Postdoc / Research Fellow / PhD candidate (Acoustic Heart Failure Monitoring: Deep Learning)

About TUMCREATE

TUMCREATE is a multidisciplinary research platform of the Technical University of Munich (TUM) at the Singapore Campus for Research Excellence and Technological Enterprise (CREATE). We are partnering with universities, public agencies, and industry to advance future technologies.

The HEARTwise project is a public-private research initiative funded by Singapore's National Research Foundation, bringing together clinical cardiologists (CADENCE / NUHS, NHCS), engineers from TUM and TUMCREATE, and MedTech partners. The project develops a multimodal platform for non-invasive, early decompensation detection in ambulant HF patients - combining wearable mobility sensing, heart and lung sound recognition, and integration with additional digital biomarker sources.

As a member of the HEARTwise team, you will collaborate within an interdisciplinary consortium comprising leading experts in cardiology, telemedicine, biomedical engineering, acoustics, data privacy, and industry.

Please visit www.tum-create.edu.sg for more information about TUMCREATE.

Background & Scientific Context

Heart failure (HF) affects 2.5% of Singapore's population and is the leading cardiac cause of hospital admission - presenting a decade earlier and more aggressively than in Europe or North America. The 30% rehospitalization rate within 90 days of discharge is both a major clinical burden and a window of opportunity for early, data-driven intervention.

Job Summary

We are seeking a highly motivated and skilled researcher with proven expertise in machine learning for audio analysis. Within HEARTwise at TUMCREATE, you will lead the development of artificial intelligence algorithms for audio classification of heart and lung sounds. Through multimodal integration with data from mobility sensing and other biomarkers, you will create a reliable tool for early decompensation detection in ambulant HF patients. You will be working in a team developing a tool for audio data collection in the clinic, running a clinical study for data collection, working with clinicians to create an annotated database, and validating your classification prototype in a pilot clinical diagnostic study.

This is a fixed-term contract until March 2029 when applicable.

Key Responsibilities

• Develop prototype algorithms for audio classification of heart and lung sounds.
• Contribute to clinical data collection, database building, multimodal data analysis,clinical prototype app development.
• Verify the technology against requirements and validate in a pilot clinical study.
• Collaborate with experts in cardiology, telemedicine, biomedical engineering, acoustics, and data privacy
• Write scientific manuscripts, contribute to project reports, and present findings at international conferences and stakeholder meetings.

Key Competencies

• At least 1 year of proven expertise in deep learning signal analysis,preferably with audio data.
• Excellent written and verbal communication skills in English, with the ability to communicate complex analytical results to multidisciplinary teams.
• Demonstrated ability to work independently and collaboratively in a fast-paced, international research environment.

Qualification Requirements

• We are seeking a researcher with a PhD degree in electrical engineering,physics, data science, computer science, or similar, and a proven track record in machine learning and (audio) signal processing. Fresh graduates and PhD candidates will also be considered.
• Proficiency in Python/Matlab and anaconda (or similar).
• Experience in C/C++ and database interfacing and in user or clinical studies is desirable.
• Ideally, the candidate has an interdisciplinary education including acoustics that facilitates collaboration with biomedical engineers.

What we offer

• A collaborative and inclusive world-class research environment.
• Opportunities for professional development.
• Engagement with a leading international team addressing interdisciplinary challenges in heart failure management.

Applications

Please submit the following compulsory documents to [Confidential Information]:

• Cover letter
• CV
• University transcripts

Only shortlisted candidates will be contacted.

We look forward to your application!