How to explain: from theory to practice

Today we live in an age where the internet and artificial intelligence allow us to search for information through impressive amounts of data, opening up revolutionary new ways to make sense of reality and understand our world. However, it is still an area of improvement to exploit the full potential of large amounts of explainable information by distilling it automatically in an intuitive and user-centred explanation. For instance, different people (or artificial agents) may search for and request different types of information in a different order, so it is unlikely that a short explanation can suffice for all needs in the most generic case. Moreover, dumping a large portion of explainable information in a one-size-fits-all representation may also be sub-optimal, as the needed information may be scarce and dispersed across hundreds of pages. The aim of this work is to investigate how to automatically generate (user-centred) explanations from heterogeneous and large collections of data, with a focus on the concept of explanation in a broad sense, as a critical artefact for intelligence, regardless of whether it is human or robotic. Our approach builds on and extends Achinstein’s philosophical theory of explanations, where explaining is an illocutionary (i.e., broad but relevant) act of usefully answering questions. Specifically, we provide the theoretical foundations of Explanatory Artificial Intelligence (YAI), formally defining a user-centred explanatory tool and the space of all possible explanations, or explanatory space, generated by it. We present empirical results in support of our theory, showcasing the implementation of YAI tools and strategies for assessing explainability. To justify and evaluate the proposed theories and models, we considered case studies at the intersection of artificial intelligence and law, particularly European legislation. Our tools helped produce better explanations of software documentation and legal texts for humans and complex regulations for reinforcement learning agents.

Adapting fruit production to climate change: how to increase intrinsic fruit quality and save water through deficit irrigation scheduling.

Fruit crops are an important resource for food security, since more than being nutrient they are also a source of natural antioxidant compounds, such as polyphenols and vitamins. However, fruit crops are also among the cultivations threatened by the harmful effects of climate change This study had the objective of investigating the physiological effects of deficit irrigation on apple (2020-2021), sour cherry (2020-2021-2022) and apricot (2021-2022) trees, with a special focus on fruit nutraceutical quality. On each trial, the main physiological parameters were monitored along the growing season: i) stem and leaf water potentials; ii) leaf gas exchanges; iii) fruit and shoot growth. At harvest, fruit quality was evaluated especially in terms of fruit size, flesh firmness and soluble solids content. Moreover, it was performed: i) total phenolic content determination; ii) anthocyanidin concentration evaluation; and iii) untargeted metabolomic study. Irrigation scheduling in apricot, apple and sour cherry is surely overestimated by the decision support system available in Emilia-Romagna region. The water stress imposed on different fruit crops, each during two years of study, showed as a general conclusion that the decrease in the irrigation water did not show a straightforward decrease in plant physiological performance. This can be due to the miscalculation of the real water needs of the considered fruit crops. For this reason, there is the need to improve this important tool for an appropriate water irrigation management. Furthermore, there is also the need to study the behaviour of fruit crops under more severe deficit irrigations. In fact, it is likely that the application of lower water amounts will enhance the synthesis of specialized metabolites, with positive repercussion on human health. These hypotheses must be verified.

Deep learning and spatial transcriptomics to investigate thyroid tumors

There are many diseases that affect the thyroid gland, and among them are carcinoma. Thyroid cancer is the most common endocrine neoplasm and the second most frequent cancer in the 0-49 age group. This thesis deals with two studies I conducted during my PhD. The first concerns the development of a Deep Learning model to be able to assist the pathologist in screening of thyroid cytology smears. This tool created in collaboration with Prof. Diciotti, affiliated with the DEI-UNIBO "Guglielmo Marconi" Department of Electrical Energy and Information Engineering, has an important clinical implication in that it allows patients to be stratified between those who should undergo surgery and those who should not. The second concerns the application of spatial transcriptomics on well-differentiated thyroid carcinomas to better understand their invasion mechanisms and thus to better comprehend which genes may be involved in the proliferation of these tumors. This project specifically was made possible through a fruitful collaboration with the Gustave Roussy Institute in Paris. Studying thyroid carcinoma deeply is essential to improve patient care, increase survival rates, and enhance the overall understanding of this prevalent cancer. It can lead to more effective prevention, early detection, and treatment strategies that benefit both patients and the healthcare system.

Developing a radiomic based machine learning model to identify patients with resected non-small-cell lung cancer at high risk of relapse

Background There is a wide variation of recurrence risk of Non-small-cell lung cancer (NSCLC) within the same Tumor Node Metastasis (TNM) stage, suggesting that other parameters are involved in determining this probability. Radiomics allows extraction of quantitative information from images that can be used for clinical purposes. The primary objective of this study is to develop a radiomic prognostic model that predicts a 3 year disease free-survival (DFS) of resected Early Stage (ES) NSCLC patients. Material and Methods 56 pre-surgery non contrast Computed Tomography (CT) scans were retrieved from the PACS of our institution and anonymized. Then they were automatically segmented with an open access deep learning pipeline and reviewed by an experienced radiologist to obtain 3D masks of the NSCLC. Images and masks underwent to resampling normalization and discretization. From the masks hundreds Radiomic Features (RF) were extracted using Py-Radiomics. Hence, RF were reduced to select the most representative features. The remaining RF were used in combination with Clinical parameters to build a DFS prediction model using Leave-one-out cross-validation (LOOCV) with Random Forest. Results and Conclusion A poor agreement between the radiologist and the automatic segmentation algorithm (DICE score of 0.37) was found. Therefore, another experienced radiologist manually segmented the lesions and only stable and reproducible RF were kept. 50 RF demonstrated a high correlation with the DFS but only one was confirmed when clinicopathological covariates were added: Busyness a Neighbouring Gray Tone Difference Matrix (HR 9.610). 16 clinical variables (which comprised TNM) were used to build the LOOCV model demonstrating a higher Area Under the Curve (AUC) when RF were included in the analysis (0.67 vs 0.60) but the difference was not statistically significant (p=0,5147).