PhD position in machine learning preterm lung health

University of Basel
· Department of Biomedical Engineering – Analytics & Informatics for Child Health (AICH)

Apply by 28 Nov 2026

University of Basel is advertising PhD position in machine learning for preterm lung health 100 in Basel, addressing the doctoral project uses machine learning with longitudinal measures of lung growth and pulmonary health in children born preterm, connecting respiratory outcomes with environmental exposure and cohort information.

The University of Basel's Analytics & Informatics for Child Health (AICH) group and University Children's Hospital Basel offer a four-year doctoral project on machine-learning prediction of long-term lung health after preterm birth. The Basel–Bern Infant Lung Development (BILD) cohort links infant pulmonary function, breath metabolomics, urine oxidative stress markers, medical history, high-resolution air pollution exposure and pulmonary follow-up into adolescence. The researcher will develop multimodal and longitudinal models for early-life risk stratification and disease phenotypes, seeking more informative markers than categorical bronchopulmonary dysplasia definitions.

External cohorts in Italy and Australia provide generalisability tests. Duties include interpretable ML, temporal analysis, missing-data handling, reproducible Python code, clinical collaboration, scientific publishing and conference presentations. The group offers HPC access and Graduate School Medical and Health Sciences training. Applicants need an MSc in a quantitative discipline, Python and ML framework proficiency, statistical modelling and excellent English. German is advantageous but not required.

Who This Opportunity Is Best Suited For:

Candidates comfortable with multi-modal clinical data, longitudinal prediction, uncertainty and translational biomedical questions are especially aligned, including those without a medical degree who want to work closely with clinicians.

Potential Challenges / Who May Find It Less Suitable:

Patient cohorts can be small, incomplete and heterogeneous; reproducibility and out-of-sample validation are not optional methodological details.

Experience with clinical or epidemiological datasets is advantageous.

Overall Opportunity Assessment:

This project is distinctive for its unusually rich birth-cohort measurements, independent international validation cohorts and direct connection to pediatric respiratory follow-up decisions.

machine learning longitudinal cohorts pulmonary function

Level PhD

Time Commitment Full-time

Duration 4 years

Organization Type University

MSc in Computer Science, Data Science, Statistics, Biomedical Engineering or related field;
Python, PyTorch/scikit-learn and statistical modelling; strong written/spoken English. German advantageous, not mandatory.

Technical Skills machine learning longitudinal cohorts pulmonary function