NUS Asst Prof Lei Li Models Heart Function with π Precision for MRI Digital Twins on Pi Day

Discovering the Mathematical Magic Behind NUS's Digital Heart Twins

On Pi Day, March 14—or 3.14 to math enthusiasts—the National University of Singapore (NUS) spotlighted a fascinating intersection of ancient mathematics and cutting-edge healthcare. Assistant Professor Lei Li from the Department of Biomedical Engineering at NUS College of Design and Engineering (CDE) leads the Digital Heart Lab, where the constant π plays a starring role in modeling heart function through magnetic resonance imaging (MRI). This research not only honors the timeless relevance of π but also advances personalized medicine in Singapore's thriving biomedical sector.

Lei Li's work transforms raw MRI data into dynamic digital twins—virtual replicas of individual hearts that simulate motion, blood flow, and electrical activity. These models promise to revolutionize cardiology by enabling doctors to predict disease progression and test treatments virtually, reducing risks for patients.

Singapore, with its world-class universities like NUS, is positioning itself as an Asia-Pacific hub for AI-driven health innovations. This research exemplifies how NUS faculty are bridging engineering, mathematics, and medicine to address global health challenges.

Who is Assistant Professor Lei Li?

Assistant Professor Lei Li (李雷) is a rising star in biomedical engineering. She joined NUS as an Assistant Professor in October 2024, following roles as a Lecturer at the University of Southampton and Postdoctoral Research Assistant at the University of Oxford's Institute of Biomedical Engineering. Holding a PhD from Shanghai Jiao Tong University (SJTU) in 2021, where she earned the Outstanding Doctoral Graduate Development Scholarship, Li has amassed over 3,695 citations on Google Scholar.

Her accolades include IEEE Transactions on Medical Imaging (TMI) Distinguished Reviewer awards, 2nd Prize of Shanghai Science and Technology Award (2024), and Rising Star of Women in Engineering at the Asian Deans’ Forum (2023). Li serves as Area Chair for MICCAI 2025 and Board Member of SIG-Cardiac and Women in MICCAI (WiM).

"Our goal is not for AI to replace mathematics," Li explains. "Our AI algorithms build on imaging theory and mathematical modelling to translate image data with interpretable markers of heart function." This philosophy drives her lab's mission at the forefront of NUS's biomedical research ecosystem.

• PhD in Biomedical Engineering, SJTU (2021)
• Postdoc, University of Oxford (2021-2023)
• Lecturer, University of Southampton (2024)
• Assistant Professor, NUS BME (2024-present)

The Digital Heart Lab: NUS's Hub for Cardiac Innovation

Established under Lei Li's leadership, the Digital Heart Lab (DHlab) at NUS pioneers AI-powered cardiac digital twins. The lab integrates multi-modal data—MRI, electrocardiograms (ECG), and more—to create patient-specific heart models. These virtual hearts replicate real-time dynamics, from contraction to arrhythmias, aiding in non-invasive diagnostics.

Key projects include solving the ECG inverse problem for tissue property inference and personalized topology-informed ECG electrode localization from incomplete MRIs. DHlab's workflow: acquire MRI/ECG data, segment anatomy, simulate electrophysiology/biomechanics, validate against clinical outcomes.

The lab recruits PhD students, postdocs, and research fellows, offering opportunities in AI for healthcare. Recent hires focus on cardiac arrhythmia treatment via digital twins. For aspiring researchers, check NUS research positions or postdoc openings.

π's Pivotal Role in MRI Reconstruction

MRI doesn't capture light like cameras; it measures electromagnetic signals from hydrogen atoms in a magnetic field. The Fourier transform—core to image reconstruction—involves π in equations linking frequency, phase, and spatial position. Specifically, π scales angular frequency in the inverse discrete Fourier transform (IDFT): the k-space to image conversion relies on e^(-i*2π*...) terms.

This ensures precise, artifact-free images vital for heart modeling. Li notes, "Higher resolution imaging and advanced AI depend on a reliable mathematical backbone. Constants like π represent stable scientific frameworks supporting new technologies." In cardiac twins, π recurs in cylindrical coordinates for blood flow and spherical harmonics for shape analysis.

Step-by-step MRI process:

• Apply magnetic field and radiofrequency pulses to align/excite protons.
• Detect gradient echoes; Fourier transform (with π) reconstructs spatial data.
• AI enhances: segment myocardium, quantify strain/ejection fraction.

Building Cardiac Digital Twins: A Step-by-Step Breakdown

Digital twins start with cine MRI sequences capturing heart cycles. Li's team:

1. Image Acquisition & Preprocessing: 4D MRI for anatomy/motion; denoise, register.
2. Segmentation: AI (e.g., AtrialJSQnet) delineates atria, scars, ventricles.
3. Geometry & Mesh: Finite element mesh; π in curvature computations.
4. Physics Simulation: Biomechanics (active strain), electrophysiology (bidomain model), hemodynamics (Navier-Stokes).
5. Personalization: Inverse problems calibrate parameters to match patient ECG/MRI.
6. Validation & Prediction: Simulate interventions; predict fibrosis, arrhythmias.

Publication: "Towards Enabling Cardiac Digital Twins of Myocardial Infarction" (IEEE TMI, 2024). For details, visit DHlab site.

Photo by Mia de Jesus on Unsplash

Transforming Cardiology: Real-World Applications

These twins enable:

• Risk stratification for atrial fibrillation (AFib), heart failure.
• Virtual ablation planning, device implantation.
• Drug response simulation, reducing trial-and-error.

In Singapore, where cardiovascular diseases claim 5,000 lives yearly, this aligns with MOH's precision medicine push. Li's models could cut readmissions by 20-30% via tailored therapies.

Explore careers in this field via faculty positions or clinical research roles at NUS.

Singapore's Biomedical Ecosystem and NUS Leadership

NUS BME ranks top globally; DHlab complements NTU's AI health initiatives, Duke-NUS clinical trials. Singapore's $25B biomed sector employs 80,000; digital twins boost R&D.

Govt funding via NRF, A*STAR supports. Li collaborates with Oxford, Southampton—global ties enhancing SG's rep. For advice, see academic CV tips.

Key Publications and Achievements

• "Personalized 4D Whole Heart Geometry Reconstruction from Cine MRI" (MedIA, 2025).
• "Solving the Inverse Problem of Electrocardiography for Cardiac Digital Twins" (IEEE Rev BME, 2024).
• MICCAI challenges: 1st in Dehazing Echo (2025).

Lab news: Featured NUS Pi Day; WHO GI-AI4H talks.

Career Opportunities in AI-Driven Biomedicine at NUS

DHlab seeks PhDs/postdocs in cardiac AI. Singapore's ecosystem offers higher ed jobs, from research assistants to faculty. Skills: Python, PyTorch, cardiac physiology. Rate profs at Rate My Professor; career advice here.

Future Horizons: Scaling Digital Twins for Global Impact

Li envisions population-scale twins for drug discovery, preventive care. Challenges: data privacy, computational cost—addressed via federated learning. SG's Smart Nation initiative accelerates translation.

Prospective students: apply via NUS grad portal; explore scholarships.

Photo by Hyundai Motor Group on Unsplash

Conclusion: Pioneering Hearts with Math and AI

Lei Li's Pi Day-highlighted research underscores NUS's math-AI-health fusion. As digital twins mature, they promise safer, precise cardiology. Join this revolution—check higher ed jobs, university jobs, rate professors, or career advice at AcademicJobs.com.