Hi there! I'm Gianluca, a passionate Research Scientist involved in conquering cancer through AI
Nov 2024 - Present
Berlin, Germany
In the Model-Centric AI research team of the Oncology group, I have been:
- π©» Developing AI models to advance the suit of LunitΒ΄s products and the SOTA in medical image analysis.
- π» Contributing to the internal research codebases with high-quality code and development standards.
- π€ Collaborating with research engineers and medical doctors for practical medical applications.
- π° Pushing Lunit's technological advancement and publishing in top-tier journals and conferences
Sep 2021 - Oct 2024
Pisa, Italy
In the Signals and Images Laboratory, I have:
- β« Proposed bias-mitigation frameworks for robust DL image classification in out-of-distribution settings under domain shift. Based on causal inference, feature disentanglement, contrastive learning, and injection of prior knowledge. Applied to large RWD CXR datasets.
- π Investigated techniques to discover causal disposition signals in images via Attention-inspired CNNs for cancer prediction from medical imaging (prostate MRI and HE digital pathology).
- π§ Proposed biologically inspired context-aware image classifiers akin to human vision.
- π€ Experienced with DDPMs (generative AI) to synthesize MRI images of prostate cancer.
- π Led a systematic literature review to study the intersection of causality and Explainable AI.
- π€± Explored the applicability of prototypical part learning in medical imaging by experimenting with ProtoPNet on a breast mass classification from mammogram images.
Apr - Jun 2023
Edinburgh, UK
At the VIOS Collaboratory led by Prof. Tsaftaris, I have:
- π Experimented with representation learning, causal reasoning in neural networks, and diffusion models.
βοΈ Collaborated in international research groups- π₯οΈ Group-coded with four other PhD students for a MICCAI 2023 challenge.
Pisa, Italy (Nov 2021 - Nov 2024)
Thesis: βHuman-aligned Deep Learning: Explainability, Causality, and Biological Inspiration.β
Pisa, Italy (Oct 2018 - Jul 2021)
Thesis: βStudy and development of advanced models integrating radiomic features and clinical data for outcome prediction in non-small cell lung cancer patients treated for brain metastases with stereotactic radiotherapy.β
Bologna, Italy (Sep 2015 - Oct 2018)
Thesis: βDevelopment of a Graphical User Interface in MATLAB for the visualization and spectral analysis of EEG and ECG signals.β
to do
-
[Project Name 1](Link to GitHub Repo)
Short description of the project and technologies used.
π Features: [Key Features]
π» Built with: [Tech Stack] -
[Project Name 2](Link to GitHub Repo)
Short description of the project and technologies used.
π Features: [Key Features]
π» Built with: [Tech Stack]
When Iβm not coding, Iβm into photography, cooking, rock music, and exploring the outdoors. Hereβs a peek at my stuff: