This project aims to develop and clinically validate a real-time blood analysis and diagnostic platform using an integrated microfluidic medical device. The platform uses precision microfluidic technology to process and profile the biological activity of white blood cells, which is related to inflammatory conditions ranging from diabetes and sepsis to leukemia.
The Postdoctoral Research Associate - Data Scientist will lead the acquisition, preprocessing, analysis, and modelling of medical device-generated data, and will work closely with clinicians, scientists, and engineers to translate AI/machine learning models into working, deployable software for clinical testing and point-of-care use.
This role offers a rare opportunity to bridge clinical data science, medical device engineering, and translational research, moving from experimental and clinical sample data to robust AI-enabled diagnostic workflows and deployable prototypes.
The Postdoctoral Research Associate - Data Scientist is responsible for developing end-to-end data and AI/ML workflows for the medical diagnostic platform, from device data acquisition and signal/data preprocessing to model optimisation, validation, software integration, and deployment.
Responsibilities include:
- Work closely with clinicians, scientists, software engineers, mechanical/electrical engineers, and data scientists to define clinical use cases, data requirements, model outputs, and deployment workflows for the medical device platform.
- Design and implement reliable data acquisition pipelines from medical device hardware, sensors, imaging or signal outputs, and associated experimental or clinical metadata.
- Develop preprocessing, quality-control, annotation, feature-engineering, and data-management workflows for noisy, multimodal, longitudinal, or time-series biomedical data.
- Analyse experimental and clinical sample data to characterise device and biological performance, identify artefacts, benchmark platform performance, and support evidence generation for translational studies.
- Develop, train, validate, and optimise machine learning, deep learning, statistical, or hybrid models for diagnostic, predictive, or decision-support tasks using appropriate metrics, cross-validation, robustness checks, and uncertainty analyses.
- Translate research models into reproducible and deployable software components, including model inference pipelines, APIs, dashboards, reports, or lightweight user interfaces for clinical and laboratory testing.
- Support software-hardware integration for data acquisition, model inference, device feedback, and prototype deployment; assist with basic UX refinement based on clinician and end-user feedback.
- Maintain high-quality documentation of datasets, code repositories, model versions, experimental protocols, validation results, data dictionaries, and technical reports.
- Prepare manuscripts, presentations, grant reports, and invention disclosures; contribute to literature reviews, patent/IP searches, and responsible research dissemination where relevant.
- Support clinical study workflows by coordinating with clinical collaborators and applying good data governance practices for sensitive clinical and experimental data.
- Ph.D. in Data Science, Computer Science, Artificial Intelligence/Machine Learning, Biomedical Engineering, Electrical Engineering, Bioinformatics, Statistics, Applied Mathematics, Computational Biology, or a related field. Candidates close to completion may be considered if the degree will be conferred before appointment.
- Demonstrated research experience in data science, machine learning, deep learning, signal/image processing, biomedical data analysis, time-series analysis, clinical AI, or related computational research.
- Strong programming skills in Python; experience with common scientific and ML frameworks such as NumPy, pandas, scikit-learn, PyTorch or TensorFlow, Jupyter, and Git/GitHub.
- Experience with data acquisition, data cleaning, feature engineering, model optimisation, evaluation, reproducibility, version control, and scientific software development.
- Familiarity with clinical or biomedical datasets, medical device data, biosensor/microfluidic data, imaging data, physiological signals, or multimodal data integration is highly advantageous.
- Experience translating ML models into usable software prototypes, such as REST APIs, dashboards.
- Working knowledge of software-hardware interfacing, instrument-control software, serial/USB/Bluetooth communication, microcontrollers, embedded systems, or edge AI deployment would be advantageous.
- Ability to work closely with clinicians and multidisciplinary teams to interpret findings, gather end-user feedback, define clinical workflow needs, and improve the usability of AI/ML outputs.
- Strong written and oral communication skills, including the ability to write scientific manuscripts, technical documentation, reports, research proposals, and presentations for technical and non-technical audiences.
- Strong commitment to robust, ethical, and responsible AI/ML development, including attention to data privacy, model bias, interpretability, clinical validity, reproducibility, and safe deployment.
- Able and committed to work in Singapore, and willing to meet regularly with collaborators at NUS, NUH, and partnering hospitals.
Relevant Skills
Python, R/MATLAB, SQL, NumPy, pandas, scikit-learn, PyTorch/TensorFlow, Jupyter, data acquisition, biomedical signal processing, time-series analysis, image processing/computer vision, data preprocessing, feature engineering, model validation, model optimisation, uncertainty/robustness analysis, explainable AI, data visualisation, dashboards, REST APIs, Docker, Linux, Git/GitHub, MLOps, UX prototyping, software-hardware interfacing, and clinical data governance.
Interested applicants are invited to send in their full CV/resume, cover letter and list of three references (to include reference names and contact information). We regret that only shortlisted candidates will be notified.
