United Kingdom
I am a machine learning scientist with 5+ years of experience developing and deploying machine learning solutions for biomedical applications. My expertise spans computer vision, biomedical image analysis, and neuroscience, underpinned by a strong understanding of clinical research and trials. I thrive in highly interdisciplinary environments and collaborated closely with clinicians, computer scientists, epidemiologists, and physicists in my PhD to translate cutting-edge ML research into clinically meaningful solutions. Beyond my research, I bring strong communication, leadership, and organisational skills. I have organised and hosted multiple scientific and career-focused events through the BBMS Society at Cambridge, engaging academic researchers (including Nobel laureates), biotech founders, venture capital investors, consultants, and international policymakers in the healthcare ecosystem. I excel in roles that require both technical depth and the ability to communicate complex ideas clearly to diverse audiences.
• Thesis title: Applications of Machine Learning to Neuroimage Analysis in Cerebral Small Vessel Disease. (Link below) • Developed, validated and deployed LUMEN, the first open-source deep learning pipeline for 3D segmentation and morphology quantification of cerebral lenticulostriate arteries (LSAs) from 7T TOF MRI. Reduced analysis time by 15× from 10 hrs (manual 3D segmentation) to 40 min per subject, enabling the 3D morphology analysis of LSAs in clinical research into cerebrovascular disease pathology. (Link to repo below) • Built multimodal deep learning survival models for dementia prediction in cerebral small vessel disease using multiple modalities of MRI images and clinical data. Investigated different modality fusion strategies and embedding extraction using foundation diffusion autoencoders pretrained on UK Biobank. Repo: https://github.com/RuiLiGitLove/SVDementia-Public. • Co-authored 5+ peer-reviewed journal and conference articles. Full list available on Google Scholar (link below). • Presented at multiple national and international conferences, including ISMRM 2024, VasCog 2025&2023. Received 4 awards for best poster presentation. • Awarded the Alan Turing Institute Enrichment Scheme Award (£1,500) in 2022. • Fully funded by Trinity College.
• Supervised a 3rd-year engineering course on Inference. • Covered various topics in machine learning, including Bayesian regression, classification, unsupervised clustering, dimensionality reduction, sequence modelling.
• Led the student society committee in organising multiple scientific and career-focused events, including the flagship BBMS Annual Conference (150+ attendees), career roundtables, society formal dinners, and various networking events. • Engaged with leading biomedical science researchers (including Nobel laureates), biotech founders, venture capital investors, consultants, and international policymakers in the healthcare ecosystem. • Initiated and facilitated the establishment of a strategic partnership between BBMS and Roche, resulting in funding, on-campus events, and company visits. • Jointly raised £10,000+ funding for the society in one year.
• Developed a deep learning pipeline (TensorFlow Keras) to classify fluorescent E. coli barcodes from fluorescence microscopy images, achieving 0.85-0.99 test accuracy. • Used experimental data generated from the lab. Collaborated with PhD wet-lab scientists to understand the data generation and biology of E. coli barcoding. • Gained extensive working experience using TensorFlow Keras and scikit-image in Python. • Awarded the Part IIB (Master’s) Project Prize for best Master's research projects.
• Trained a deep learning algorithm to identify circadian patterns in proteomics data, incorporating protein-protein interaction data from STRING database to aid the finding of joint periodicity. • Modelled protein expression trajectories using Gaussian processes.
• Selected as one of 65 scholars for the fully funded Amgen Scholars Europe Program in bioscience research. • Worked with an optogenetic platform, Cyberloop, for real-time, single-cell stimulation and observation. • Developed an image calibration pipeline using MATLAB to correct the image distortion in the projection system for this optogenetic platform, achieving 100% cell-targeting accuracy. • Presented a scientific poster at the 2019 Amgen Scholars European Symposium in Cambridge, UK.