Joana Ip
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Joana Ip has a total of 52 co-authors. Here are the top co-authors by number of shared publications:
Ana Mascarenhas Gaivão
Radiology Department, Champalimaud Clinical Center, Champalimaud Foundation, Lisbon, Portugal
2 shared publications- Ana Sofia Castro Verde
Computational Clinical Imaging Group, Champalimaud Research, Champalimaud Foundation
2 shared publications - Carlos Bilreiro
Radiology Department, Champalimaud Clinical Center, Champalimaud Foundation, Lisbon, Portugal
2 shared publications Daniele Regge
Candiolo Cancer Institute, FPO-IRCCS, Candiolo, Turin, Italy
2 shared publications- José Guilherme de Almeida
Computational Clinical Imaging Group, Champalimaud Research, Champalimaud Foundation
2 shared publications - Kostas Marias
Computational BioMedicine Laboratory (CBML), Foundation for Research and Technology—Hellas (FORTH), 70013 Heraklion, Greece
2 shared publications
Latest Publications
- 1 September 2025
- Published by Radiological Society of North America (RSNA) in Radiology: Imaging Cancer
- Vol. 7 (5) , e240507
Improving Clinically Significant Prostate Cancer Detection with a Multimodal Machine Learning Approach: A Large-Scale Multicenter Study
A multimodal machine learning model combining radiomic, radiologic, and clinical variables reduced unnecessary prostate biopsies while maintaining high sensitivity for clinically significant cancer detection.
- 1 September 2025
- Published by Radiological Society of North America (RSNA) in Radiology: Imaging Cancer
- Vol. 7 (5) , e240507
Improving Clinically Significant Prostate Cancer Detection with a Multimodal Machine Learning Approach: A Large-Scale Multicenter Study
A multimodal machine learning model combining radiomic, radiologic, and clinical variables reduced unnecessary prostate biopsies while maintaining high sensitivity for clinically significant cancer detection.
- Open Access
Breath Insights: Advancing Lung Cancer Early-Stage Detection Through AI Algorithms in Non-Invasive VOC Profiling Trials
Background: Lung cancer (LC) is the leading cause of cancer-related deaths worldwide. Effective screening strategies for early diagnosis that could improve disease prognosis are lacking. Non-invasive breath analysis of volatile organic compounds (VOC) is a potential method for earlier LC detection. This study explores the association of VOC profiles with artificial intelligence (AI) to achieve a sensitive, specific, and fast method for LC detection. Patients and methods: Exhaled breath air samples were collected from 123 healthy individuals and 73 LC patients at two clinical sites. The enrolled patients had LC diagnosed with different stages. Breath samples were collected before undergoing any treatment, including surgery, and analyzed using gas chromatography coupled to ion-mobility spectrometry (GC-IMS). AI methods classified the overall chromatographic profiles. Results: GC-IMS is highly sensitive, yielding detailed chromatographic profiles. AI methods ranked the sets of exhaled breath profiles across both groups through training and validation steps, while qualitative information was deliberately not taking part nor influencing the results. The K-nearest neighbor (KNN) algorithm classified the groups with an accuracy of 90% (sensitivity = 87%, specificity = 92%). Narrowing the LC group to those only in early-stage IA, the accuracy was 90% (sensitivity = 90%, specificity = 93%). Conclusions: Evaluation of the global exhaled breath profiles using AI algorithms enabled LC detection and demonstrated that qualitative information may not be required, thus easing the frustration that many studies have experienced so far. The results show that this approach coupled with screening protocols may improve earlier detection of LC and hence its prognosis.
Open AccessBreath Insights: Advancing Lung Cancer Early-Stage Detection Through AI Algorithms in Non-Invasive VOC Profiling Trials
Background: Lung cancer (LC) is the leading cause of cancer-related deaths worldwide. Effective screening strategies for early diagnosis that could improve disease prognosis are lacking. Non-invasive breath analysis of volatile organic compounds (VOC) is a potential method for earlier LC detection. This study explores the association of VOC profiles with artificial intelligence (AI) to achieve a sensitive, specific, and fast method for LC detection. Patients and methods: Exhaled breath air samples were collected from 123 healthy individuals and 73 LC patients at two clinical sites. The enrolled patients had LC diagnosed with different stages. Breath samples were collected before undergoing any treatment, including surgery, and analyzed using gas chromatography coupled to ion-mobility spectrometry (GC-IMS). AI methods classified the overall chromatographic profiles. Results: GC-IMS is highly sensitive, yielding detailed chromatographic profiles. AI methods ranked the sets of exhaled breath profiles across both groups through training and validation steps, while qualitative information was deliberately not taking part nor influencing the results. The K-nearest neighbor (KNN) algorithm classified the groups with an accuracy of 90% (sensitivity = 87%, specificity = 92%). Narrowing the LC group to those only in early-stage IA, the accuracy was 90% (sensitivity = 90%, specificity = 93%). Conclusions: Evaluation of the global exhaled breath profiles using AI algorithms enabled LC detection and demonstrated that qualitative information may not be required, thus easing the frustration that many studies have experienced so far. The results show that this approach coupled with screening protocols may improve earlier detection of LC and hence its prognosis.
- 1 May 2025
- Published by Radiological Society of North America (RSNA) in Radiology: Artificial Intelligence
- Vol. 7 (3) , e230555
Impact of Scanner Manufacturer, Endorectal Coil Use, and Clinical Variables on Deep Learning–assisted Prostate Cancer Classification Using Multiparametric MRI
Aggressiveness of prostate cancer could be predicted using biparametric MRI and deep learning with negligible expert input, but performance was affected by scanner manufacturer and scanning protocol.
- 1 May 2025
- Published by Radiological Society of North America (RSNA) in Radiology: Artificial Intelligence
- Vol. 7 (3) , e230555
Impact of Scanner Manufacturer, Endorectal Coil Use, and Clinical Variables on Deep Learning–assisted Prostate Cancer Classification Using Multiparametric MRI
Aggressiveness of prostate cancer could be predicted using biparametric MRI and deep learning with negligible expert input, but performance was affected by scanner manufacturer and scanning protocol.
- 23 November 2024
- Published by Springer Nature in Virchows Archiv
- Vol. 487 (1) , 47-59
Clinical and histopathological features of immune checkpoint inhibitor-induced lung toxicity
A new era in cancer therapy emerged with the arrival of immune-checkpoint inhibitors (ICIs), followed by a new cadre of immune-related adverse events that affect up to 40% of patients. Literature on the pathological features associated with these events is still limited. Therefore, to expand our knowledge of the histopathologic spectrum of pulmonary changes, we conducted a case study series analysis on 16 non-neoplastic lung samples collected during or after ICI therapy. A set of predefined histological features related to the four different “compartments” (interstitium, pneumocyte, alveolar space, and bronchial mucosa), the CD4/CD8 T cell ratio, and the SP263 PD-L1 expression in the immune cells was assessed in three study categories [ICI + radiotherapy with/without chemotherapy (RT-based), ICI + chemotherapy (CT-based), ICI-based monotherapy (ICI-mono)]. Our results identified interstitial thickening, interstitial lymphocytic infiltrate, pneumocyte desquamation, intra-alveolar fibrin, or foamy macrophages in at least half of the cases in each of the study categories; all five features were present in 4 (RT-based), 3 (CT-based) and 1 (ICI-mono) patients. Hyaline membrane was a frequent finding in CT-based (80%) and ICI-mono (100%) compared to RT-based (44%) category. Moreover, CD4/CD8 ratio was ≤ 1 for almost all cases of the three study categories. Finally, a positive SP263 PD-L1 expression was identified in 50% or more of each study category. In conclusion, our results indicate that histopathologic findings in patients treated with ICI therapy are not diagnostic and varied. Additionally, these results are in line with recent studies showing an expansion on CD8+ T cell subset in patients under ICI treatment and highlight the synergism of polytherapy.
- 23 November 2024
- Published by Springer Nature in Virchows Archiv
- Vol. 487 (1) , 47-59
Clinical and histopathological features of immune checkpoint inhibitor-induced lung toxicity
A new era in cancer therapy emerged with the arrival of immune-checkpoint inhibitors (ICIs), followed by a new cadre of immune-related adverse events that affect up to 40% of patients. Literature on the pathological features associated with these events is still limited. Therefore, to expand our knowledge of the histopathologic spectrum of pulmonary changes, we conducted a case study series analysis on 16 non-neoplastic lung samples collected during or after ICI therapy. A set of predefined histological features related to the four different “compartments” (interstitium, pneumocyte, alveolar space, and bronchial mucosa), the CD4/CD8 T cell ratio, and the SP263 PD-L1 expression in the immune cells was assessed in three study categories [ICI + radiotherapy with/without chemotherapy (RT-based), ICI + chemotherapy (CT-based), ICI-based monotherapy (ICI-mono)]. Our results identified interstitial thickening, interstitial lymphocytic infiltrate, pneumocyte desquamation, intra-alveolar fibrin, or foamy macrophages in at least half of the cases in each of the study categories; all five features were present in 4 (RT-based), 3 (CT-based) and 1 (ICI-mono) patients. Hyaline membrane was a frequent finding in CT-based (80%) and ICI-mono (100%) compared to RT-based (44%) category. Moreover, CD4/CD8 ratio was ≤ 1 for almost all cases of the three study categories. Finally, a positive SP263 PD-L1 expression was identified in 50% or more of each study category. In conclusion, our results indicate that histopathologic findings in patients treated with ICI therapy are not diagnostic and varied. Additionally, these results are in line with recent studies showing an expansion on CD8+ T cell subset in patients under ICI treatment and highlight the synergism of polytherapy.