Gaetano Scebba

Gaetano Scebba

Data Scientist

Novartis

About me

I am Data Scientist in Translational Medicine at Novartis. My work focuses on improving the probability of success of drug development by using machine learning and heterogeneous health data, including continuous measurement from wearable devices, compound descriptors, clinical trial reports.

Prior to joining Novartis, I was a doctoral researcher in biomedical engineering at ETH Zurich, where I worked in close collaboration with students and medical researchers from Balgrist Clinic and University Hospital Zurich, on advancing physiological monitoring by utilizing machine learning and data from cameras and wearable devices. I hold a PhD in Biomedical Engineering (2021) from ETH Zurich, Switzerland, a MSc in Biomedical Engineering (2016) from Polytechnic University of Milan, Italy and a BSc in Biomedical Engineering (2013) from University of Pisa, Italy.

Research Interests

Biomedical Informatics, Digital Health, AI for Drug Development

News

I presented our work on Large Language Models at BioTechX conference (October 2023) - [link]
I held a tutorial on Foundations of ML for health data at Google Health GenAI hackathon (September 2023)
Our analysis on motor-cognitive dual tasking in Alzheimer's disease has been presented at the Alzheimer's Association International Conference (July 2023) - [link]
Our patent application "A system for recording a high-quality wound image and for real-time image quality assessment" was published at the European Patent Office (November 2022) - [link]
I have joined Novartis as Data Scientist in Translational Medicine (July 2022)
Our paper "Detect-and-Segment: a deep learning approach to automate wound image segmentation" has been accepted to Informatics in Medicine Unlocked (February 2022) - [link]

Projects

RADAR-AD

Remote Assessment of Disease and Relapse - Alzheimer’’s Disease

VitalCam

Multisensory Camera for Health Monitoring

WOU

An intelligent decision support system for wound healing

Publications

Motor-cognitive dual tasking in the clinical setting: a sensitive measure of functional impairment in early Alzheimer''s disease

This investigation is part of the ongoing Remote Assessment of Disease and Relapse – Alzheimer’’s Disease (RADAR-AD) study.

A system for recording a high-quality wound image and for real-time image quality assessment

We propose a system for recording supporting users to collect high quality wound images.

Detect-and-Segment: a Deep Learning Approach to Automate Wound Image Segmentation

We propose a deep learning approach based on chaining of object detection and segmentation architectures to automate wound image segmentation and improve its robustness.

Wound Image Quality From a Mobile Health Tool for Home-Based Chronic Wound Management With Real-Time Quality Feedback Randomized Feasibility Study

We propose a mHealth tool for the remote self-assessment of digital ulcers in patients with systemic sclerosis.