility, focusing on high-stakes security applications such as anomaly detection in public transportation. Because the accuracy required to identify potentially dangerous objects often surpasses the capabilities of current models, there is an unavoidable incidence of false positives. We suggest a "learning to defer" approach as a solution. Our technique uses the deep features and label relative importance of a pre-trained classifier (DenseNet/ImageNET-1k) to create a unique item "fingerprint". We then employ a zero-shot meta-learning approach to calibrate the system, enabling it to distinguish between normal background items and genuine anomalies by assigning a similarity score. This method significantly reduces the false "new object" alarms that would otherwise overwhelm human operators. Our proof-of-concept demonstrates that the system is computationally light and can be easily adapted to specific environments and integrated into existing classification modules.Source: LECTURE NOTES IN COMPUTER SCIENCE, vol. 16170, pp. 323-334. Roma, Italy, 15-19/09/2025
DOI: 10.1007/978-3-032-11381-8_28
Project(s): FAITH
Metrics:
See at:
CNR IRIS | link.springer.com | CNR IRIS | CNR IRIS
2026
Contribution to journal
Open Access
Multiomics approach for patient stratification and novel target identification in metastatic clear cell renal carcinoma (MeetUro 31): preliminary analysis of radiomics features—A Meet-URO and AIRC study (NCT05782400)
Procopio Giuseppe, Stellato Marco, Barella Marco, Claps Melanie, Guadalupi Valentina, Rametta Alessandro, Basso Umberto, Buti Sebastiano, Vignani Francesca, Fratino Lucia, Nole Franco, Di Napoli Marilena, Zucali Paolo Andrea, Verzonie Lena, Germanese Danila, Di Maio Massimo, Del Re Marzia, De Cecco Loris, Colantonio Sara, Romei Chiara
Artificial Intelligence can integrate clinic-pathological features, radiomics, genomic and transcriptomic analysis to define an optimal allocation strategy in first line treatment of metastatic renal cell carcinoma (mRCC). Methods: This is a multicenter Italian prospective translational study including patients (pts) with clear cell mRCC receiving first-line treatment as per investigator’s choice. Tumor tissue was collected at baseline, plasma samples and CT scan were collected at baseline and every 3 months until progression. Due to the short follow up, here we report the preliminary analysis of the radiomic features to identify signatures associated with Objective Response Rate (ORR). A subset of non-analytically correlated radiomic features was extracted from the selected regions of interest. This subset included first-order statistics, three-dimensional shape descriptors, and texture-based features. All features were computed on the original images using PyRadiomics v.3.1.0. The radiomic analysis pipeline consisted of feature variance filtering, multicollinearity reduction, data harmonization and standardization, and feature importance estimation through a Random Forest-based algorithm. Results: 100 pts were enrolled. For the radiomic analysis, 68 patients were included to ensure a more reliable data harmonization process and to improve the robustness of subsequent analyses. 18 (26%) received IO-IO, 38 (56%) received IO-TKI, 12 (18) received TKI monotherapy as first line treatment. According to IMDC score, 16(24%) were good risk, 39(57%) intermediate and 13(19%) poor. The most common site of metastasis were lung (55%, 38), bone (23%, 16), nodes (20%, 14/68) and liver (13%, 9). In the overall population, ORR was 48% (33), 44% (18) in the IO-TKI group, 44% (8) in the IO-IO group and 42% (5) in the TKI group. The two most influential features identified by the Random Forest model were original_firstorder_Mean and original_glcm_Contrast (0.59 accuracy, 0.58 precision, 0.58 recall, 0.58 F1 score, 0.49 AUROC). Higher values of these features—reflecting increased tissue density and heterogeneity—were associated with a higher ORR. Conclusions: This preliminary analysis suggests that 2 radiomic signatures are associated with higher ORR and are promising as early biomarkers of response in mRCC. However, they do not appear to provide optimal predictive value when used alone, and should therefore be integrated with clinical, genomic, and transcriptomic data to refine predictive modeling. Acknowledgments: We thank AIRC (Associazione Italiana Ricerca sul Cancro) for the support received to conduct this trial. Clinical trial information: NCT05782400.Source: JOURNAL OF CLINICAL ONCOLOGY, vol. 44 (issue 7_suppl), p. 425
DOI: 10.1200/jco.2026.44.7_suppl.425
Metrics:
Procopio Giuseppe, Stellato Marco, Barella Marco, Claps Melanie, Guadalupi Valentina, Rametta Alessandro, Basso Umberto, Buti Sebastiano, Vignani Francesca, Fratino Lucia, Nole Franco, Di Napoli Marilena, Zucali Paolo Andrea, Verzonie Lena, Germanese Danila, Di Maio Massimo, Del Re Marzia, De Cecco Loris, Colantonio Sara, Romei Chiara
Artificial Intelligence can integrate clinic-pathological features, radiomics, genomic and transcriptomic analysis to define an optimal allocation strategy in first line treatment of metastatic renal cell carcinoma (mRCC). Methods: This is a multicenter Italian prospective translational study including patients (pts) with clear cell mRCC receiving first-line treatment as per investigator’s choice. Tumor tissue was collected at baseline, plasma samples and CT scan were collected at baseline and every 3 months until progression. Due to the short follow up, here we report the preliminary analysis of the radiomic features to identify signatures associated with Objective Response Rate (ORR). A subset of non-analytically correlated radiomic features was extracted from the selected regions of interest. This subset included first-order statistics, three-dimensional shape descriptors, and texture-based features. All features were computed on the original images using PyRadiomics v.3.1.0. The radiomic analysis pipeline consisted of feature variance filtering, multicollinearity reduction, data harmonization and standardization, and feature importance estimation through a Random Forest-based algorithm. Results: 100 pts were enrolled. For the radiomic analysis, 68 patients were included to ensure a more reliable data harmonization process and to improve the robustness of subsequent analyses. 18 (26%) received IO-IO, 38 (56%) received IO-TKI, 12 (18) received TKI monotherapy as first line treatment. According to IMDC score, 16(24%) were good risk, 39(57%) intermediate and 13(19%) poor. The most common site of metastasis were lung (55%, 38), bone (23%, 16), nodes (20%, 14/68) and liver (13%, 9). In the overall population, ORR was 48% (33), 44% (18) in the IO-TKI group, 44% (8) in the IO-IO group and 42% (5) in the TKI group. The two most influential features identified by the Random Forest model were original_firstorder_Mean and original_glcm_Contrast (0.59 accuracy, 0.58 precision, 0.58 recall, 0.58 F1 score, 0.49 AUROC). Higher values of these features—reflecting increased tissue density and heterogeneity—were associated with a higher ORR. Conclusions: This preliminary analysis suggests that 2 radiomic signatures are associated with higher ORR and are promising as early biomarkers of response in mRCC. However, they do not appear to provide optimal predictive value when used alone, and should therefore be integrated with clinical, genomic, and transcriptomic data to refine predictive modeling. Acknowledgments: We thank AIRC (Associazione Italiana Ricerca sul Cancro) for the support received to conduct this trial. Clinical trial information: NCT05782400.Source: JOURNAL OF CLINICAL ONCOLOGY, vol. 44 (issue 7_suppl), p. 425
DOI: 10.1200/jco.2026.44.7_suppl.425
Metrics:
See at:
CNR IRIS | www.scilove.app | Journal of Clinical Oncology | CNR IRIS
2026
Journal article
Open Access
Integrating multimodal learning and explainable AI for enhanced and interpretable prostate lesion classification
Giovannoni Claudio, Metta Carlo, Berti Andrea, Colantonio Sara, Monreale Anna, Pratesi Francesca, Rinzivillo Salvatore
Artificial Intelligence systems could find many important applications in the medical field, holding excellent potential for improving disease diagnosis, treatment identification and selection. These opportunities are often jeopardized by the lack of interpretability of such systems, slowing down AI adoption. To overcome the issue, we first introduce an analytical framework exploiting multimodal deep learning for the classification of prostate lesions using Magnetic Resonance Imaging (MRI) data and clinical information on the patients. Then, we propose a multimodal explainability approach based on visual explanations to interpret the proposed model decision-making process and identify how the different modalities contribute to each specific prediction. Our findings, based on the PI-CAI Grand Challenge dataset, demonstrate the potential of combining multimodal data with eXplainable AI (XAI) to enhance prostate cancer diagnosis, improving model predictive performance, interpretability and understanding in treatment decision-making.Source: MACHINE LEARNING, vol. 115 (issue 4)
DOI: 10.1007/s10994-026-07033-x
Project(s): Critical Action Planning over Extreme-Scale Data, Foundations of Trustworthy AI - Integrating Reasoning, Learning and Optimization, Future Artificial Intelligence Research, It takes two to tango: a synergistic approach to human-machine decision making, Science and technology for the explanation of AI decision making, SoBigData++: European Integrated Infrastructure for Social Mining and Big Data Analytics, Strengthening the Italian RI for Social Mining and Big Data Analytics
Metrics:
Giovannoni Claudio, Metta Carlo, Berti Andrea, Colantonio Sara, Monreale Anna, Pratesi Francesca, Rinzivillo Salvatore
Artificial Intelligence systems could find many important applications in the medical field, holding excellent potential for improving disease diagnosis, treatment identification and selection. These opportunities are often jeopardized by the lack of interpretability of such systems, slowing down AI adoption. To overcome the issue, we first introduce an analytical framework exploiting multimodal deep learning for the classification of prostate lesions using Magnetic Resonance Imaging (MRI) data and clinical information on the patients. Then, we propose a multimodal explainability approach based on visual explanations to interpret the proposed model decision-making process and identify how the different modalities contribute to each specific prediction. Our findings, based on the PI-CAI Grand Challenge dataset, demonstrate the potential of combining multimodal data with eXplainable AI (XAI) to enhance prostate cancer diagnosis, improving model predictive performance, interpretability and understanding in treatment decision-making.Source: MACHINE LEARNING, vol. 115 (issue 4)
DOI: 10.1007/s10994-026-07033-x
Project(s): Critical Action Planning over Extreme-Scale Data, Foundations of Trustworthy AI - Integrating Reasoning, Learning and Optimization, Future Artificial Intelligence Research, It takes two to tango: a synergistic approach to human-machine decision making, Science and technology for the explanation of AI decision making, SoBigData++: European Integrated Infrastructure for Social Mining and Big Data Analytics, Strengthening the Italian RI for Social Mining and Big Data Analytics
Metrics:
See at:
CNR IRIS | link.springer.com | CNR IRIS | CNR IRIS
2025
Journal article
Open Access
The role of causality in explainable artificial intelligence
Carloni G., Berti A., Colantonio S.
Causality and eXplainable Artificial Intelligence (XAI) have developed as separate fields in computer science, even though the underlying concepts of causation and explanation share common ancient roots. This is further enforced by the lack of review works jointly covering these two fields. In this paper, we investigate the literature to try to understand how and to what extent causality and XAI are intertwined. More precisely, we seek to uncover what kinds of relationships exist between the two concepts and how one can benefit from them, for instance, in building trust in AI systems. As a result, three main perspectives are identified. In the first one, the lack of causality is seen as one of the major limitations of current AI and XAI approaches, and the "optimal" form of explanations is investigated. The second is a pragmatic perspective and considers XAI as a tool to foster scientific exploration for causal inquiry, via the identification of pursue-worthy experimental manipulations. Finally, the third perspective supports the idea that causality is propaedeutic to XAI in three possible manners: exploiting concepts borrowed from causality to support or improve XAI, utilizing counterfactuals for explainability, and considering accessing a causal model as explaining itself. To complement our analysis, we also provide relevant software solutions used to automate causal tasks. We believe our work provides a unified view of the two fields of causality and XAI by highlighting potential domain bridges and uncovering possible limitations.Source: WILEY INTERDISCIPLINARY REVIEWS. DATA MINING AND KNOWLEDGE DISCOVERY
DOI: 10.1002/widm.70015
Project(s): ProCAncer-I
Metrics:
Carloni G., Berti A., Colantonio S.
Causality and eXplainable Artificial Intelligence (XAI) have developed as separate fields in computer science, even though the underlying concepts of causation and explanation share common ancient roots. This is further enforced by the lack of review works jointly covering these two fields. In this paper, we investigate the literature to try to understand how and to what extent causality and XAI are intertwined. More precisely, we seek to uncover what kinds of relationships exist between the two concepts and how one can benefit from them, for instance, in building trust in AI systems. As a result, three main perspectives are identified. In the first one, the lack of causality is seen as one of the major limitations of current AI and XAI approaches, and the "optimal" form of explanations is investigated. The second is a pragmatic perspective and considers XAI as a tool to foster scientific exploration for causal inquiry, via the identification of pursue-worthy experimental manipulations. Finally, the third perspective supports the idea that causality is propaedeutic to XAI in three possible manners: exploiting concepts borrowed from causality to support or improve XAI, utilizing counterfactuals for explainability, and considering accessing a causal model as explaining itself. To complement our analysis, we also provide relevant software solutions used to automate causal tasks. We believe our work provides a unified view of the two fields of causality and XAI by highlighting potential domain bridges and uncovering possible limitations.Source: WILEY INTERDISCIPLINARY REVIEWS. DATA MINING AND KNOWLEDGE DISCOVERY
DOI: 10.1002/widm.70015
Project(s): ProCAncer-I
Metrics:
See at:
CNR IRIS | wires.onlinelibrary.wiley.com | CNR IRIS | CNR IRIS
2025
Other
Open Access
SI-Lab Annual Research Report 2024
Awais Ch Muhammad, Baiamonte A., Benassi A., Berti A., Bertini G., Buongiorno R, Bulotta D., Cafiso M., Carboni A., Carloni G., Caudai C., Colantonio S., Conti F., Daoudagh S., Del Corso G., Fusco G., Galesi G., Germanese D., Gravili S., Ignesti G., Kuruoglu E. E., Lazzini G., Leone G. R., Leporini B., Magrini M., Martinelli M., Omrani Ali Reza, Pachetti E., Papini O., Paradisi P., Pardini F., Pascali M. A., Pieri G., Reggiannini M., Righi M., Salerno E., Salvetti O., Scozzari A., Sebastiani L., Straface S., Tampucci M., Tarabella L., Tonazzini A., Moroni D.
The Signal & Images Laboratory (SI-Lab) is an interdisciplinary research group in computer vision, signal analysis, intelligent vision systems and multimedia data understanding. It is part of the Institute of Information Science and Technologies (ISTI) of the National Research Council of Italy (CNR). This report accounts for the research activities of the Signal and Images Laboratory of the Institute of Information Science and Technologies during the year 2024.DOI: 10.32079/isti-ar-2025/002
Metrics:
Awais Ch Muhammad, Baiamonte A., Benassi A., Berti A., Bertini G., Buongiorno R, Bulotta D., Cafiso M., Carboni A., Carloni G., Caudai C., Colantonio S., Conti F., Daoudagh S., Del Corso G., Fusco G., Galesi G., Germanese D., Gravili S., Ignesti G., Kuruoglu E. E., Lazzini G., Leone G. R., Leporini B., Magrini M., Martinelli M., Omrani Ali Reza, Pachetti E., Papini O., Paradisi P., Pardini F., Pascali M. A., Pieri G., Reggiannini M., Righi M., Salerno E., Salvetti O., Scozzari A., Sebastiani L., Straface S., Tampucci M., Tarabella L., Tonazzini A., Moroni D.
The Signal & Images Laboratory (SI-Lab) is an interdisciplinary research group in computer vision, signal analysis, intelligent vision systems and multimedia data understanding. It is part of the Institute of Information Science and Technologies (ISTI) of the National Research Council of Italy (CNR). This report accounts for the research activities of the Signal and Images Laboratory of the Institute of Information Science and Technologies during the year 2024.DOI: 10.32079/isti-ar-2025/002
Metrics:
2025
Journal article
Open Access
AI Model Passport: data and system traceability framework for transparent AI in health
Kalokyri V., Tachos N. S., Kalantzopoulos C. N., Sfakianakis S., Kondylakis H., Zaridis D. I., Colantonio S., Regge D., Papanikolaou N., Marias K., Fotiadis D. I., Tsiknakis M.
The increasing integration of Artificial Intelligence (AI) into health and biomedical systems necessitates robust frameworks for transparency, accountability, and ethical compliance. Existing frameworks often rely on human-readable, manual documentation which limits scalability, comparability, and machine interpretability across projects and platforms. They also fail to provide a unique, verifiable identity for AI models to ensure their provenance and authenticity across systems and use cases, limiting reproducibility and stakeholder trust. This paper introduces the concept of the AI Model Passport, a structured and standardized documentation framework that acts as a digital identity and verification tool for AI models. It captures essential metadata to uniquely identify, verify, trace and monitor AI models across their lifecycle - from data acquisition and preprocessing to model design, development and deployment. In addition, an implementation of this framework is presented through AIPassport, an MLOps tool developed within the ProCAncer-I EU project for medical imaging applications. AIPassport automates metadata collection, ensures proper versioning, decouples results from source scripts, and integrates with various development environments. Its effectiveness is showcased through a lesion segmentation use case using data from the ProCAncer-I dataset, illustrating how the AI Model Passport enhances transparency, reproducibility, and regulatory readiness while reducing manual effort. This approach aims to set a new standard for fostering trust and accountability in AI-driven healthcare solutions, aspiring to serve as the basis for developing transparent and regulation compliant AI systems across domains.Source: COMPUTATIONAL AND STRUCTURAL BIOTECHNOLOGY JOURNAL, vol. 28, pp. 386-404
DOI: 10.1016/j.csbj.2025.09.041
Project(s): ProCAncer-I
Metrics:
Kalokyri V., Tachos N. S., Kalantzopoulos C. N., Sfakianakis S., Kondylakis H., Zaridis D. I., Colantonio S., Regge D., Papanikolaou N., Marias K., Fotiadis D. I., Tsiknakis M.
The increasing integration of Artificial Intelligence (AI) into health and biomedical systems necessitates robust frameworks for transparency, accountability, and ethical compliance. Existing frameworks often rely on human-readable, manual documentation which limits scalability, comparability, and machine interpretability across projects and platforms. They also fail to provide a unique, verifiable identity for AI models to ensure their provenance and authenticity across systems and use cases, limiting reproducibility and stakeholder trust. This paper introduces the concept of the AI Model Passport, a structured and standardized documentation framework that acts as a digital identity and verification tool for AI models. It captures essential metadata to uniquely identify, verify, trace and monitor AI models across their lifecycle - from data acquisition and preprocessing to model design, development and deployment. In addition, an implementation of this framework is presented through AIPassport, an MLOps tool developed within the ProCAncer-I EU project for medical imaging applications. AIPassport automates metadata collection, ensures proper versioning, decouples results from source scripts, and integrates with various development environments. Its effectiveness is showcased through a lesion segmentation use case using data from the ProCAncer-I dataset, illustrating how the AI Model Passport enhances transparency, reproducibility, and regulatory readiness while reducing manual effort. This approach aims to set a new standard for fostering trust and accountability in AI-driven healthcare solutions, aspiring to serve as the basis for developing transparent and regulation compliant AI systems across domains.Source: COMPUTATIONAL AND STRUCTURAL BIOTECHNOLOGY JOURNAL, vol. 28, pp. 386-404
DOI: 10.1016/j.csbj.2025.09.041
Project(s): ProCAncer-I
Metrics:
See at:
CNR IRIS | www.sciencedirect.com | CNR IRIS
2025
Journal article
Restricted
NAVIGATOR: a regional multimodal imaging biobank initiative powered by AI tools for precision medicine in oncology
Aghakhanyan G., Barucci A., Pascali M. A., Assante M., Bagnacci G., Bertelli E., Caputo F. P., Cuibari M. E., Carlini E., Carpi R., Caudai C., Cioni D., Colantonio S., Colcelli V., Dell'Amico A., Vecchio V. D., Gangi D. D., Faggioni L., Formica V., Francischello R., Frosini L., Kotsa C., Lipari G., Manghi P., Martino V. D., Marzi C., Mazzei M. A., Mangiacrapa F., Meglio N. D., Miele V., Molinaro E., Paiar F., Pagano P., Panichi G., Pasquinelli F., Peccerillo B., Perrella A., Piccioli T., Oliviero A., Olivoni M., Rucci D., Tampucci M., Tumminello L., Volpini F., Zanuzzi A., Fanni S. C., Neri E.
The NAVIGATOR project established an Italian regional imaging biobank and interactive research platform designed to support precision oncology through the integration of multimodal imaging, clinical, and omics data. The platform goes beyond a static repository, offering a secure Virtual Research Environment (VRE) where users can upload data, test AI algorithms, and execute complete analytical pipelines. The platform incorporates artificial intelligence (AI)-driven radiomics and deep learning methodologies to enable biomarker extraction, disease stratification, and predictive modeling. This manuscript presents the development and implementation of the NAVIGATOR infrastructure, including its data governance framework, ethical and legal considerations, and application to three oncological use cases: prostate, rectal, and gastric cancers. To date, the biobank has collected imaging and clinical data from over 700 patients across these cohorts. AI models were deployed within a dedicated VRE to facilitate image analysis, feature extraction, and classification tasks. The project addresses critical challenges related to data harmonization, regulatory compliance, privacy safeguards and fairness in AI systems. NAVIGATOR demonstrates the feasibility of integrating AI methodologies within imaging biobanks and provides a scalable framework to advance oncological research and support clinical decision-making.Source: EUROPEAN JOURNAL OF RADIOLOGY, vol. 191 (issue 112327)
DOI: 10.1016/j.ejrad.2025.112327
Project(s): An Imaging Biobank to Precisely Prevent and Predict cancer, and facilitate the Participation of oncologic patients to Diagnosis and Treatment
Metrics:
Aghakhanyan G., Barucci A., Pascali M. A., Assante M., Bagnacci G., Bertelli E., Caputo F. P., Cuibari M. E., Carlini E., Carpi R., Caudai C., Cioni D., Colantonio S., Colcelli V., Dell'Amico A., Vecchio V. D., Gangi D. D., Faggioni L., Formica V., Francischello R., Frosini L., Kotsa C., Lipari G., Manghi P., Martino V. D., Marzi C., Mazzei M. A., Mangiacrapa F., Meglio N. D., Miele V., Molinaro E., Paiar F., Pagano P., Panichi G., Pasquinelli F., Peccerillo B., Perrella A., Piccioli T., Oliviero A., Olivoni M., Rucci D., Tampucci M., Tumminello L., Volpini F., Zanuzzi A., Fanni S. C., Neri E.
The NAVIGATOR project established an Italian regional imaging biobank and interactive research platform designed to support precision oncology through the integration of multimodal imaging, clinical, and omics data. The platform goes beyond a static repository, offering a secure Virtual Research Environment (VRE) where users can upload data, test AI algorithms, and execute complete analytical pipelines. The platform incorporates artificial intelligence (AI)-driven radiomics and deep learning methodologies to enable biomarker extraction, disease stratification, and predictive modeling. This manuscript presents the development and implementation of the NAVIGATOR infrastructure, including its data governance framework, ethical and legal considerations, and application to three oncological use cases: prostate, rectal, and gastric cancers. To date, the biobank has collected imaging and clinical data from over 700 patients across these cohorts. AI models were deployed within a dedicated VRE to facilitate image analysis, feature extraction, and classification tasks. The project addresses critical challenges related to data harmonization, regulatory compliance, privacy safeguards and fairness in AI systems. NAVIGATOR demonstrates the feasibility of integrating AI methodologies within imaging biobanks and provides a scalable framework to advance oncological research and support clinical decision-making.Source: EUROPEAN JOURNAL OF RADIOLOGY, vol. 191 (issue 112327)
DOI: 10.1016/j.ejrad.2025.112327
Project(s): An Imaging Biobank to Precisely Prevent and Predict cancer, and facilitate the Participation of oncologic patients to Diagnosis and Treatment
Metrics:
See at:
European Journal of Radiology | Archivio della Ricerca - Università di Pisa | CNR IRIS | CNR IRIS | Archivio della Ricerca - Università di Pisa
2025
Contribution to book
Open Access
AI models in cancer diagnosis and prognosis
Filos D., Chouvarda I., Sykiotis S., Tzortzis I., Rallis I., Doulamis A., Doulamis N., De Almeida J. G., Rodrigues N., Rodrigues A. C., Papanikolaou N., Loncar-Turukalo T., Jakovljevic N., Lazic I., Rapaic M., Hautaniemi S., Caudai C., Del Corso G., Germanese D., Pachetti E., Pascali M. A., Colantonio S.
The increasing volume of collected cancer imaging data, together with the development of technological tools based on Artificial Intelligence (AI), offers unprecedented opportunities for enhancing cancer management and improving clinical workflows. This chapter explores the approaches adopted by two projects within the AI for Health Imaging (AI4HI) cluster, INCISIVE and ProCancer-I, targeting various cancer types. In total, sixteen models were implemented across two projects, focusing on prostate, breast, lung, and ovarian cancers. These models were designed for lesion segmentation, patient stratification, and predicting metastasis risk or radiotherapy side effects, utilizing various DL and ML architectures such as YOLO, ResUnet++, and U-Net. Diverse imaging modalities were used, including Magnetic Resonance Imaging, Computed Tomography, and Mammography, while whole-slide images were used for the detection and classification of cell types in histopathological images. Radiomics was employed for classification and prediction by extracting features from imaging data, with harmonization techniques applied to improve model generalizability. Although some models incorporated clinical data, most relied on imaging features, highlighting the potential for improved performance by integrating multimodal data. To further enhance model performance and generalizability, comprehensive repositories with detailed clinical and follow-up data are needed. Additionally, addressing model fairness, explainability, and biological validation is essential for gaining acceptance within the clinical community.DOI: 10.1007/978-3-031-89963-8_7
Project(s): An AI Platform integrating imaging data and models, supporting precision care through prostate cancer’s continuum
Metrics:
Filos D., Chouvarda I., Sykiotis S., Tzortzis I., Rallis I., Doulamis A., Doulamis N., De Almeida J. G., Rodrigues N., Rodrigues A. C., Papanikolaou N., Loncar-Turukalo T., Jakovljevic N., Lazic I., Rapaic M., Hautaniemi S., Caudai C., Del Corso G., Germanese D., Pachetti E., Pascali M. A., Colantonio S.
The increasing volume of collected cancer imaging data, together with the development of technological tools based on Artificial Intelligence (AI), offers unprecedented opportunities for enhancing cancer management and improving clinical workflows. This chapter explores the approaches adopted by two projects within the AI for Health Imaging (AI4HI) cluster, INCISIVE and ProCancer-I, targeting various cancer types. In total, sixteen models were implemented across two projects, focusing on prostate, breast, lung, and ovarian cancers. These models were designed for lesion segmentation, patient stratification, and predicting metastasis risk or radiotherapy side effects, utilizing various DL and ML architectures such as YOLO, ResUnet++, and U-Net. Diverse imaging modalities were used, including Magnetic Resonance Imaging, Computed Tomography, and Mammography, while whole-slide images were used for the detection and classification of cell types in histopathological images. Radiomics was employed for classification and prediction by extracting features from imaging data, with harmonization techniques applied to improve model generalizability. Although some models incorporated clinical data, most relied on imaging features, highlighting the potential for improved performance by integrating multimodal data. To further enhance model performance and generalizability, comprehensive repositories with detailed clinical and follow-up data are needed. Additionally, addressing model fairness, explainability, and biological validation is essential for gaining acceptance within the clinical community.DOI: 10.1007/978-3-031-89963-8_7
Project(s): An AI Platform integrating imaging data and models, supporting precision care through prostate cancer’s continuum
Metrics:
See at:
CNR IRIS | link.springer.com | doi.org | CNR IRIS | CNR IRIS
2025
Journal article
Open Access
Effective reduction of unnecessary biopsies through a deep-learning-assisted aggressive prostate cancer detector
Gaivão A. M., Bireiro C., Santiago I., Joana Ip, Belião S., Matos C., Vanneschi L., Tsiknakis M., Marias K., Regge D., Silva S., Sfakianakis S., Kalokyri V., Trivizakis E., Kalliatakis G., Dimitriadis A., Fotiadis D., Tachos N., Mylona E., Zaridis D., Kalantzopoulos C., Papanikolaou N., De Almeida J. G., Castro Verde A., Rodrigues A. C., Rodrigues N., Chambel M., Huisman H., De Rooij M., Saha A., Twilt J. J., Futterer J., Martí-Bonmatí L., Cerdá-Alberich L., Ribas G., Navarro S., Marfil M., Neri E., Aringhieri G., Tumminello L., Mendola M., Akata V., Özmen M., Karaosmanoglu A. D., Atak F., Karcaaltincaba M., Vilanova J. C., Usinskiene J., Briediene R., Untanas A., Slidevska K., Vasilis K., Georgios G., Koh D. -M., Emsley R., Vit S., Ribeiro A., Doran S., Jacobs T., García-Martí G., Giannini V., Mazzetti S., Cappello G., Maimone G., Napolitano V., Colantonio S., Pascali M. A., Pachetti E., Del Corso G., Germanese D., Berti A., Carloni G., Kalpathy-Cramer J., Bridge C., Correia J., Hernandez W., Giavri Z., Pollalis C., Agraniotis D., Jiménez Pastor A., Munuera Mora J., Saillant C., Henne T., Marquez R.
Despite being one of the most prevalent cancers, prostate cancer (PCa) shows a significantly high survival rate, provided there is timely detection and treatment. Currently, several screening and diagnostic tests are required to be carried out in order to detect PCa. These tests are often invasive, requiring either a biopsy (Gleason score and ISUP) or blood tests (PSA). Computational methods have been shown to help this process, using multiparametric MRI (mpMRI) data to detect PCa, effectively providing value during the diagnosis and monitoring stages. While delineating lesions requires a high degree of experience and expertise from the radiologists, being subject to a high degree of inter-observer variability, often leading to inconsistent readings, these computational models can leverage the information from mpMRI to locate the lesions with a high degree of certainty. By considering as positive samples only those that have an ISUP2 we can train aggressive index lesion detection models. The main advantage of this approach is that, by focusing only on aggressive disease, the output of such a model can also be seen as an indication for biopsy, effectively reducing unnecessary biopsy screenings. In this work, we utilize both the highly heterogeneous ProstateNet dataset, and the PI-CAI dataset, to develop accurate aggressive disease detection models.Source: SCIENTIFIC REPORTS, vol. 15 (issue 1)
DOI: 10.1038/s41598-025-99795-y
Project(s): ProCAncer-I
Metrics:
Gaivão A. M., Bireiro C., Santiago I., Joana Ip, Belião S., Matos C., Vanneschi L., Tsiknakis M., Marias K., Regge D., Silva S., Sfakianakis S., Kalokyri V., Trivizakis E., Kalliatakis G., Dimitriadis A., Fotiadis D., Tachos N., Mylona E., Zaridis D., Kalantzopoulos C., Papanikolaou N., De Almeida J. G., Castro Verde A., Rodrigues A. C., Rodrigues N., Chambel M., Huisman H., De Rooij M., Saha A., Twilt J. J., Futterer J., Martí-Bonmatí L., Cerdá-Alberich L., Ribas G., Navarro S., Marfil M., Neri E., Aringhieri G., Tumminello L., Mendola M., Akata V., Özmen M., Karaosmanoglu A. D., Atak F., Karcaaltincaba M., Vilanova J. C., Usinskiene J., Briediene R., Untanas A., Slidevska K., Vasilis K., Georgios G., Koh D. -M., Emsley R., Vit S., Ribeiro A., Doran S., Jacobs T., García-Martí G., Giannini V., Mazzetti S., Cappello G., Maimone G., Napolitano V., Colantonio S., Pascali M. A., Pachetti E., Del Corso G., Germanese D., Berti A., Carloni G., Kalpathy-Cramer J., Bridge C., Correia J., Hernandez W., Giavri Z., Pollalis C., Agraniotis D., Jiménez Pastor A., Munuera Mora J., Saillant C., Henne T., Marquez R.
Despite being one of the most prevalent cancers, prostate cancer (PCa) shows a significantly high survival rate, provided there is timely detection and treatment. Currently, several screening and diagnostic tests are required to be carried out in order to detect PCa. These tests are often invasive, requiring either a biopsy (Gleason score and ISUP) or blood tests (PSA). Computational methods have been shown to help this process, using multiparametric MRI (mpMRI) data to detect PCa, effectively providing value during the diagnosis and monitoring stages. While delineating lesions requires a high degree of experience and expertise from the radiologists, being subject to a high degree of inter-observer variability, often leading to inconsistent readings, these computational models can leverage the information from mpMRI to locate the lesions with a high degree of certainty. By considering as positive samples only those that have an ISUP2 we can train aggressive index lesion detection models. The main advantage of this approach is that, by focusing only on aggressive disease, the output of such a model can also be seen as an indication for biopsy, effectively reducing unnecessary biopsy screenings. In this work, we utilize both the highly heterogeneous ProstateNet dataset, and the PI-CAI dataset, to develop accurate aggressive disease detection models.Source: SCIENTIFIC REPORTS, vol. 15 (issue 1)
DOI: 10.1038/s41598-025-99795-y
Project(s): ProCAncer-I
Metrics:
See at:
doaj.org | doi.org | CNR IRIS | www.nature.com | Archivio Istituzionale della Ricerca (AperTO) - Università di Torino | Archivio Istituzionale della Ricerca (AperTO) - Università di Torino | CNR IRIS | pubmed.ncbi.nlm.nih.gov
2025
Book
Open Access
Trustworthy AI in cancer imaging research
Chouvarda I., Colantonio S., Tsakou G., Yang G.
The book covers multiple aspects and challenges, from legal to technical and validation, in the emerging topic of AI in cancer imaging, bringing together the experience of top researchers and flagship projects. The aim of this book is to address the important questions: “How to design AI that is trustworthy”, and “How to validate AI trustworthiness” in the scope of AI for cancer imaging research. The book discusses overall considerations and the generation of a framework; preparing for trustworthy AI, including the data and metadata for quality, transparency and traceability; implementing trustworthy AI with algorithms and Decision Support Systems; and validating trustworthy AI. Chapters 2 and 3 are available open access under a Creative Commons Attribution 4.0 International License via Springerlink. This is an ideal resource for researchers from technical and clinical research sites, postgraduate students, and healthcare professionals in cancer imaging and beyond.DOI: 10.1007/978-3-031-89963-8
Project(s): An AI Platform integrating imaging data and models, supporting precision care through prostate cancer’s continuum
Metrics:
Chouvarda I., Colantonio S., Tsakou G., Yang G.
The book covers multiple aspects and challenges, from legal to technical and validation, in the emerging topic of AI in cancer imaging, bringing together the experience of top researchers and flagship projects. The aim of this book is to address the important questions: “How to design AI that is trustworthy”, and “How to validate AI trustworthiness” in the scope of AI for cancer imaging research. The book discusses overall considerations and the generation of a framework; preparing for trustworthy AI, including the data and metadata for quality, transparency and traceability; implementing trustworthy AI with algorithms and Decision Support Systems; and validating trustworthy AI. Chapters 2 and 3 are available open access under a Creative Commons Attribution 4.0 International License via Springerlink. This is an ideal resource for researchers from technical and clinical research sites, postgraduate students, and healthcare professionals in cancer imaging and beyond.DOI: 10.1007/978-3-031-89963-8
Project(s): An AI Platform integrating imaging data and models, supporting precision care through prostate cancer’s continuum
Metrics:
See at:
CNR IRIS | link.springer.com | CNR IRIS
2025
Other
Open Access
ISTI-day 2025 Proceedings
Del Corso G., Pedrotti A., Federico G., Gennaro C., Carrara F., Amato G., Di Benedetto M., Gabrielli E., Belli D., Matrullo Zoe, Miori V., Tolomei Gabriele, Waheed T., Marchetti E., Calabrò Antonello., Rossetti G., Stella Massimo, Cazabet Rémy, Abramski K., Cau E., Citraro S., Failla A., Mesina V., Morini V., Pansanella V., Colantonio S., Germanese D., Pascali M. A., Bianchi L., Messina N., Falchi F., Barsellotti L., Pacini G., Cassese M., Puccetti G., Esuli A., Volpi L., Moreo Alejandro, Sebastiani F., Sperduti G., Nguyen Dong, Broccia G., Ter Beek M. H., Ferrari A., Massink M., Belmonte Gina, Ciancia V., Papini O., Canapa G., Catricalà B., Manca M., Paternò F., Santoro C., Zedda E., Gallo S., Maenza S., Mattioli A., Simeoli L., Rucci D., Carlini E., Dazzi P., Kavalionak H., Mordacchini M., Rulli C., Muntean Cristina Ioana, Nardini F. M., Perego R., Rocchietti G., Lettich F., Renso C., Pugliese C., Casini G., Haldimann Jonas, Meyer Thomas, Assante M., Candela L., Dell'Amico A., Frosini L., Mangiacrapa F., Oliviero A., Pagano P., Panichi G., Peccerillo B., Procaccini M., Mannocci A., Manghi P., Lonetti F., Kang Dongjae, Di Giandomenico F., Jee Eunkyoung, Lazzini G., Conti F., Scopigno R., D'Acunto M., Moroni D., Cafiso M., Paradisi P., Callieri M., Pavoni G., Corsini M., De Falco A., Sala F., Saraceni Q., Gattiglia Gabriele
ISTI-Day is an annual information and networking event organized by the Institute of Information Science and Technologies "A. Faedo" (ISTI) of the Italian National Research Council (CNR). This event features an opening talk of the Director of the Dept. DIITET (Emilio F. Campana) as well as an overview of the Institute's activities presented by the ISTI Director (Roberto Scopigno). Those institutional segments are complemented by dedicated presentations and round tables featuring former staff members, as well as internal and external collaborators. To foster a network of knowledge and collaboration among newcomers, the 2025 ISTI Day edition also includes a large poster session that provides a comprehensive overview of current research activities. Each of the 13 laboratories contributes 1–3 posters, highlighting the most innovative work and offering early-career researchers a platform for discussion. Thus these proceedings include the posters selected for ISTI-Day 2025, reflecting the diverse and innovative nature of the Institute's research.
Del Corso G., Pedrotti A., Federico G., Gennaro C., Carrara F., Amato G., Di Benedetto M., Gabrielli E., Belli D., Matrullo Zoe, Miori V., Tolomei Gabriele, Waheed T., Marchetti E., Calabrò Antonello., Rossetti G., Stella Massimo, Cazabet Rémy, Abramski K., Cau E., Citraro S., Failla A., Mesina V., Morini V., Pansanella V., Colantonio S., Germanese D., Pascali M. A., Bianchi L., Messina N., Falchi F., Barsellotti L., Pacini G., Cassese M., Puccetti G., Esuli A., Volpi L., Moreo Alejandro, Sebastiani F., Sperduti G., Nguyen Dong, Broccia G., Ter Beek M. H., Ferrari A., Massink M., Belmonte Gina, Ciancia V., Papini O., Canapa G., Catricalà B., Manca M., Paternò F., Santoro C., Zedda E., Gallo S., Maenza S., Mattioli A., Simeoli L., Rucci D., Carlini E., Dazzi P., Kavalionak H., Mordacchini M., Rulli C., Muntean Cristina Ioana, Nardini F. M., Perego R., Rocchietti G., Lettich F., Renso C., Pugliese C., Casini G., Haldimann Jonas, Meyer Thomas, Assante M., Candela L., Dell'Amico A., Frosini L., Mangiacrapa F., Oliviero A., Pagano P., Panichi G., Peccerillo B., Procaccini M., Mannocci A., Manghi P., Lonetti F., Kang Dongjae, Di Giandomenico F., Jee Eunkyoung, Lazzini G., Conti F., Scopigno R., D'Acunto M., Moroni D., Cafiso M., Paradisi P., Callieri M., Pavoni G., Corsini M., De Falco A., Sala F., Saraceni Q., Gattiglia Gabriele
ISTI-Day is an annual information and networking event organized by the Institute of Information Science and Technologies "A. Faedo" (ISTI) of the Italian National Research Council (CNR). This event features an opening talk of the Director of the Dept. DIITET (Emilio F. Campana) as well as an overview of the Institute's activities presented by the ISTI Director (Roberto Scopigno). Those institutional segments are complemented by dedicated presentations and round tables featuring former staff members, as well as internal and external collaborators. To foster a network of knowledge and collaboration among newcomers, the 2025 ISTI Day edition also includes a large poster session that provides a comprehensive overview of current research activities. Each of the 13 laboratories contributes 1–3 posters, highlighting the most innovative work and offering early-career researchers a platform for discussion. Thus these proceedings include the posters selected for ISTI-Day 2025, reflecting the diverse and innovative nature of the Institute's research.
See at:
CNR IRIS | www.isti.cnr.it | CNR IRIS
2025
Journal article
Open Access
A review of methods for trustworthy AI in medical imaging: The FUTURE-AI guidelines
Kondylakis H., Osuala R., Puig-Bosch X., Lazrak N., Diaz O., Kushibar K., Chouvarda I., Charalambous S., Starmans Martijn Pa, Colantonio S., Tachos N., Joshi S., Woodruff Henry C., Salahuddin Z., Tsakou G., Aussó S., Alberich Leonor C., Papanikolaou N., Lambin P., Marias K., Tsiknakis M., Fotiadis Dimitrios I., Martí-Bonmatí L., Lekadir K.
Recent advancements in artificial intelligence (AI) and the vast data generated by modern clinical systems have driven the development of AI solutions in medical imaging, encompassing image reconstruction, segmentation, diagnosis, and treatment planning. Despite these successes and potential, many stakeholders worry about the risks and ethical implications of imaging AI, viewing it as complex, opaque, and challenging to understand, use, and trust in critical clinical applications. The FUTURE-AI guideline for trustworthy AI in healthcare was established based on six guiding principles: Fairness, Universality, Traceability, Usability, Robustness, and Explainability. Through international consensus, a set of recommendations was defined, covering the entire lifecycle of medical AI tools, from design, development, and validation to regulation, deployment, and monitoring. In this paper, we describe how these specific recommendations can be instantiated in the domain of medical imaging, providing an overview of current best practices along with guidelines and concrete metrics on how those recommendations could be met, offering a valuable resource to the international medical imaging community.Source: IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS
DOI: 10.1109/jbhi.2025.3614546
Project(s): ProCAncer-I
Metrics:
Kondylakis H., Osuala R., Puig-Bosch X., Lazrak N., Diaz O., Kushibar K., Chouvarda I., Charalambous S., Starmans Martijn Pa, Colantonio S., Tachos N., Joshi S., Woodruff Henry C., Salahuddin Z., Tsakou G., Aussó S., Alberich Leonor C., Papanikolaou N., Lambin P., Marias K., Tsiknakis M., Fotiadis Dimitrios I., Martí-Bonmatí L., Lekadir K.
Recent advancements in artificial intelligence (AI) and the vast data generated by modern clinical systems have driven the development of AI solutions in medical imaging, encompassing image reconstruction, segmentation, diagnosis, and treatment planning. Despite these successes and potential, many stakeholders worry about the risks and ethical implications of imaging AI, viewing it as complex, opaque, and challenging to understand, use, and trust in critical clinical applications. The FUTURE-AI guideline for trustworthy AI in healthcare was established based on six guiding principles: Fairness, Universality, Traceability, Usability, Robustness, and Explainability. Through international consensus, a set of recommendations was defined, covering the entire lifecycle of medical AI tools, from design, development, and validation to regulation, deployment, and monitoring. In this paper, we describe how these specific recommendations can be instantiated in the domain of medical imaging, providing an overview of current best practices along with guidelines and concrete metrics on how those recommendations could be met, offering a valuable resource to the international medical imaging community.Source: IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS
DOI: 10.1109/jbhi.2025.3614546
Project(s): ProCAncer-I
Metrics:
See at:
IEEE Journal of Biomedical and Health Informatics | CNR IRIS | ieeexplore.ieee.org | IEEE Journal of Biomedical and Health Informatics | IEEE Journal of Biomedical and Health Informatics | CNR IRIS
2025
Conference article
Open Access
Towards trustworthy AI in the public transport domain
Leone G. R., Carboni A., Del Corso G., Gravili S., Moroni D., Pascali M. A., Colantonio S.
In the context of rapidly evolving urban landscapes, the demand for enhanced mobility services has become increasingly critical. Traditional transportation systems struggle to keep pace with the growing complexity of commuting patterns and the diverse needs of urban residents. While AI can play a strong role in addressing these emerging demands, a parallel need for trustworthy services is also arising, which must be adequately met to ultimately provide equitable and ethical services to society. Based on these considerations, we explore the relevant dimensions of AI trustworthiness and propose how they can be transferred and demonstrated in a large-scale pilot focused on public transportation and exploiting advanced visual analytics paradigms based on pervasive computing. To this end, we present the FAITH risk management framework, ongoing activities, and preliminary results towards its implementation in the pilot project.Source: CEUR WORKSHOP PROCEEDINGS, vol. 4121. Trieste, Italy, 23-24/06/2025
Leone G. R., Carboni A., Del Corso G., Gravili S., Moroni D., Pascali M. A., Colantonio S.
In the context of rapidly evolving urban landscapes, the demand for enhanced mobility services has become increasingly critical. Traditional transportation systems struggle to keep pace with the growing complexity of commuting patterns and the diverse needs of urban residents. While AI can play a strong role in addressing these emerging demands, a parallel need for trustworthy services is also arising, which must be adequately met to ultimately provide equitable and ethical services to society. Based on these considerations, we explore the relevant dimensions of AI trustworthiness and propose how they can be transferred and demonstrated in a large-scale pilot focused on public transportation and exploiting advanced visual analytics paradigms based on pervasive computing. To this end, we present the FAITH risk management framework, ongoing activities, and preliminary results towards its implementation in the pilot project.Source: CEUR WORKSHOP PROCEEDINGS, vol. 4121. Trieste, Italy, 23-24/06/2025
See at:
ceur-ws.org | CNR IRIS | CNR IRIS
2025
Other
Open Access
Time series scribe: open-source platform for enhanced annotation in multi-channel signal processing
Del Corso G., Kanka S., Bardelli S., Visaggi P., Colantonio S.
Multi-channel temporal signal annotation for training predictive models requires open source software tools that are easy to use and adaptable to different real-world applications. This is especially true in the analysis of biological and medical signals, where the annotation process is not carried out by data scientists. Therefore, this technical report proposes an open source Python software package that allows the annotation of time signals through the use of an intuitive graphical interface. In particular, the annotation of the refluxes in a ph-impedance signal has been taken as a case study.DOI: 10.32079/isti-tr-2025/001
Metrics:
Del Corso G., Kanka S., Bardelli S., Visaggi P., Colantonio S.
Multi-channel temporal signal annotation for training predictive models requires open source software tools that are easy to use and adaptable to different real-world applications. This is especially true in the analysis of biological and medical signals, where the annotation process is not carried out by data scientists. Therefore, this technical report proposes an open source Python software package that allows the annotation of time signals through the use of an intuitive graphical interface. In particular, the annotation of the refluxes in a ph-impedance signal has been taken as a case study.DOI: 10.32079/isti-tr-2025/001
Metrics:
2025
Journal article
Open Access
Differences in technical and clinical perspectives on AI validation in cancer imaging: mind the gap!
Chouvarda I., Colantonio S., Verde Ana S. C., Jimenez-Pastor A., Cerdá-Alberich L., Metz Y., Zacharias L., Nabhani-Gebara S., Bobowicz M., Tsakou G., Lekadir K., Tsiknakis M., Martí-Bonmati L., Papanikolaou N.
Abstract: Good practices in artificial intelligence (AI) model validation are key for achieving trustworthy AI. Within the cancer imaging domain, attracting the attention of clinical and technical AI enthusiasts, this work discusses current gaps in AI validation strategies, examining existing practices that are common or variable across technical groups (TGs) and clinical groups (CGs). The work is based on a set of structured questions encompassing several AI validation topics, addressed to professionals working in AI for medical imaging. A total of 49 responses were obtained and analysed to identify trends and patterns. While TGs valued transparency and traceability the most, CGs pointed out the importance of explainability. Among the topics where TGs may benefit from further exposure are stability and robustness checks, and mitigation of fairness issues. On the other hand, CGs seemed more reluctant towards synthetic data for validation and would benefit from exposure to cross-validation techniques, or segmentation metrics. Topics emerging from the open questions were utility, capability, adoption and trustworthiness. These findings on current trends in AI validation strategies may guide the creation of guidelines necessary for training the next generation of professionals working with AI in healthcare and contribute to bridging any technical-clinical gap in AI validation. Relevance statement: This study recognised current gaps in understanding and applying AI validation strategies in cancer imaging and helped promote trust and adoption for interdisciplinary teams of technical and clinical researchers. Key Points: Clinical and technical researchers emphasise interpretability, external validation with diverse data, and bias awareness in AI validation for cancer imaging. In cancer imaging AI research, clinical researchers prioritise explainability, while technical researchers focus on transparency and traceability, and see potential in synthetic datasets. Researchers advocate for greater homogenisation of AI validation practices in cancer imaging.Source: EUROPEAN RADIOLOGY EXPERIMENTAL, vol. 9 (issue 1)
DOI: 10.1186/s41747-024-00543-0
Project(s): A European Cancer Image Platform Linked to Biological and Health Data for Next-Generation Artificial Intelligence and Precision Medicine in Oncology, INCISIVE, CHAIMELEON , ProCAncer-I , RadioVal , PRIMAGE
Metrics:
Chouvarda I., Colantonio S., Verde Ana S. C., Jimenez-Pastor A., Cerdá-Alberich L., Metz Y., Zacharias L., Nabhani-Gebara S., Bobowicz M., Tsakou G., Lekadir K., Tsiknakis M., Martí-Bonmati L., Papanikolaou N.
Abstract: Good practices in artificial intelligence (AI) model validation are key for achieving trustworthy AI. Within the cancer imaging domain, attracting the attention of clinical and technical AI enthusiasts, this work discusses current gaps in AI validation strategies, examining existing practices that are common or variable across technical groups (TGs) and clinical groups (CGs). The work is based on a set of structured questions encompassing several AI validation topics, addressed to professionals working in AI for medical imaging. A total of 49 responses were obtained and analysed to identify trends and patterns. While TGs valued transparency and traceability the most, CGs pointed out the importance of explainability. Among the topics where TGs may benefit from further exposure are stability and robustness checks, and mitigation of fairness issues. On the other hand, CGs seemed more reluctant towards synthetic data for validation and would benefit from exposure to cross-validation techniques, or segmentation metrics. Topics emerging from the open questions were utility, capability, adoption and trustworthiness. These findings on current trends in AI validation strategies may guide the creation of guidelines necessary for training the next generation of professionals working with AI in healthcare and contribute to bridging any technical-clinical gap in AI validation. Relevance statement: This study recognised current gaps in understanding and applying AI validation strategies in cancer imaging and helped promote trust and adoption for interdisciplinary teams of technical and clinical researchers. Key Points: Clinical and technical researchers emphasise interpretability, external validation with diverse data, and bias awareness in AI validation for cancer imaging. In cancer imaging AI research, clinical researchers prioritise explainability, while technical researchers focus on transparency and traceability, and see potential in synthetic datasets. Researchers advocate for greater homogenisation of AI validation practices in cancer imaging.Source: EUROPEAN RADIOLOGY EXPERIMENTAL, vol. 9 (issue 1)
DOI: 10.1186/s41747-024-00543-0
Project(s): A European Cancer Image Platform Linked to Biological and Health Data for Next-Generation Artificial Intelligence and Precision Medicine in Oncology, INCISIVE
, CHAIMELEON , ProCAncer-I , RadioVal , PRIMAGE Metrics:
See at:
European Radiology Experimental | European Radiology Experimental | CNR IRIS | European Radiology Experimental | CNR IRIS
2025
Other
Open Access
U-ProBE: Uncertainty Probabilistic Bayesian Estimate
Bandini L., Del Corso G., Colantonio S., Caudai C.
In this technical report we have designed and developed a Python software suite (U-ProBE: Uncertainty Probabilistic Bayesian Estimate) for analyzing Deep Learning models with predictions affected by uncertainty (i.e., Bayesian Probabilistic Models). The suite is equipped with an intuitive graphical interface that is simple to use even for non-experts and designed to support a growing pool of users who need to evaluate a model’s performance and, above all, its uncertainty.DOI: 10.32079/isti-tr-2025/006
Project(s): ProCAncer-I
Metrics:
Bandini L., Del Corso G., Colantonio S., Caudai C.
In this technical report we have designed and developed a Python software suite (U-ProBE: Uncertainty Probabilistic Bayesian Estimate) for analyzing Deep Learning models with predictions affected by uncertainty (i.e., Bayesian Probabilistic Models). The suite is equipped with an intuitive graphical interface that is simple to use even for non-experts and designed to support a growing pool of users who need to evaluate a model’s performance and, above all, its uncertainty.DOI: 10.32079/isti-tr-2025/006
Project(s): ProCAncer-I
Metrics:
2025
Conference article
Open Access
Leveraging AI for Signal and Image Analysis in Medicine and Health
Marco Cafiso, Andrea Carboni, Claudia Caudai, Sara Colantonio, Francesco Conti, Mario D’acunto, Said Daoudagh, Giulio Del Corso, Danila Germanese, Giacomo Ignesti, Gianmarco Lazzini, Giuseppe Riccardo Leone, Massimo Magrini, Davide Moroni, Francesca Pardini, Maria Antonietta Pascali, Paolo Paradisi, Federico Volpini
The integration of artificial intelligence (AI) into the medical domain is driving innovation and progress in healthcare. This paper summarizes the research activities that a multidisciplinary research group within the Signals and Images Lab of the Institute of Information Science and Technologies of the National Research Council of Italy is carrying out to explore the great potential of AI in several applications, e.g., in the analysis of biomedical data, and in the development of tools for enhancing trustworthiness and reliability of AI based systems. From cancer diagnosis and grading, to the analysis of body physiological signals to improve the understanding of dance movement therapy as an approach to healthy aging, this work highlights the paradigm shift that AI has brought into medicine and healthcare.Source: CEUR WORKSHOP PROCEEDINGS, vol. 4121. Trieste, June 23-24, 2025
Marco Cafiso, Andrea Carboni, Claudia Caudai, Sara Colantonio, Francesco Conti, Mario D’acunto, Said Daoudagh, Giulio Del Corso, Danila Germanese, Giacomo Ignesti, Gianmarco Lazzini, Giuseppe Riccardo Leone, Massimo Magrini, Davide Moroni, Francesca Pardini, Maria Antonietta Pascali, Paolo Paradisi, Federico Volpini
The integration of artificial intelligence (AI) into the medical domain is driving innovation and progress in healthcare. This paper summarizes the research activities that a multidisciplinary research group within the Signals and Images Lab of the Institute of Information Science and Technologies of the National Research Council of Italy is carrying out to explore the great potential of AI in several applications, e.g., in the analysis of biomedical data, and in the development of tools for enhancing trustworthiness and reliability of AI based systems. From cancer diagnosis and grading, to the analysis of body physiological signals to improve the understanding of dance movement therapy as an approach to healthy aging, this work highlights the paradigm shift that AI has brought into medicine and healthcare.Source: CEUR WORKSHOP PROCEEDINGS, vol. 4121. Trieste, June 23-24, 2025
See at:
ceur-ws.org | CNR IRIS | CNR IRIS
2025
Other
Open Access
Boosting few-shot learning with disentangled self-supervised learning and meta-learning for medical image classification
Pachetti E., Tsaftaris S. A., Colantonio S.
Background and objective: Employing deep learning models in critical domainssuch as medical imaging poses challenges associated with the limitedavailability of training data. We present a strategy for improving theperformance and generalization capabilities of models trained in low-dataregimes. Methods: The proposed method starts with a pre-training phase, wherefeatures learned in a self-supervised learning setting are disentangled toimprove the robustness of the representations for downstream tasks. We thenintroduce a meta-fine-tuning step, leveraging related classes betweenmeta-training and meta-testing phases but varying the granularity level. Thisapproach aims to enhance the model's generalization capabilities by exposing itto more challenging classification tasks during meta-training and evaluating iton easier tasks but holding greater clinical relevance during meta-testing. Wedemonstrate the effectiveness of the proposed approach through a series ofexperiments exploring several backbones, as well as diverse pre-training andfine-tuning schemes, on two distinct medical tasks, i.e., classification ofprostate cancer aggressiveness from MRI data and classification of breastcancer malignity from microscopic images. Results: Our results indicate thatthe proposed approach consistently yields superior performance w.r.t. ablationexperiments, maintaining competitiveness even when a distribution shift betweentraining and evaluation data occurs. Conclusion: Extensive experimentsdemonstrate the effectiveness and wide applicability of the proposed approach.We hope that this work will add another solution to the arsenal of addressinglearning issues in data-scarce imaging domains.DOI: 10.48550/arxiv.2403.17530
Metrics:
Pachetti E., Tsaftaris S. A., Colantonio S.
Background and objective: Employing deep learning models in critical domainssuch as medical imaging poses challenges associated with the limitedavailability of training data. We present a strategy for improving theperformance and generalization capabilities of models trained in low-dataregimes. Methods: The proposed method starts with a pre-training phase, wherefeatures learned in a self-supervised learning setting are disentangled toimprove the robustness of the representations for downstream tasks. We thenintroduce a meta-fine-tuning step, leveraging related classes betweenmeta-training and meta-testing phases but varying the granularity level. Thisapproach aims to enhance the model's generalization capabilities by exposing itto more challenging classification tasks during meta-training and evaluating iton easier tasks but holding greater clinical relevance during meta-testing. Wedemonstrate the effectiveness of the proposed approach through a series ofexperiments exploring several backbones, as well as diverse pre-training andfine-tuning schemes, on two distinct medical tasks, i.e., classification ofprostate cancer aggressiveness from MRI data and classification of breastcancer malignity from microscopic images. Results: Our results indicate thatthe proposed approach consistently yields superior performance w.r.t. ablationexperiments, maintaining competitiveness even when a distribution shift betweentraining and evaluation data occurs. Conclusion: Extensive experimentsdemonstrate the effectiveness and wide applicability of the proposed approach.We hope that this work will add another solution to the arsenal of addressinglearning issues in data-scarce imaging domains.DOI: 10.48550/arxiv.2403.17530
Metrics:
See at:
arXiv.org e-Print Archive | IRIS Cnr | IRIS Cnr | IRIS Cnr | doi.org | CNR IRIS