Typology
operator:
and / or
Language
operator:
and / or
Date
operator:
and / or
Rights
operator:
and / or
2023
Conference article
Open Access
On the applicability of prototypical part learning in medical images: breast masses classification using ProtoPNet
Carloni G., Berti A., Iacconi C., Pascali M. A., Colantonio S.
Deep learning models have become state-of-the-art in many areas, ranging from computer vision to agriculture research. However, concerns have been raised with respect to the transparency of their decisions, especially in the image domain. In this regard, Explainable Artificial Intelligence has been gaining popularity in recent years. The ProtoPNet model, which breaks down an image into prototypes and uses evidence gathered from the prototypes to classify an image, represents an appealing approach. Still, questions regarding its effectiveness arise when the application domain changes from real-world natural images to gray-scale medical images. This work explores the applicability of prototypical part learning in medical imaging by experimenting with ProtoPNet on a breast masses classification task. The two considered aspects were the classification capabilities and the validity of explanations. We looked for the optimal model's hyperparameter configuration via a random search. We trained the model in a five-fold CV supervised framework, with mammogram images cropped around the lesions and ground-truth labels of benign/malignant masses. Then, we compared the performance metrics of ProtoPNet to that of the corresponding base architecture, which was ResNet18, trained under the same framework. In addition, an experienced radiologist provided a clinical viewpoint on the quality of the learned prototypes, the patch activations, and the global explanations. We achieved a Recall of 0.769 and an area under the receiver operating characteristic curve of 0.719 in our experiments. Even though our findings are non-optimal for entering the clinical practice yet, the radiologist found ProtoPNet's explanations very intuitive, reporting a high level of satisfaction. Therefore, we believe that prototypical part learning offers a reasonable and promising trade-off between classification performance and the quality of the related explanation.Source: ICPR 2022 - International Conference on Pattern Recognition - ICPR 2022 International Workshops and Challenges, pp. 539–557, Montreal, Canada, 21-25/08/2022
DOI: 10.1007/978-3-031-37660-3_38
Metrics:
Carloni G., Berti A., Iacconi C., Pascali M. A., Colantonio S.
Deep learning models have become state-of-the-art in many areas, ranging from computer vision to agriculture research. However, concerns have been raised with respect to the transparency of their decisions, especially in the image domain. In this regard, Explainable Artificial Intelligence has been gaining popularity in recent years. The ProtoPNet model, which breaks down an image into prototypes and uses evidence gathered from the prototypes to classify an image, represents an appealing approach. Still, questions regarding its effectiveness arise when the application domain changes from real-world natural images to gray-scale medical images. This work explores the applicability of prototypical part learning in medical imaging by experimenting with ProtoPNet on a breast masses classification task. The two considered aspects were the classification capabilities and the validity of explanations. We looked for the optimal model's hyperparameter configuration via a random search. We trained the model in a five-fold CV supervised framework, with mammogram images cropped around the lesions and ground-truth labels of benign/malignant masses. Then, we compared the performance metrics of ProtoPNet to that of the corresponding base architecture, which was ResNet18, trained under the same framework. In addition, an experienced radiologist provided a clinical viewpoint on the quality of the learned prototypes, the patch activations, and the global explanations. We achieved a Recall of 0.769 and an area under the receiver operating characteristic curve of 0.719 in our experiments. Even though our findings are non-optimal for entering the clinical practice yet, the radiologist found ProtoPNet's explanations very intuitive, reporting a high level of satisfaction. Therefore, we believe that prototypical part learning offers a reasonable and promising trade-off between classification performance and the quality of the related explanation.Source: ICPR 2022 - International Conference on Pattern Recognition - ICPR 2022 International Workshops and Challenges, pp. 539–557, Montreal, Canada, 21-25/08/2022
DOI: 10.1007/978-3-031-37660-3_38
Metrics:
See at:
ISTI Repository 
| link.springer.com 
| CNR ExploRA
2023
Journal article
Open Access
Brain metastases from NSCLC treated with stereotactic radiotherapy: prediction mismatch between two different radiomic platforms
Carloni G., Garibaldi C., Marvaso G., Volpe S., Zaffaroni M., Pepa M., Isaksson L. J., Colombo F., Durante S., Lo Presti G., Raimondi S., Spaggiari L. J., De Marinis F., Piperno G., Vigorito S., Gandini S., Cremonesi M., Positano V., Jereczek-Fossa B. A.
Background and purpose. Radiomics enables the mining of quantitative features from medical images. The influence of the radiomic feature extraction software on the final performance of models is still a poorly understood topic. This study aimed to investigate the ability of radiomic features extracted by two different radiomic platforms to predict clinical outcomes in patients treated with radiosurgery for brain metastases from non-small cell lung cancer. We developed models integrating pre-treatment magnetic resonance imaging (MRI)-derived radiomic features and clinical data. Materials and methods. Pre-radiotherapy gadolinium enhanced axial T1-weighted MRI scans were used. MRI images were re-sampled, intensity-shifted, and histogram-matched before radiomic extraction by means of two different platforms (PyRadiomics and SOPHiA Radiomics). We adopted LASSO Cox regression models for multivariable analyses by creating radiomic, clinical, and combined models using three survival clinical endpoints (local control, distant progression, and overall survival). The statistical analysis was repeated 50 times with different random seeds and the median concordance index was used as performance metric of the models. Results. We analysed 276 metastases from 148 patients. The use of the two platforms resulted in differences in both the quality and the number of extractable features. That led to mismatches in terms of end-to-end performance, statistical significance of radiomic scores, and clinical covariates found significant in combined models. Conclusion. This study shed new light on how extracting radiomic features from the same images using two different platforms could yield several discrepancies. That may lead to acute consequences on drawing conclusions, comparing results across the literature, and translating radiomics into clinical practice.Source: Radiotherapy and oncology 178 (2023). doi:10.1016/j.radonc.2022.11.013
DOI: 10.1016/j.radonc.2022.11.013
Metrics:
Carloni G., Garibaldi C., Marvaso G., Volpe S., Zaffaroni M., Pepa M., Isaksson L. J., Colombo F., Durante S., Lo Presti G., Raimondi S., Spaggiari L. J., De Marinis F., Piperno G., Vigorito S., Gandini S., Cremonesi M., Positano V., Jereczek-Fossa B. A.
Background and purpose. Radiomics enables the mining of quantitative features from medical images. The influence of the radiomic feature extraction software on the final performance of models is still a poorly understood topic. This study aimed to investigate the ability of radiomic features extracted by two different radiomic platforms to predict clinical outcomes in patients treated with radiosurgery for brain metastases from non-small cell lung cancer. We developed models integrating pre-treatment magnetic resonance imaging (MRI)-derived radiomic features and clinical data. Materials and methods. Pre-radiotherapy gadolinium enhanced axial T1-weighted MRI scans were used. MRI images were re-sampled, intensity-shifted, and histogram-matched before radiomic extraction by means of two different platforms (PyRadiomics and SOPHiA Radiomics). We adopted LASSO Cox regression models for multivariable analyses by creating radiomic, clinical, and combined models using three survival clinical endpoints (local control, distant progression, and overall survival). The statistical analysis was repeated 50 times with different random seeds and the median concordance index was used as performance metric of the models. Results. We analysed 276 metastases from 148 patients. The use of the two platforms resulted in differences in both the quality and the number of extractable features. That led to mismatches in terms of end-to-end performance, statistical significance of radiomic scores, and clinical covariates found significant in combined models. Conclusion. This study shed new light on how extracting radiomic features from the same images using two different platforms could yield several discrepancies. That may lead to acute consequences on drawing conclusions, comparing results across the literature, and translating radiomics into clinical practice.Source: Radiotherapy and oncology 178 (2023). doi:10.1016/j.radonc.2022.11.013
DOI: 10.1016/j.radonc.2022.11.013
Metrics:
See at:
ISTI Repository | www.sciencedirect.com | CNR ExploRA
2023
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 (2023).
Project(s): ProCAncer-I
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 (2023).
Project(s): ProCAncer-I
See at:
arxiv.org | ISTI Repository | CNR ExploRA
2023
Conference article
Open Access
Causality-Driven One-Shot learning for prostate cancer grading from MRI
Carloni G., Pachetti E., Colantonio S.
In this paper, we present a novel method to automatically classify medical images that learns and leverages weak causal signals in the image. Our framework consists of a convolutional neural network backbone and a causality-extractor module that extracts cause-effect relationships between feature maps that can inform the model on the appearance of a feature in one place of the image, given the presence of another feature within some other place of the image. To evaluate the effectiveness of our approach in low-data scenarios, we train our causality-driven architecture in a One-shot learning scheme, where we propose a new meta-learning procedure entailing meta-training and meta-testing tasks that are designed using related classes but at different levels of granularity. We conduct binary and multi-class classification experiments on a publicly available dataset of prostate MRI images. To validate the effectiveness of the proposed causality-driven module, we perform an ablation study and conduct qualitative assessments using class activation maps to highlight regions strongly influencing the network's decision-making process. Our findings show that causal relationships among features play a crucial role in enhancing the model's ability to discern relevant information and yielding more reliable and interpretable predictions. This would make it a promising approach for medical image classification tasks.Source: ICCV 2023 - International Conference on Computer Vision. Computer Vision for Automated Medical Diagnosis Workshop, Parigi, Francia, 02/10/2023
Carloni G., Pachetti E., Colantonio S.
In this paper, we present a novel method to automatically classify medical images that learns and leverages weak causal signals in the image. Our framework consists of a convolutional neural network backbone and a causality-extractor module that extracts cause-effect relationships between feature maps that can inform the model on the appearance of a feature in one place of the image, given the presence of another feature within some other place of the image. To evaluate the effectiveness of our approach in low-data scenarios, we train our causality-driven architecture in a One-shot learning scheme, where we propose a new meta-learning procedure entailing meta-training and meta-testing tasks that are designed using related classes but at different levels of granularity. We conduct binary and multi-class classification experiments on a publicly available dataset of prostate MRI images. To validate the effectiveness of the proposed causality-driven module, we perform an ablation study and conduct qualitative assessments using class activation maps to highlight regions strongly influencing the network's decision-making process. Our findings show that causal relationships among features play a crucial role in enhancing the model's ability to discern relevant information and yielding more reliable and interpretable predictions. This would make it a promising approach for medical image classification tasks.Source: ICCV 2023 - International Conference on Computer Vision. Computer Vision for Automated Medical Diagnosis Workshop, Parigi, Francia, 02/10/2023
See at:
arxiv.org | ISTI Repository | CNR ExploRA
2023
Conference article
Open Access
Cine cardiac MRI reconstruction using a convolutional recurrent network with refinement
Xue Y., Du Y., Carloni G., Pachetti E., Jordan C., Tsaftaris S. A.
Cine Magnetic Resonance Imaging (MRI) allows for understanding of the heart's function and condition in a non-invasive manner. Undersampling of the k-space is employed to reduce the scan duration, thus increasing patient comfort and reducing the risk of motion artefacts, at the cost of reduced image quality. In this challenge paper, we investigate the use of a convolutional recurrent neural network (CRNN) architecture to exploit temporal correlations in supervised cine cardiac MRI reconstruction. This is combined with a single-image super-resolution refinement module to improve single coil reconstruction by 4.4% in structural similarity and 3.9% in normalised mean square error compared to a plain CRNN implementation. We deploy a high-pass filter to our l1 loss to allow greater emphasis on high-frequency details which are missing in the original data. The proposed model demonstrates considerable enhancements compared to the baseline case and holds promising potential for further improving cardiac MRI reconstruction.Source: MICCAI 2023 - 26th International Conference on Medical Image Computing and Computer Assisted Intervention, Vancouver, Canada, 08-12/10/2023
Xue Y., Du Y., Carloni G., Pachetti E., Jordan C., Tsaftaris S. A.
Cine Magnetic Resonance Imaging (MRI) allows for understanding of the heart's function and condition in a non-invasive manner. Undersampling of the k-space is employed to reduce the scan duration, thus increasing patient comfort and reducing the risk of motion artefacts, at the cost of reduced image quality. In this challenge paper, we investigate the use of a convolutional recurrent neural network (CRNN) architecture to exploit temporal correlations in supervised cine cardiac MRI reconstruction. This is combined with a single-image super-resolution refinement module to improve single coil reconstruction by 4.4% in structural similarity and 3.9% in normalised mean square error compared to a plain CRNN implementation. We deploy a high-pass filter to our l1 loss to allow greater emphasis on high-frequency details which are missing in the original data. The proposed model demonstrates considerable enhancements compared to the baseline case and holds promising potential for further improving cardiac MRI reconstruction.Source: MICCAI 2023 - 26th International Conference on Medical Image Computing and Computer Assisted Intervention, Vancouver, Canada, 08-12/10/2023
See at:
ISTI Repository | CNR ExploRA
2023
Conference article
Open Access
Exploring the potentials and challenges of AI in supporting clinical diagnostics and remote assistance for the health and well-being of individuals
Berti A., Buongiorno R., Carloni G., Caudai C., Del Corso G., Germanese D., Pachetti E., Pascali M. A., Colantonio S.
Innovative technologies powered by Artificial Intelligence have the big potential to support new models of care delivery, disease prevention and quality of life promotion. The ultimate goal is a paradigm shift towards more personalized, accessible, effective, and sustainable care and health systems. Nevertheless, despite the advances in the field over the last years, the adoption and deployment of AI technologies remains limited in clinical practice and real-world settings. This paper summarizes the 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 for exploring both the potential of AI in health and well-being as well as the challenges to their uptake in real-world settingsSource: Ital-IA 2023 - Italia Intelligenza Artificiale. Thematic Workshops of the 3rd CINI National Lab AIIS Conference on Artificial Intelligence - 2023, Pisa, Italy, 29-30/05/2023
Project(s): ProCAncer-I
Berti A., Buongiorno R., Carloni G., Caudai C., Del Corso G., Germanese D., Pachetti E., Pascali M. A., Colantonio S.
Innovative technologies powered by Artificial Intelligence have the big potential to support new models of care delivery, disease prevention and quality of life promotion. The ultimate goal is a paradigm shift towards more personalized, accessible, effective, and sustainable care and health systems. Nevertheless, despite the advances in the field over the last years, the adoption and deployment of AI technologies remains limited in clinical practice and real-world settings. This paper summarizes the 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 for exploring both the potential of AI in health and well-being as well as the challenges to their uptake in real-world settingsSource: Ital-IA 2023 - Italia Intelligenza Artificiale. Thematic Workshops of the 3rd CINI National Lab AIIS Conference on Artificial Intelligence - 2023, Pisa, Italy, 29-30/05/2023
Project(s): ProCAncer-I
See at:
ceur-ws.org | ISTI Repository | CNR ExploRA
2022
Conference article
Open Access
Data models for an imaging bio-bank for colorectal, prostate and gastric cancer: the NAVIGATOR project
Berti A., Carloni G., Colantonio S., Pascali M. A., Manghi P., Pagano P., Buongiorno R., Pachetti E., Caudai C., Di Gangi D., Carlini E., Falaschi Z., Ciarrocchi E., Neri E., Bertelli E., Miele V., Carpi R., Bagnacci G., Di Meglio N., Mazzei M. A., Barucci A.
Researchers nowadays may take advantage of broad collections of medical data to develop personalized medicine solutions. Imaging bio-banks play a fundamental role, in this regard, by serving as organized repositories of medical images associated with imaging biomarkers. In this context, the NAVIGATOR Project aims to advance colorectal, prostate, and gastric oncology translational research by leveraging quantitative imaging and multi-omics analyses. As Project's core, an imaging bio-bank is being designed and implemented in a web-accessible Virtual Research Environment (VRE). The VRE serves to extract the imaging biomarkers and further process them within prediction algorithms. In our work, we present the realization of the data models for the three cancer use-cases of the Project. First, we carried out an extensive requirements analysis to fulfill the necessities of the clinical partners involved in the Project. Then, we designed three separate data models utilizing entity-relationship diagrams. We found diagrams' modeling for colorectal and prostate cancers to be more straightforward, while gastric cancer required a higher level of complexity. Future developments of this work would include designing a common data model following the Observational Medical Outcomes Partnership Standards. Indeed, a common data model would standardize the logical infrastructure of data models and make the bio-bank easily interoperable with other bio-banks.Source: BHI '22 - IEEE-EMBS International Conference on Biomedical and Health Informatics, Ioannina, Greece, 27-30/09/2022
DOI: 10.1109/bhi56158.2022.9926910
Metrics:
Berti A., Carloni G., Colantonio S., Pascali M. A., Manghi P., Pagano P., Buongiorno R., Pachetti E., Caudai C., Di Gangi D., Carlini E., Falaschi Z., Ciarrocchi E., Neri E., Bertelli E., Miele V., Carpi R., Bagnacci G., Di Meglio N., Mazzei M. A., Barucci A.
Researchers nowadays may take advantage of broad collections of medical data to develop personalized medicine solutions. Imaging bio-banks play a fundamental role, in this regard, by serving as organized repositories of medical images associated with imaging biomarkers. In this context, the NAVIGATOR Project aims to advance colorectal, prostate, and gastric oncology translational research by leveraging quantitative imaging and multi-omics analyses. As Project's core, an imaging bio-bank is being designed and implemented in a web-accessible Virtual Research Environment (VRE). The VRE serves to extract the imaging biomarkers and further process them within prediction algorithms. In our work, we present the realization of the data models for the three cancer use-cases of the Project. First, we carried out an extensive requirements analysis to fulfill the necessities of the clinical partners involved in the Project. Then, we designed three separate data models utilizing entity-relationship diagrams. We found diagrams' modeling for colorectal and prostate cancers to be more straightforward, while gastric cancer required a higher level of complexity. Future developments of this work would include designing a common data model following the Observational Medical Outcomes Partnership Standards. Indeed, a common data model would standardize the logical infrastructure of data models and make the bio-bank easily interoperable with other bio-banks.Source: BHI '22 - IEEE-EMBS International Conference on Biomedical and Health Informatics, Ioannina, Greece, 27-30/09/2022
DOI: 10.1109/bhi56158.2022.9926910
Metrics:
See at:
ISTI Repository | ieeexplore.ieee.org | CNR ExploRA
2022
Report
Open Access
SI-Lab annual research report 2021
Righi M., Leone G. R., Carboni A., Caudai C., Colantonio S., Kuruoglu E. E., Leporini B., Magrini M., Paradisi P., Pascali M. A., Pieri G., Reggiannini M., Salerno E., Scozzari A., Tonazzini A., Fusco G., Galesi G., Martinelli M., Pardini F., Tampucci M., Berti A., Bruno A., Buongiorno R., Carloni G., Conti F., Germanese D., Ignesti G., Matarese F., Omrani A., Pachetti E., Papini O., Benassi A., Bertini G., Coltelli P., Tarabella L., Straface S., Salvetti O., Moroni D.
The Signal & Images Laboratory 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 2021.Source: ISTI Annual reports, 2022
DOI: 10.32079/isti-ar-2022/003
Metrics:
Righi M., Leone G. R., Carboni A., Caudai C., Colantonio S., Kuruoglu E. E., Leporini B., Magrini M., Paradisi P., Pascali M. A., Pieri G., Reggiannini M., Salerno E., Scozzari A., Tonazzini A., Fusco G., Galesi G., Martinelli M., Pardini F., Tampucci M., Berti A., Bruno A., Buongiorno R., Carloni G., Conti F., Germanese D., Ignesti G., Matarese F., Omrani A., Pachetti E., Papini O., Benassi A., Bertini G., Coltelli P., Tarabella L., Straface S., Salvetti O., Moroni D.
The Signal & Images Laboratory 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 2021.Source: ISTI Annual reports, 2022
DOI: 10.32079/isti-ar-2022/003
Metrics:
See at:
ISTI Repository | CNR ExploRA