Abstract
Lung cancer is one of the leading causes of cancer-related deaths worldwide, mainly due to late diagnosis. Screening programs can benefit from Computer-Aided Diagnosis (CAD) systems that detect and classify lung nodules using Computed Tomography (CT) scans. A great proportion of the literature proposes deep learning models based on single and private datasets with no evaluation of their generalisation capability. The main goal of this work is to study and address the lack of generalisation to out-of-domain data (source domain different from the target domain). In this work, we propose using a ResNet architecture with 2.5D inputs capable of maintaining the spatial information of the nodules (3 input channels based on the anatomical planes). Secondly, we apply domain-specific data augmentation tailored for CT scans. Combined with data augmentation, using 2.5D inputs achieves the best results, both in in-domain data (LIDC-IDRI: N=1377 nodules; and LNDb: N=183 nodules) and in out-of-domain data (LUNGx: N=73 nodules). In in-domain data, an Area Under the Curve (AUC) of 0.914 was achieved in the internal test set and 0.746 in one of the external test sets. Notably, in out-of-domain data, where the ground-truth labels have been confirmed by biopsy, whereas the training data only involved radiologist annotation regarding the likelihood of malignancy, AUC improves from 0.576 to 0.695, reaching a performance close to that of radiology experts. In the future, strategies should be applied to deal with the level of uncertainty of lung nodule annotations based solely on the observation of the CT scans.

Abstract

Abstract
Neuroblastoma (NB) is the most common extracranial tumor in pediatric cases. The MYCN oncogene amplification (MNA) is knowingly correlated with a poor prognosis, making detecting this biomarker crucial for treatment selection and survival prediction. The current clinical protocol for MNA detection includes invasive procedures, such as biopsy. The proposed work aims to develop non-invasive techniques for predicting MNA in patients with diagnosed NB, using AI-based models and Computerized Tomography (CT) scans. Machine learning methods that use the imaging features extracted from the tumor on the CT slices were developed and compared with deep learning (DL) models. Additionally, agnostic explainable methods for imaging were applied to create explanations about the relevant information used by the DL models in the prediction. The results show a better performance for the DL approach, which achieved an AUC of 0.94 ± 0.04. The similarity in the explanations produced by the models trained with different data splits showed that feature extraction remains somewhat invariant to shifts in training data, which is relevant given the small amount of data available. Learning models were shown to have predictive potential that, with further improvements, can be integrated into predictive, explainable, and, thus, trustworthy systems to aid clinicians in the decision-making process. © 2025 IEEE.

Abstract
Segmentation of the lungs in Computed Tomography (CT) is very challenging due to changes in lung shape, size, and parenchyma pattern, as well as differences in imaging acquisition protocols. As a consequence, these models may not be robust and may decrease their performance when deployed in a clinical setting. The Continual Learning paradigm holds great promise since learning models continually acquire incoming knowledge, having the ability to adapt to changing environments. In this work, experience replay with random sampling of past data was implemented, using the original CT images and the corresponding ground-truths. Data from four different institutions were used to develop the experiments, and the models were evaluated on a cross-cohort dataset. Using raw data, the goal was to study how the datasets and their imaging patterns were related and what impact the training datasets have on one another. The catastrophic forgetting effect diminished for almost all datasets. For two of the in-domain test datasets there was forward and backward transfer, results that could be linked to a possible similarity between them. A mean DSC of 0.94 was obtained across all datasets. The results showed how the similarity or disparity between data from different institutions can influence the performance of learning models. © 2025 IEEE.

Abstract
Algorithms based on computer vision are commonly used in pre-operative procedures to help health professionals detecting blood vessels, which is also the case with the Deep Inferior Epigastric Perforators (DIEPs). These blood vessels are essential to produce a viable autologous DIEP flap, and the analysis of characteristics such as their location, diameter and course is essential to ensure the success of surgeries. This analysis is made by a team of radiology technicians and then validated by a surgeon, making it a complex process that can take up to 2 hours. The proposed algorithm called Assisted Vascular Analysis (AVA) was developed to ensure a faster alternative to the conventional methods, using automation to identify structures of interest such as the skin, umbilicus and fascia, while also requiring minimum input from the user to segment each DIEP (2 points for the subcutaneous portion and 2 for the intramuscular portion). The AVA feasibility tests where conducted using 6 Computed Tomography Angiographies (CTAs), with a total of 28 DIEP calibers obtained during surgery (ground truths) from patients that underwent breast reconstruction with a DIEP flap. The algorithm was evaluated for its capability to segment the DIEPs and measure their caliber, comparing the results with the ground truth calibers and the manual mapping done by the radiology technicians. The Root Mean Square Error (RMSE) metric shows that the calibers obtained by the AVA algorithm (0.57 millimeters) and the radiology technicians (0.46 millimeters) are very similar, with the radiology technicians gaining a smaller edge of 0.11 millimeters. These results are very promising, since the errors are inferior to the average image resolution (0.88 millimeters). It was also demonstrated that the AVA algorithm is a faster alternative to manual segmentation, taking around 10 minutes to fully analyze each CTA, while the radiology technicians takes around 1 hour to do the DIEP mapping and caliber measurements. In conclusion, AVA is a validated algorithm to segment DIEP vessels and a faster alternative compared with conventional methods. © 2025 IEEE.

Abstract