Francesco Ponzio

Francesco Ponzio
Politecnico di Torino | polito · DIST - Interuniversity Department of Regional and Urban Studies and Planning

Biomedical Engineering

About

14
Publications
4,793
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
76
Citations
Citations since 2016
13 Research Items
74 Citations
20162017201820192020202120220510152025
20162017201820192020202120220510152025
20162017201820192020202120220510152025
20162017201820192020202120220510152025
Introduction
The recent prominent role of deep learning (DL) in computer science has strongly influenced my research. My best research efforts have been tailored to face three important reasons that have hampered the large-scale adoption of DL solutions: (i) the dependence of DL on huge annotated training data; (ii) the absence of a principled uncertainty measure associated with DL’s predictions; (iii) the needing of DL-based solutions able to evolve as new information emerges.

Publications

Publications (14)
Article
Full-text available
Ideally, Convolutional Neural Networks (CNNs) should be trained with high quality images with minimum noise and correct ground truth labels. Nonetheless, in many real-world scenarios, such high quality is very hard to obtain, and datasets may be affected by any sort of image degradation and mislabelling issues. This negatively impacts the performan...
Article
Full-text available
Chiari I malformation is characterized by the herniation of cerebellar tonsils below the foramen magnum. It is often accompanied by syringomyelia and neurosurgical management is still controversial. In fact, it is frequent that some symptomatic patients initially undergo bony decompression of the posterior fossa and need in a short time more invasi...
Article
Full-text available
Motivation Computer-aided analysis of biological images typically requires extensive training on large-scale annotated datasets, which is not viable in many situations. In this paper, we present Generative Adversarial Network Discriminator Learner (GAN-DL), a novel self-supervised learning paradigm based on the StyleGAN2 architecture, which we empl...
Preprint
Full-text available
Computer-aided analysis of biological images typically requires extensive training on large-scale annotated datasets, which is not viable in many situations. In this paper we present GAN-DL, a Discriminator Learner based on the StyleGAN2 architecture, which we employ for self-supervised image representation learning in the case of fluorescent biolo...
Preprint
Full-text available
Convolutional Neural Networks (CNNs) are supposed to be fed with only high-quality annotated datasets. Nonetheless, in many real-world scenarios, such high quality is very hard to obtain, and datasets may be affected by any sort of image degradation and mislabelling issues. This negatively impacts the performance of standard CNNs, both during the t...
Chapter
Full-text available
Traditionally, the analysis of histological samples is visually performed by a pathologist, who inspects under the microscope the tissue samples, looking for malignancies and anomalies. This visual assessment is both time consuming and highly unreliable due to the subjectivity of the evaluation. Hence, there are growing efforts towards the automati...
Conference Paper
With the advent of digital pathology, there has been an increasing interest in providing pathologists with machine learning tools, often based on deep learning, to obtain faster and more robust image assessment. Nonetheless, the accuracy of these tools relies on the generation of large training sets of pre-labeled images. This is typically a challe...
Chapter
The early diagnosis of colorectal cancer (CRC) traditionally leverages upon the microscopic examination of histological slides by experienced pathologists, which is very time-consuming and rises many issues about the reliability of the results. In this paper we propose using Convolutional Neural Networks (CNNs), a class of deep networks that are su...
Article
Full-text available
Thanks to their capability to learn generalizable descriptors directly from images, deep Convolutional Neural Networks (CNNs) seem the ideal solution to most pattern recognition problems. On the other hand, to learn the image representation, CNNs need huge sets of annotated samples that are unfeasible in many every-day scenarios. This is the case,...
Article
Full-text available
Objectives: Changes in pupil size are governed by the autonomic nervous system but may be systematically driven by voluntary shifting the gaze in depth. Thus, the pupil accommodative response (PAR) that accompanies voluntary gaze shifts from a far to a near target might be exploited as a simple human-computer interface (HCI), bypassing the somato-m...

Network

Cited By