AI Researcher (GEMINI Project)

The GEMINI Project

The GEMINI project, "Digital Twins by Generative Artificial Intelligence to Boost Personalized Medicine and Therapeutic Innovation", led by Humanitas Research Hospital, Humanitas AI Center, and Humanitas University, has been funded by the Ministry of University and Research under the FISA program.

Over the next five years, the goal is to develop a model based on innovative and scalable AI algorithms and real-world data (Digital Twin) that replicates the biological and clinical complexity of human diseases with a high social impact (such as cancers, debilitating chronic diseases, and rare diseases).

The ultimate aim is to provide physicians and patients with a validated and transparent tool to support critical aspects of clinical management, including diagnosis, prognosis, and optimal treatment selection.

Job Description

The AI Researcher has experience in machine learning and deep learning applied to structured and unstructured medical data.

The candidate will research and develop new statistical and machine-learning methods for the analysis of medical and clinical records, applying Artificial Intelligence (AI) techniques on real-world healthcare data.

The candidate will contribute to developing new technologies for data synthesis and digital twins using a wide variety of machine learning and deep learning methods; investigating various research topics in machine learning and statistics to determine the best method for medical data synthesis and effective approach for generated data validation.

Responsibilities and main activities

• Collaborate in research and development of innovative generative data models for effective synthetic data generation and digital twins in healthcare;
• Development of statistical, machine learning and deep learning models on medical data, including time-series/longitudinal data;
• Explore, define and support the clinical validation of the statistical and machine learning models applied to real-world data;
• Exploratory data analysis and integration of highly fragmented data;
• Visualize data, report effective results and derive useful knowledge using a data-driven approach;
• Collaborate with international partners in both private industry and academia;

Skills and qualifications

• Experience in developing machine learning and deep learning techniques and algorithms (such as k-NN, Naive Bayes, Support Vector Machines, Random Forests, etc) in healthcare, also applied to time-series/longitudinal data;
• Experience in developing generative models (e.g. statistical, GAN, VAE, etc.) applied to medical data for synthetic data generation and digital twins;
• Good knowledge of Computer Vision and/or NLP is appreciated;
• Experience in applied statistics skills, such as distributions, statistical testing, regression, etc;
• Good scripting and programming skills;
• Good proficiency in Python, R programming languages;
• Experience with data science frameworks (e.g. tensorflow, pytorch, scikit learn, scipy, pandas, numpy) and visualization frameworks (e.g. plotly, seaborn, matplotlib);
• Experience with cloud (GCP, AWS, Azure) and/or distributed computing is appreciated;
• Knowledge of MLOps practices, IT infrastructures, back end frontend development is appreciated;
• Master (PhD would be a plus) in a STEM discipline;
• Fluent in written and spoken English and Italian;

Soft Skills

• Excellent team-working capabilities even with colleagues from different research areas and backgrounds;
• Strong self-motivation, commitment and proactive approach;
• Ability to meet deadlines and work autonomously in rapidly changing environments;
• Curiosity and ability of stepping outside your comfort zone.

Contract and duration

We can offer a fixed-term contract. Contract duration, salary, as well as employment level, will be defined based on candidate's profile.

All candidate data collected from the application shall be processed in accordance with applicable law: Dlgs 198/2006 e dei Dlgs 215/2003 e 216/2003;  privacy ex artt. 13 e 14 del Reg. UE 2016/679.