I am an AI researcher and PhD candidate in Electrical Engineering at Harvard University, based at Massachusetts General Hospital. I work with Prof. Na Li and Prof. Quanzheng Li.
My research develops efficient generative and multimodal models for high-dimensional biomedical data — with a focus on 3D medical image synthesis, multimodal representation learning across electrophysiology and molecular measurements, and closed-loop AI systems for biological discovery.
Before Harvard, I received BS and MS degrees in Electrical and Computer Engineering from Carnegie Mellon University, and worked as a software engineer at Google, Meta, and Duolingo.
Seeking AI Research Scientist / Research Scientist / postdoctoral roles starting 2027, focused on generative AI, multimodal learning, computer vision, and AI for scientific discovery.
News
Feb 2026 Our Science paper on cyborg pancreatic islets was featured by EurekAlert!, GEN, and BioTechniques. It was exciting to see this work reach a broader audience and highlight how flexible bioelectronics and spatial transcriptomics can be combined to study the maturation of transplanted human islet cells.
Dec 2025 Our Science paper on flexible nanoelectronics for transplanted cardiomyocytes was featured by Medical Xpress. This work combined bioelectronics, machine learning, and spatial transcriptomics to investigate how transplanted heart cells integrate with the host and give rise to arrhythmias.
Apr 2025 I received Third Place for Poster Presentation at the Eric and Wendy Schmidt Center Biomedical Science & AI Symposium. The awarded work presented a closed-loop AI framework that combines Gaussian Process Bayesian optimization with flexible bioelectronics to automatically optimize electrical stimulation protocols for cardiac organoid maturation, reducing experimental cost while accelerating tissue development.
Nov 2024 I was selected as an Early-Career Scholar Honoree and received Best Poster Award Runner-Up at the NIH BRAIN NeuroAI Workshop. The awarded work introduced a hierarchical generative model that jointly captures short-term stimulus-evoked neural dynamics and long-term representational drift from large-scale single-neuron recordings, with applications to NeuroAI and brain-computer interfaces.
Selected Research
I am broadly interested in AI methods for scientific discovery — efficient generative models, multimodal learning, and closed-loop optimization systems for biomedical data.
A framework for efficient 3D medical image synthesis that learns inter-slice feature trajectories from 2D generators, achieving ~135× lower inference cost than comparable 3D methods. Evaluated on BraTS 2023 and SynthRAD2023 across unconditional generation, missing-modality MR synthesis, and MR-to-CT translation.
A closed-loop AI platform integrating flexible bioelectronics with Gaussian Process Bayesian optimization to adaptively control electrical stimulation of cardiac organoids. Identifies optimized stimulation conditions that accelerate functional maturation, achieving ~70% higher conduction velocity than fixed-parameter baselines.
Tissue-like stretchable electronics implanted during organogenesis of human stem cell-derived pancreatic islets enable months-long, single-cell-resolved electrophysiology. Reveals maturation-dependent electrical patterns of α- and β-like cells and identifies two major electrical states distinguished by glucose threshold for action potential firing.
Ultra-flexible nanoelectronic meshes implanted alongside transplanted human iPSC-derived cardiomyocytes reveal arrhythmogenic electrical dynamics at single-cell resolution over weeks in vivo. Identifies how coinjection with RADA16 promotes sarcomere organization and mitigates arrhythmogenesis.
Uses the CARLA simulator to generate training data for rare road object detection, improving sim-to-real transfer for autonomous driving perception.
Publications & Preprints
LiFT: Lifted Inter-slice Feature Trajectories for 3D Image Generation from 2D Generators X. Zhang, Y. Zhang, P. Jin, A. Marin-Llobet, N. Li, Q. Li arXiv, 2026 Introduces a lightweight 2D-to-3D generative framework that learns inter-slice feature trajectories to produce coherent high-resolution 3D medical images at much lower inference cost than fully volumetric models. Computer VisionGenerative AI
Benchmarking spike source localization algorithms in high density probes
H. Zhao‡, X. Zhang‡, A. Marin-Llobet, X. Lin, J. Liu PLOS Computational Biology, 2026 Provides a systematic benchmark of spike source localization algorithms using simulated and experimental ground-truth datasets, identifying tradeoffs in accuracy, robustness, and runtime for long-term neural recordings. Computational Biology
Implanted flexible electronics reveal principles of human islet cell electrical maturation
Q. Li†, R. Liu†, Z. Lin†, X. Zhang†, W. Wang, I. M. Galicia-Silva, M. Liu, Z. Gao, S. D. Pollock, J. R. Alvarez-Dominguez‡, J. Liu‡ Science, 2026 Uses implanted stretchable electronics to longitudinally track single-cell electrical activity in human pancreatic organoids, revealing maturation trajectories of stem-cell-derived α and β cells. MultimodalComputational Biology
Growth-adaptive spring electronics for long-term, same-neuron mapping in the developing rat brain
A. J. Lee†, H. Sheng†, A. Marin-Llobet, Z. Wang, J. Lee, R. Liu, X. Zhang, E. Hsiao, J. Baek, A. Aljovic, D. Liu, Y. He, N. Lu, J. Liu‡ bioRxiv, 2026 Develops growth-adaptive spring electronics that maintain stable neural interfaces during brain growth, enabling weeks-long tracking of the same neurons across postnatal development. Computational Biology
Flexible nanoelectronics reveal arrhythmogenesis in transplanted human cardiomyocytes
J. Aoyama†, R. Liu†, X. Zhang†, A. Y. Zhu, P. Luanpaisanon, N. Velayutham, J. C. Garbern, F. Cao, I. Barrera, H. Fandl, M. Sokol, S. Dasariraju, E. S. Gil, E. Aleksi, T. Amanuma, J. J. Saucerman, F. Chen, J. Liu‡, R. T. Lee‡ Science, 2025 Uses flexible nanoelectronics to monitor transplanted human iPSC-derived cardiomyocytes and identify electrophysiological mechanisms underlying graft-induced arrhythmogenesis. MultimodalComputational Biology
Riemannian geometry for the classification of brain states with intracortical brain recordings
A. Marin-Llobet‡, S. Sánchez-Manso, A. Manasanch, L. Tresserras, X. Zhang, Y. Hua, H. Zhao, M. Torao-Angosto, M. V. Sanchez-Vives‡, L. Dalla Porta‡ Advanced Intelligent Systems, 2025 Shows that Riemannian geometry on covariance features can classify intracortical brain states efficiently and interpretably, outperforming conventional baselines with much lower training cost. Computational Biology
An autonomous AI agent for universal behavior analysis
A. Aljović†, Z. Lin†, W. Wang, X. Zhang, A. Marin-Llobet, N. Liang, B. Canales, J. Lee, J. Baek, R. Liu, C. Li, N. Li, J. Liu‡ bioRxiv, 2025 Introduces BehaveAgent, a multimodal AI agent that automates behavior analysis from video across species and tasks without task-specific retraining or manual annotation. AI Agents
Plastic-elastomer heterostructure for robust flexible brain-computer interfaces
X. Lin, X. Zhang, Z. Wang, J. Chen, J. Lee, A. J. Lee, H. Yang, A. Remy, H. Shen, Y. He, H. Zhao, X. Zhang, W. Wang, A. Aljović, J. J. Vlassak, N. Lu, J. Liu‡ bioRxiv, 2025 Introduces a plastic–elastomer heterostructure platform that combines mechanical robustness with tissue-level flexibility for scalable long-term brain-computer interfaces. Computational Biology
Spatial transcriptomics AI agent charts hPSC-pancreas maturation in vivo
Z. Lin†, W. Wang†, A. Marin-Llobet, Q. Li, S. D. Pollock, X. Sui, A. Aljovic, J. Lee, J. Baek, N. Liang, X. Zhang, C. K. Wang, J. Huang, M. Liu, Z. Gao, H. Sheng, J. Du, S. J. Lee, B. Wang, Y. He, J. Ding, X. Wang, J. R. Alvarez-Dominguez‡, J. Liu‡ bioRxiv, 2025 Presents STAgent, an autonomous multimodal agent that automates spatial transcriptomics analysis and charts in vivo maturation of hPSC-derived pancreatic tissues. MultimodalAI AgentsComputational Biology
Spike sorting AI agent
Z. Lin†, A. Marin-Llobet†, J. Baek, Y. He, J. Lee, W. Wang, X. Zhang, A. J. Lee, N. Liang, J. Du, J. Ding, N. Li, J. Liu‡ bioRxiv, 2025 Develops an autonomous AI agent for spike sorting that streamlines preprocessing, detection, clustering, and validation to improve scalability and reproducibility of neural data analysis. AI AgentsComputational Biology
DeviceAgent: An autonomous multimodal AI agent for flexible bioelectronics
J. Lee†, Z. Lin†, W. Wang†, J. Baek†, A. J. Lee, A. Aljović, A. Marin-Llobet, X. Zhang, R. Liu, N. Li, J. Liu‡ bioRxiv, 2025 Introduces a multimodal AI agent that integrates LLMs, VLMs, and domain tools to automate flexible bioelectronics design, fabrication planning, defect inspection, and signal analysis. MultimodalAI AgentsComputational Biology
An AI Agent for cell-type specific brain computer interfaces
A. Marin-Llobet†, Z. Lin†, J. Baek†, A. Aljovic, X. Zhang, A. J. Lee, W. Wang, J. Lee, H. Shen, Y. He, N. Li, J. Liu‡ bioRxiv, 2025 Uses pretrained vision-language models to link extracellular electrophysiological features with molecular cell identities, enabling AI-assisted cell-type-specific brain-computer interfaces. MultimodalAI AgentsComputational Biology
An AI-Cyborg System for Adaptive Intelligent Modulation of Organoid Maturation
R. Liu†, Z. Ren†, X. Zhang†, Q. Li, W. Wang, Z. Lin, R. T. Lee, J. Ding, N. Li‡, J. Liu‡ bioRxiv, 2024 Builds a closed-loop AI-cyborg platform that combines flexible bioelectronics with Bayesian optimization to adaptively modulate organoid maturation in real time. MultimodalComputational Biology
In vivo neural stimulation and recording using flexible bioelectronics
R. Liu, X. Zhang, H. Sheng, J. Liu‡ IEEE IEDM, 2024 Demonstrates soft, tissue-level flexible bioelectronics for minimally invasive in vivo neural recording and stimulation in the brain and spinal cord. Computational Biology
Drift to remember
J. Du, X. Zhang, H. Shen, X. Xian, G. Wang, J. Zhang, Y. Yang, N. Li, J. Liu, J. Ding‡ arXiv, 2024 Proposes DriftNet, a lifelong-learning framework inspired by representational drift that mitigates catastrophic forgetting across image, language, and large-model adaptation tasks. Computer VisionGenerative AI
Realigning representational drift in mouse visual cortex by flexible brain-machine interfaces
S. Zhao†, H. Shen†, S. Qin, S. Jiang, X. Tang, M. Lee, X. Zhang, J. Lee, J. Chen, J. Liu‡ bioRxiv, 2024 Uses flexible brain-machine interfaces to study and potentially realign representational drift in mouse visual cortex over longitudinal neural recordings. Computer VisionMultimodalComputational Biology
Explainable multi-task learning for multi-modality biological data analysis
X. Tang†, J. Zhang†, Y. He†, X. Zhang, Z. Lin, S. Partarrieu, E. B. Hanna, Z. Ren, H. Shen, Y. Yang, X. Wang, N. Li, J. Ding‡, J. Liu‡ Nature Communications, 2023 Introduces UnitedNet, an explainable multi-task deep learning model for integrating single-cell multimodal data and performing cross-modal biological inference. Computer VisionMultimodalComputational Biology
CARLA Simulated Data for Rare Road Object Detection
T. Bu†, X. Zhang†, C. Mertz, J. M. Dolan IEEE ITSC, 2021 Shows that synthetic data generated in the CARLA simulator can improve rare-object detection when real labeled examples are scarce. Computer VisionGenerative AI
