Edoardo Giacomello

Edoardo Giacomello, Michele Cataldo, Daniele Loiacono et al.

Deep Learning has established in the latest years as a successful approach to address a great variety of tasks. Healthcare is one of the most promising field of application for Deep Learning approaches since it would allow to help clinicians to analyze patient data and perform diagnoses. However, despite the vast amount of data collected every year in hospitals and other clinical institutes, privacy regulations on sensitive data - such as those related to health - pose a serious challenge to the application of these methods. In this work, we focus on strategies to cope with privacy issues when a consortium of healthcare institutions needs to train machine learning models for identifying a particular disease, comparing the performances of two recent distributed learning approaches - Federated Learning and Split Learning - on the task of Automated Chest X-Ray Diagnosis. In particular, in our analysis we investigated the impact of different data distributions in client data and the possible policies on the frequency of data exchange between the institutions.

Edoardo Giacomello, Pier Luca Lanzi, Daniele Loiacono et al.

Chest X-ray (CXR) is perhaps the most frequently-performed radiological investigation globally. In this work, we present and study several machine learning approaches to develop automated CXR diagnostic models. In particular, we trained several Convolutional Neural Networks (CNN) on the CheXpert dataset, a large collection of more than 200k CXR labeled images. Then, we used the trained CNNs to compute embeddings of the CXR images, in order to train two sets of tree-based classifiers from them. Finally, we described and compared three ensembling strategies to combine together the classifiers trained. Rather than expecting some performance-wise benefits, our goal in this work is showing that the above two methodologies, i.e., the extraction of image embeddings and models ensembling, can be effective and viable to solve tasks that require medical imaging understanding. Our results in that perspective are encouraging and worthy of further investigation.

Edoardo Giacomello, Daniele Loiacono, Luca Mainardi

Deep Learning is a promising approach to either automate or simplify several tasks in the healthcare domain. In this work, we introduce SegAN-CAT, an approach to brain tumor segmentation in Magnetic Resonance Images (MRI), based on Adversarial Networks. In particular, we extend SegAN, successfully applied to the same task in a previous work, in two respects: (i) we used a different model input and (ii) we employed a modified loss function to train the model. We tested our approach on two large datasets, made available by the Brain Tumor Image Segmentation Benchmark (BraTS). First, we trained and tested some segmentation models assuming the availability of all the major MRI contrast modalities, i.e., T1-weighted, T1 weighted contrast-enhanced, T2-weighted, and T2-FLAIR. However, as these four modalities are not always all available for each patient, we also trained and tested four segmentation models that take as input MRIs acquired only with a single contrast modality. Finally, we proposed to apply transfer learning across different contrast modalities to improve the performance of these single-modality models. Our results are promising and show that not SegAN-CAT is able to outperform SegAN when all the four modalities are available, but also that transfer learning can actually lead to better performances when only a single modality is available.

Edoardo Giacomello, Pier Luca Lanzi, Daniele Loiacono

We applied Generative Adversarial Networks (GANs) to learn a model of DOOM levels from human-designed content. Initially, we analysed the levels and extracted several topological features. Then, for each level, we extracted a set of images identifying the occupied area, the height map, the walls, and the position of game objects. We trained two GANs: one using plain level images, one using both the images and some of the features extracted during the preliminary analysis. We used the two networks to generate new levels and compared the results to assess whether the network trained using also the topological features could generate levels more similar to human-designed ones. Our results show that GANs can capture intrinsic structure of DOOM levels and appears to be a promising approach to level generation in first person shooter games.