Design of experiment in drug development: optimal allocations in multi-arm clinical trials and design space for multi-step processes

In this thesis, we deal with the design of experiments in the drug development process, focusing on the design of clinical trials for treatment comparisons (Part I) and the design of preclinical laboratory experiments for proteins development and manufacturing (Part II). In Part I we propose a multi-purpose design methodology for sequential clinical trials. We derived optimal allocations of patients to treatments for testing the efficacy of several experimental groups by also taking into account ethical considerations. We first consider exponential responses for survival trials and we then present a unified framework for heteroscedastic experimental groups that encompasses the general ANOVA set-up. The very good performance of the suggested optimal allocations, in terms of both inferential and ethical characteristics, are illustrated analytically and through several numerical examples, also performing comparisons with other designs proposed in the literature. Part II concerns the planning of experiments for processes composed of multiple steps in the context of preclinical drug development and manufacturing. Following the Quality by Design paradigm, the objective of the multi-step design strategy is the definition of the manufacturing design space of the whole process and, as we consider the interactions among the subsequent steps, our proposal ensures the quality and the safety of the final product, by enabling more flexibility and process robustness in the manufacturing.

Mind the gaps: cognitive-inspired AI for high-level visual sensemaking. Towards abstract concept image classification

The abundance of visual data and the push for robust AI are driving the need for automated visual sensemaking. Computer Vision (CV) faces growing demand for models that can discern not only what images "represent," but also what they "evoke." This is a demand for tools mimicking human perception at a high semantic level, categorizing images based on concepts like freedom, danger, or safety. However, automating this process is challenging due to entropy, scarcity, subjectivity, and ethical considerations. These challenges not only impact performance but also underscore the critical need for interoperability. This dissertation focuses on abstract concept-based (AC) image classification, guided by three technical principles: situated grounding, performance enhancement, and interpretability. We introduce ART-stract, a novel dataset of cultural images annotated with ACs, serving as the foundation for a series of experiments across four key domains: assessing the effectiveness of the end-to-end DL paradigm, exploring cognitive-inspired semantic intermediaries, incorporating cultural and commonsense aspects, and neuro-symbolic integration of sensory-perceptual data with cognitive-based knowledge. Our results demonstrate that integrating CV approaches with semantic technologies yields methods that surpass the current state of the art in AC image classification, outperforming the end-to-end deep vision paradigm. The results emphasize the role semantic technologies can play in developing both effective and interpretable systems, through the capturing, situating, and reasoning over knowledge related to visual data. Furthermore, this dissertation explores the complex interplay between technical and socio-technical factors. By merging technical expertise with an understanding of human and societal aspects, we advocate for responsible labeling and training practices in visual media. These insights and techniques not only advance efforts in CV and explainable artificial intelligence but also propel us toward an era of AI development that harmonizes technical prowess with deep awareness of its human and societal implications.