Chiara De Nardo
Research Subjects
Chiara De Nardo has a total of 13 co-authors. Here are the top co-authors by number of shared publications:
- Angelina Pernazza
Department of Radiological, Oncological and Pathological Sciences, Sapienza University of Rome, Rome, Italy
1 shared publication - Carlo Catalano
Department of Radiological Sciences, Oncology and Pathology, Policlinico Umberto I, Sapienza University, Rome, Italy
1 shared publication - Federica Cicciarelli
Department of Radiological, Oncological and Pathological Sciences, Sapienza, University of Rome, Policlinico “Umberto I”, Rome, Italy
1 shared publication - Federica Pediconi
Department of Radiological, Oncological and Pathological Sciences, Sapienza-University of Rome, Rome, Italy.
1 shared publication - Francesca Arienzo
Pathology Unit, Bambino Gesù Children’s Hospital, IRCCS, 00165 Rome, Italy
1 shared publication - Francesca Galati
Department of Radiological, Oncological and Pathological Sciences, Sapienza University of Rome, Piazzale Aldo Moro 5, 00185 Rome, Italy
1 shared publication
Latest Publications
- Open Access
- 29 June 2026
- Published by Oxford University Press (OUP) in BJR|Artificial Intelligence
- Vol. 3 (1)
Predictive Deep Learning Model based on Contrast-Enhanced Mammography for Breast Cancer Diagnosis: A Pilot Study
Objectives: The interpretation of contrast-enhanced mammography (CEM) images heavily depends on radiologists’ expertise, highlighting the need for automated tools to assist clinical decision-making.This pilot study aimed to develop a deep learning model using CEM images to predict the histological diagnosis of breast cancer. Methods: We retrospectively analyzed patients who underwent contrast-enhanced mammography (CEM) followed by histopathological assessment (October 2022-May 2023) across two centers. CEM images from center 1 were used for model development, including training and 10-fold cross-validation, while images from center 2 served as an independent external test set. The breast region was manually segmented, and two deep learning architectures were implemented as ensemble classifiers, using biopsy results as reference standard. Performance metrics included accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and area under the ROC curve (AUC). Results: A total of 106 CEM images were retrospectively analyzed (50 from center 1 and 56 from center 2), obtained using different mammography systems. Histopathology identified 51 lesions (48%) as malignant and 55 (52%) as benign. Cross-validation yielded a ROC-AUC of 75% [58.4–91.6], accuracy of 68.7% [58.3–79], sensitivity of 68.9% [52.2–85.6], specificity of 67.8% [65.4–70.2], PPV of 69.3% [55.5–83.1], and NPV of 73.5% [55.2–91.8] (p < 0.05). External testing on images from center 2 (27 malignant, 29 benign) achieved an accuracy of 64.3%. Conclusions: The study showed promising performance but requires further development and dataset expansion. The model has the potential to be integrated into clinical practice. Advances in knowledge: A deep learning model based CEM provides a tool to support physicians in the decision-making process and reduce invasiveness in breast cancer management.
- 29 June 2026
- Published by Oxford University Press (OUP) in BJR|Artificial Intelligence
- Vol. 3 (1)
Open AccessPredictive Deep Learning Model based on Contrast-Enhanced Mammography for Breast Cancer Diagnosis: A Pilot Study
Objectives: The interpretation of contrast-enhanced mammography (CEM) images heavily depends on radiologists’ expertise, highlighting the need for automated tools to assist clinical decision-making.This pilot study aimed to develop a deep learning model using CEM images to predict the histological diagnosis of breast cancer. Methods: We retrospectively analyzed patients who underwent contrast-enhanced mammography (CEM) followed by histopathological assessment (October 2022-May 2023) across two centers. CEM images from center 1 were used for model development, including training and 10-fold cross-validation, while images from center 2 served as an independent external test set. The breast region was manually segmented, and two deep learning architectures were implemented as ensemble classifiers, using biopsy results as reference standard. Performance metrics included accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and area under the ROC curve (AUC). Results: A total of 106 CEM images were retrospectively analyzed (50 from center 1 and 56 from center 2), obtained using different mammography systems. Histopathology identified 51 lesions (48%) as malignant and 55 (52%) as benign. Cross-validation yielded a ROC-AUC of 75% [58.4–91.6], accuracy of 68.7% [58.3–79], sensitivity of 68.9% [52.2–85.6], specificity of 67.8% [65.4–70.2], PPV of 69.3% [55.5–83.1], and NPV of 73.5% [55.2–91.8] (p < 0.05). External testing on images from center 2 (27 malignant, 29 benign) achieved an accuracy of 64.3%. Conclusions: The study showed promising performance but requires further development and dataset expansion. The model has the potential to be integrated into clinical practice. Advances in knowledge: A deep learning model based CEM provides a tool to support physicians in the decision-making process and reduce invasiveness in breast cancer management.
- Open Access
- 12 September 2023
- Published by Springer Nature in Egyptian Journal of Radiology and Nuclear Medicine
- Vol. 54 (1) , 1-5
“Cheerios in the lung” as first metastases from endometrial endometrioid adenocarcinoma with adequate response to immunotherapy
Background: Metastases from endometrial carcinoma to the lungs are uncommon and cavitating metastases are even rarer. In some cases, lung cavitating metastases may resemble diffuse “Cheerios in the lung.” Case presentation: A 58-year-old woman, smoker, with a history of hysteroannessiectomy and pelvic lymphadenectomy for a low stage endometrial endometrioid adenocarcinoma, came under our attention for dry cough without fever that persisted for over three months, and computed tomography (CT) revealed diffuse cavitating lung metastases resembling "Cheerios in the lung". After standard chemotherapy proved ineffective, Lenvatinib plus Pembrolizumab (immunotherapy) was administrated, leading to the disappearance of all lung nodules, leaving behind a diffuse emphysematous-like pattern. Conclusions: This case is a rare example of metastatic endometrioid adenocarcinoma, manifested with chronic dry cough, with “Cheerios in the lung” appearence. The metastases responded well to immunotherapy, resulting in the regression of nodules into areas of diffuse emphysematous-like changes.
- 12 September 2023
- Published by Springer Nature in Egyptian Journal of Radiology and Nuclear Medicine
- Vol. 54 (1) , 1-5
Open Access“Cheerios in the lung” as first metastases from endometrial endometrioid adenocarcinoma with adequate response to immunotherapy
Background: Metastases from endometrial carcinoma to the lungs are uncommon and cavitating metastases are even rarer. In some cases, lung cavitating metastases may resemble diffuse “Cheerios in the lung.” Case presentation: A 58-year-old woman, smoker, with a history of hysteroannessiectomy and pelvic lymphadenectomy for a low stage endometrial endometrioid adenocarcinoma, came under our attention for dry cough without fever that persisted for over three months, and computed tomography (CT) revealed diffuse cavitating lung metastases resembling "Cheerios in the lung". After standard chemotherapy proved ineffective, Lenvatinib plus Pembrolizumab (immunotherapy) was administrated, leading to the disappearance of all lung nodules, leaving behind a diffuse emphysematous-like pattern. Conclusions: This case is a rare example of metastatic endometrioid adenocarcinoma, manifested with chronic dry cough, with “Cheerios in the lung” appearence. The metastases responded well to immunotherapy, resulting in the regression of nodules into areas of diffuse emphysematous-like changes.