Department of Engineering / Profiles / Dr Pingfan Song

Department of Engineering

Dr Pingfan Song


Pingfan Song

Senior Research Associate in Trustworthy Machine Learning

Academic Division: Information Engineering

Research group: Computational and Biological Learning

Telephone: +44 1223 7 48520



Research interests

Dr Pingfan Song has been working on trustworthy machine learning which aims to advance work on key technical underpinnings of interpretability/transparency, fairness, and robustness of machine learning systems in order to push forward the next-generation AI. In addition to machine learning, he is also a professional in image processing, sparse modeling and sampling theory. His research has been applied to multi-disciplinary fields such as medical imaging, biological imaging, and other computational imaging tasks and inverse problems.

Research projects

  • Trustworthy Machine Learning

Machine learning systems are increasingly being deployed across society, in ways that affect many lives. Trustworthy machine learning is a key component of safe, ethical and responsible AI. This EPSRC-funded Turing AI Fellowship aims to advance work on key technical underpinnings of interpretability/transparency, fairness, and robustness of machine learning systems, and develop timely key applications in real world healthcare settings, such as medical imaging diagnosis systems.

  • Computational imaging for light-field microscopy

This BBSRC funded cross-disciplinary project aims to integrate fields of signal and image processing with neurophysiology and optical engineering to develop advanced high-speed light-field microscopy. We have developed a robust 3D neuron localization approach, a fast volumetric reconstruction approach, a subcellular resolution voltage imaging approach, and two interpretable deep learning approaches for 3D imaging.

  • Hybrid deep learning for Magnetic Resonance Fingerprinting (MRF)

A hybrid deep learning approach was proposed for accelerating magnetic resonance fingerprinting (MRF), a new type of quantitative magnetic resonance imaging (MRI) technology.

  • Machine learning for fast Magnetic Resonance Imaging (MRI)

A sparsity-driven learning algorithm was proposed to capture the dependency correlation between different MRI contrasts, such as T1/T2-weighted images to achieve fast and high-quality MRI reconstruction.

  • Transfer learning for multimodal image processing

A multimodal image processing framework was proposed to capture inherent correlations between heterogeneous image modalities such as RGB/multispectral/depth images. The proposed technology enables transferring appropriate high-resolution information from the guidance image to the target image of different modality without introducing notable artifacts.

  • Compressive Sensing with optimal bounds by incorporating side information

Side information was exploited to optimize measurement matrix design for Compressive Sensing systems, leading to optimal sampling performance. An advanced convex optimization tool -- Gaussian width was leveraged to establish tight theoretical CS bounds for sparse signal reconstruction.

  • Sub-Nyquist sampling system design based on Compressed Sensing

Based on Compressed Sensing theory, a Random Demodulation (RD) system was designed to achieve sub-Nyquist sampling for multi-tone (frequency-sparse) signals and successful reconstruction.

Other positions

  • STEM ambassador in UK.
  • Mentor of in2scienceUK, providing free teaching, career guidance and advice for A-level students.  in2scienceUK is an award winning charity which empowers students from disadvantaged backgrounds to achieve their potential and progress to STEM and research careers.
  • Co-president of London PhD Network (LPN), a non-profit academic association promoting interdisciplinary communication.


Pingfan Song received his B.S. and M.S degree both from Harbin Institute of Technology (HIT), China, and his Ph.D. degree from University College London (UCL), UK. He was a research associate at Imperial College London, and is now a senior research associate at University of Cambridge.

Department role and responsibilities