Abstract
Neuroblastoma (NB) is the most common extracranial solid tumor in childhood. Genomic amplification of MYCN is associated with poor outcomes and is detected in 16% of all NB cases. CT scans and MRI are the imaging techniques recommended for diagnosis and disease staging. The assessment of imaging features such as tumor volume, shape, and local extension represent relevant prognostic information. Radiogenomics have shown powerful results in the assessment of the genotype based on imaging findings automatically extracted from medical images. In this work, random forest was used to classify the MYCN amplification using radiomic features extracted from CT slices in a population of 46 NB patients. The learning model showed an area under the curve (AUC) of 0.85 ± 0.13, suggesting that radiomic-based methodologies might be helpful in the extraction of information that is not accessible by human naked eyes but could aid the clinicians on the diagnosis and treatment plan definition. Clinical relevance - This approach represents a random forest-based model to predict the MYCN amplification in NB patients that could give a faster, earlier, and repeatable analysis of the tumor along the time.

Abstract
Deep Learning (DL) based classification algorithms have been shown to achieve top results in clinical diagnosis, namely with lung cancer datasets. However, the complexity and opaqueness of the models together with the still scant training datasets call for the development of explainable modeling methods enabling the interpretation of the results. To this end, in this paper we propose a novel interpretability approach and demonstrate how it can be used on a malignancy lung cancer DL classifier to assess its stability and congruence even when fed a low amount of image samples. Additionally, by disclosing the regions of the medical images most relevant to the resulting classification the approach provides important insights to the correspondent clinical meaning apprehended by the algorithm. Explanations of the results provided by ten different models against the same test sample are compared. These attest the stability of the approach and the algorithm focus on the same image regions.

Abstract
The use of deep learning methods in medical imaging has been able to deliver promising results; however, the success of such models highly relies on large, properly annotated datasets. The annotation of medical images is a laborious, expensive, and time-consuming process. This difficulty is increased for the mutations status label since these require additional exams (usually biopsies) to be obtained. On the other hand, raw images, without annotations, are extensively collected as part of the clinical routine. This work investigated methods that could mitigate the labelled data scarcity problem by using both labelled and unlabelled data to improve the efficiency of predictive models. A semi-supervised learning (SSL) approach was developed to predict epidermal growth factor receptor (EGFR) mutation status in lung cancer in a less invasive manner using 3D CT scans.The proposed approach consists of combining a variational autoencoder (VAE) and exploiting the power of adversarial training, intending that the features extracted from unlabelled data to discriminate images can help in the classification task. To incorporate labelled and unlabelled images, adversarial training was used, extending a traditional variational autoencoder. With the developed method, a mean AUC of 0.701 was achieved with the best-performing model, with only 14% of the training data being labelled. This SSL approach improved the discrimination ability by nearly 7 percentage points over a fully supervised model developed with the same amount of labelled data, confirming the advantage of using such methods when few annotated examples are available.

Abstract
In the healthcare domain, datasets are often private and lack large amounts of samples, making it difficult to cope with the inherent patient data heterogeneity. As an attempt to mitigate data scarcity, generative models are being used due to their ability to produce new data, using a dataset as a reference. However, synthesis studies often rely on a 2D representation of data, a seriously limited form of information when it comes to lung computed tomography scans where, for example, pathologies like nodules can manifest anywhere in the organ. Here, we develop a 3D Progressive Growing Generative Adversarial Network capable of generating thoracic CT volumes at a resolution of 1283, and analyze the model outputs through a quantitative metric (3D Muli-Scale Structural Similarity) and a Visual Turing Test. Clinical relevance - This paper is a novel application of the 3D PGGAN model to synthesize CT lung scans. This preliminary study focuses on synthesizing the entire volume of the lung rather than just the lung nodules. The synthesized data represent an attempt to mitigate data scarcity which is one of the major limitations to create learning models with good generalization in healthcare.

Abstract

Abstract
Lung cancer is the leading cause of cancer death worldwide. Early low-dose computed tomography (CT) screening can decrease its mortality rate and computer-aided diagnoses systems may make these screenings more accessible. Radiomic features and supervised machine learning have traditionally been employed in these systems. Contrary to supervised methods, unsupervised learning techniques do not require large amounts of annotated data which are labor-intensive to gather and long training times. Therefore, recent approaches have used unsupervised methods, such as clustering, to improve the performance of supervised models. However, an analysis of purely unsupervised methods for malignancy prediction of lung nodules from CT images has not been performed. This work studies nodule malignancy in the LIDC-IDRI image collection of chest CT scans using established radiomic features and unsupervised learning methods based on k-Means, Spectral Clustering, and Gaussian Mixture clustering. All tested methods resulted in clusters of high homogeneity malignancy. Results suggest convex feature distributions and well-separated feature subspaces associated with different diagnoses. Furthermore, diagnosis uncertainty may be explained by common characteristics captured by radiomic features. The k-Means and Gaussian Mixture models are able to generalize to unseen data, achieving a balanced accuracy of 87.23% and 86.96% when inference was tested. These results motivate the usage of unsupervised approaches for malignancy prediction of lung nodules, such as cluster-then-label models. Clinical Relevance - Unsupervised clustering of radiomic features of lung nodules in chest CT scans can differentiate between malignant and benign cases and reflects experts' diagnosis uncertainty