A Fully Funded PhD in machine learning and large language models (LLMs)

Social network you want to login/join with:

A Fully Funded PhD in machine learning and large language models (LLMs), Lausanne

Client:

Location:

Job Category:

Other

Job Reference:

369d07e3e5a8

Job Views:

2

Posted:

17.04.2025

Expiry Date:

01.06.2025

Job Description:

Contexte

The Lausanne University Hospital (CHUV) is one of five Swiss university hospitals. Through its collaboration with the Faculty of Biology and Medicine of the University of Lausanne and the EPFL, CHUV plays a leading role in the areas of medical care, medical research and training.

Dre Nazanin Sédille and Professor Oliver Y. Chén’s teams develop new machine-learning and statistical methods and study large-scale data in health and disease. Our data are from diverse sources from the Service of Clinical Chemistry. Their focus is threefold: (a) Building new, methodologically exciting models to address real-world problems; (b) using these methods to (i) study the interplays between large-scale multimodal, multivariate, high-dimensional features, and when/how they may be associated with diseases cross-sectionally and longitudinally and (ii) identify markers that support patient diagnosis and prognosis; (c) translating our algorithms into clinical decision support and patient health management apps.

Mission

The PhD student will primarily work on three interlinked projects in collaboration between CHUV and UNIL on data related to cardiovascular and metabolomic diseases.

1. Building better biomarkers for cardiovascular and metabolomic diseases: Large language models (LLMs) designs oftentimes rely on machine learning and biology in relative isolation. Here, we aim to design LLMs to discover biomarkers for cardiovascular and metabolomic diseases using insights from both machine intelligence and biology. Using these new methodological frameworks, we aim to identify cardiovascular biomarkers (e.g., ECG and echocardiogram) related to heart disease risk factors and clinical outcomes such as disease onset or progression. In parallel, we aim to identify diabetic biomarkers (e.g., glucose metabolomic) related to, for example, renal insufficiency, and clinical outcomes such as diabetic onset or progression. Finally, we aim to quantify the disease pathways from risk factors to clinical outcomes via the discovered biomarkers.
2. Disease prediction using multimodal data: Large language models (LLMs) for disease prediction using single-modal data may overlook the comprehensive disease landscapes underpinned by multiple data sources. Here, we will design LLMs to identify, from multivariate, multimodal, potentially high-dimensional biomarkers, that can together improve the overall disease prediction accuracy as well as clinical explanation of the biomarkers.
3. Longitudinal data analysis and early disease prediction: Here, knitting expertise from machine learning and clinical chemistry, we aim to develop new methods that can unveil the longitudinal trajectories of the disease profile, forecast future disease progression, and inform targeted and more timely disease management, treatment and prevention.

Profil

• A master’s degree and an undergraduate degree in disciplines relevant to (applied) mathematics, computer science, engineering, machine learning, or statistics.
• An interest in developing new methods and applications and employing them to address real-world healthcare-related problems.
• A team player.
• The working language of the group is English. A good command of the French language is mandatory.
• Strong programming skills related to machine learning, longitudinal methods, and large language models (LLMs).
• Experience in machine learning (LLMs), statistical modelling, and version control.

Nous offrons

• Joint affiliations with the Lausanne University Hospital (CHUV) and the University of Lausanne.
• An interdisciplinary environment, and a supportive team. We strive for equality, diversity, and inclusion. Our team is interdisciplinary and multicultural, and we encourage underrepresented students to apply.
• Possibility to collaborate with international universities.
• Access to courses from the CHUV and the University of Lausanne.
• Possibility to access one of the furnished apartments offered in the surrounding neighborhoods in case of relocation in Switzerland.
• Discounts proposed on social and cultural events, thanks to the “H-Oxygène” association.
• Signing up to our Mobility Plan and benefit from different advantages (discounts on public transportation, promotion of “Mobility” car fleet and discounts on electric bikes).
• Being able to enjoy our high-quality corporate restaurants, located in every hospital building, with employees’ discount.