About

James S. Duncan is the Ebenezer K. Hunt Professor of Biomedical Engineering at Yale University, with additional appointments in Electrical & Computer Engineering and Radiology & Biomedical Imaging. His research focuses on biomedical image processing and analysis, employing quantitative strategies based on geometrical, physical, and statistical models. His work includes applications in the analysis of left ventricular cardiac function, neuro-structure and function, and image-guided interventions. Dr. Duncan has been recognized as an IEEE Fellow since 2001 and was inducted into the American Institute for Medical and Biological Engineering in 2000, reflecting his significant contributions to the field of biomedical engineering and medical image analysis.

Research topics

• Computer Science
• Artificial Intelligence
• Machine Learning
• Algorithm
• Data Mining
• Neuroscience
• Theoretical computer science
• Applied mathematics
• Computer vision
• Mathematics
• Mathematical optimization
• Psychology

Selected publications

• Intramyocardial Hydrogel Delivery Decreases Left Ventricular Remodeling and Increases Angiogenesis Post Myocardial Infarction

  Circulation Cardiovascular Imaging · 2025-09-29 · 3 citations

  articleOpen access

  BACKGROUND: Intramyocardial injection of hydrogel into myocardial infarction (MI) areas can reduce left ventricular remodeling and potentially increase angiogenesis post-MI. The radiotracer 99m Tc-Maraciclatide binds to activated alpha-v-beta-3 (αvβ3)-integrin, a key factor in angiogenesis, and can be used to evaluate myocardial angiogenesis. This study used multimodality imaging to assess the effects of imageable intramyocardial hydrogel delivery on left ventricular remodeling and angiogenesis after MI. METHODS: Fourteen pigs (N=14) underwent 90 minutes of balloon occlusion and reperfusion. Five days post-MI, they were randomized to receive either intramyocardial hydrogel (n=8) or control saline injections (n=6). Contrast cine–computed tomography was used to assess biomechanical changes before and after treatment (day 1, day 5, and day 12). 99m Tc-Maraciclatide uptake was measured with gamma well counting. Scar burden and angiogenesis were evaluated through histology. RESULTS: Both groups initially showed a decrease in ejection fraction and an increase in end-diastolic volume post-MI. Hydrogel delivery on day 5 led to a reduction in end-diastolic volume and improvement in left ventricular ejection fraction by day 12. The hydrogel group also exhibited decreased compensatory radial strain in remote myocardial segments, but decreased strain in the hydrogel myocardial segments. There was increased uptake of 99m Tc-maraciclatide in the infarct segments after hydrogel delivery, associated with increased αvβ3-integrin and factor VIII expression in the hydrogel treatment group on histology. However, there was no difference in regional inflammation or scar size between the groups. CONCLUSIONS: Intramyocardial delivery of hydrogel early post-MI resulted in decreased left ventricular remodeling and increased αvβ3-integrin activation associated with an increase in angiogenesis.

  PublisherOA PDFDOI

• Causal Modeling of FMRI Time-Series for Interpretable Autism Spectrum Disorder Classification

  2025-04-14 · 2 citations

  articleOpen accessSenior author

  Autism spectrum disorder (ASD) is a neurological and developmental disorder that affects social and communicative behaviors. It emerges in early life and is generally associated with lifelong disabilities. Thus, accurate and early diagnosis could facilitate treatment outcomes for those with ASD. Functional magnetic resonance imaging (fMRI) is a useful tool that measures changes in brain signaling to facilitate our understanding of ASD. Much effort is being made to identify ASD biomarkers using various connectome-based machine learning and deep learning classifiers. However, correlation-based models cannot capture the non-linear interactions between brain regions. To solve this problem, we introduce a causality-inspired deep learning model that uses time-series information from fMRI and captures causality among ROIs useful for ASD classification. The model is compared with other baseline and state-of-the-art models with 5-fold cross-validation on the ABIDE dataset. We filtered the dataset by choosing all the images with mean FD less than 15mm to ensure data quality. Our proposed model achieved the highest average classification accuracy of 71.9% and an average AUC of 75.8%. Moreover, the inter-ROI causality interpretation of the model suggests that the left precuneus, right precuneus, and cerebellum are placed in the top 10 ROIs in inter-ROI causality among the ASD population. In contrast, these ROIs are not ranked in the top 10 in the control population. We have validated our findings with the literature and found that abnormalities in these ROIs are often associated with ASD.

  PublisherOA PDFDOI

• Towards Zero-Shot Task-Generalizable Learning on fMRI

  ArXiv.org · 2025-02-15

  preprintOpen accessSenior author

  Functional MRI measuring BOLD signal is an increasingly important imaging modality in studying brain functions and neurological disorders. It can be acquired in either a resting-state or a task-based paradigm. Compared to resting-state fMRI, task-based fMRI is acquired while the subject is performing a specific task designed to enhance study-related brain activities. Consequently, it generally has more informative task-dependent signals. However, due to the variety of task designs, it is much more difficult than in resting state to aggregate task-based fMRI acquired in different tasks to train a generalizable model. To resolve this complication, we propose a supervised task-aware network TA-GAT that jointly learns a general-purpose encoder and task-specific contextual information. The encoder-generated embedding and the learned contextual information are then combined as input to multiple modules for performing downstream tasks. We believe that the proposed task-aware architecture can plug-and-play in any neural network architecture to incorporate the prior knowledge of fMRI tasks into capturing functional brain patterns.

  PublisherOA PDFDOI

• Noninvasive Tumor Profiling: Quantitative Contrast-enhanced MRI Markers Help Predict PD-L1 and CTNNB1 Status in Hepatocellular Carcinoma

  Radiology · 2025-08-01 · 4 citations

  articleOpen access

  Quantitative multiparametric contrast-enhanced MRI parameters helped predict molecular profiles of hepatocellular carcinoma via machine learning algorithms.

  PublisherOA PDFDOI

• STNAGNN: Data-driven Spatio-temporal Brain Connectivity beyond FC.

  PubMed · 2025-04-08

  preprintOpen accessSenior author

  In recent years, graph neural networks (GNNs) have been widely applied in the analysis of brain fMRI, yet defining the connectivity between ROIs remains a challenge in noisy fMRI data. Among all approaches, Functional Connectome (FC) is the most popular method. Computed by the correlation coefficients between ROI time series, FC is a powerful and computationally efficient way to estimate ROI connectivity. However, it is well known for neglecting structural connections and causality in ROI interactions. Also, FC becomes much more noisy in the short spatio-temporal sliding-window subsequences of fMRI. Effective Connectome (EC) is proposed as a directional alternative, but is difficult to accurately estimate. Furthermore, for optimal GNN performance, usually only a small percentage of the strongest connections are selected as sparse edges, resulting in oversimplification of complex brain connections. To tackle these challenges, we propose the Spatio-Temporal Node Attention Graph Neural Network (STNAGNN) as a data-driven alternative that combines sparse predefined FC with dense data-driven spatio-temporal connections, allowing for flexible and spatio-temporal learning of ROI interaction patterns.

  PublisherOA PDF

• 2.5D Multi-View Averaging Diffusion Model for 3D Medical Image Translation: Application to Low-Count PET Reconstruction With CT-Less Attenuation Correction

  IEEE Transactions on Medical Imaging · 2025-05-15 · 4 citations

  article

  Positron Emission Tomography (PET) is an important clinical imaging tool but inevitably introduces radiation exposure to patients and healthcare providers. Reducing the tracer injection dose and eliminating the CT acquisition for attenuation correction can reduce the overall radiation dose, but often results in PET with high noise and bias. Thus, it is desirable to develop 3D methods to translate the non-attenuation-corrected low-dose PET (NAC-LDPET) into attenuation-corrected standard-dose PET (AC-SDPET). Recently, diffusion models have emerged as a new state-of-the-art deep learning method for image-to-image translation, better than traditional CNN-based methods. However, due to the high computation cost and memory burden, it is largely limited to 2D applications. To address these challenges, we developed a novel 2.5D Multi-view Averaging Diffusion Model (MADM) for 3D image-to-image translation with application on NAC-LDPET to AC-SDPET translation. Specifically, MADM employs separate diffusion models for axial, coronal, and sagittal views, whose outputs are averaged in each sampling step to ensure the 3D generation quality from multiple views. To accelerate the 3D sampling process, we also proposed a strategy to use the CNN-based 3D generation as a prior for the diffusion model. Our experimental results on human patient studies suggested that MADM can generate high-quality 3D translation images, outperforming previous CNN-based and Diffusion-based baseline methods. The code is available at https://github.com/tianqic/MADM.

  PublisherDOI

• Adapting Vision Foundation Models for Real-Time Ultrasound Image Segmentation

  Lecture notes in computer science · 2025-09-19 · 2 citations

  book-chapter
  PublisherDOI

• Adapting Vision Foundation Models for Real-time Ultrasound Image Segmentation

  ArXiv.org · 2025-03-31

  preprintOpen access

  We propose a novel approach that adapts hierarchical vision foundation models for real-time ultrasound image segmentation. Existing ultrasound segmentation methods often struggle with adaptability to new tasks, relying on costly manual annotations, while real-time approaches generally fail to match state-of-the-art performance. To overcome these limitations, we introduce an adaptive framework that leverages the vision foundation model Hiera to extract multi-scale features, interleaved with DINOv2 representations to enhance visual expressiveness. These enriched features are then decoded to produce precise and robust segmentation. We conduct extensive evaluations on six public datasets and one in-house dataset, covering both cardiac and thyroid ultrasound segmentation. Experiments show that our approach outperforms state-of-the-art methods across multiple datasets and excels with limited supervision, surpassing nnUNet by over 20\% on average in the 1\% and 10\% data settings. Our method achieves $\sim$77 FPS inference speed with TensorRT on a single GPU, enabling real-time clinical applications.

  PublisherOA PDFDOI

• Snap-and-tune: combining deep learning and test-time optimization for high-fidelity cardiovascular volumetric meshing

  ArXiv.org · 2025-06-09

  preprintOpen accessSenior author

  High-quality volumetric meshing from medical images is a key bottleneck for physics-based simulations in personalized medicine. For volumetric meshing of complex medical structures, recent studies have often utilized deep learning (DL)-based template deformation approaches to enable fast test-time generation with high spatial accuracy. However, these approaches still exhibit limitations, such as limited flexibility at high-curvature areas and unrealistic inter-part distances. In this study, we introduce a simple yet effective snap-and-tune strategy that sequentially applies DL and test-time optimization, which combines fast initial shape fitting with more detailed sample-specific mesh corrections. Our method provides significant improvements in both spatial accuracy and mesh quality, while being fully automated and requiring no additional training labels. Finally, we demonstrate the versatility and usefulness of our newly generated meshes via solid mechanics simulations in two different software platforms. Our code is available at https://github.com/danpak94/Deep-Cardiac-Volumetric-Mesh.

  PublisherOA PDFDOI

• Uncovering Memorization Effect in the Presence of Spurious Correlations

  arXiv (Cornell University) · 2025-01-01

  preprintOpen accessSenior author

  Machine learning models often rely on simple spurious features -- patterns in training data that correlate with targets but are not causally related to them, like image backgrounds in foreground classification. This reliance typically leads to imbalanced test performance across minority and majority groups. In this work, we take a closer look at the fundamental cause of such imbalanced performance through the lens of memorization, which refers to the ability to predict accurately on atypical examples (minority groups) in the training set but failing in achieving the same accuracy in the testing set. This paper systematically shows the ubiquitous existence of spurious features in a small set of neurons within the network, providing the first-ever evidence that memorization may contribute to imbalanced group performance. Through three experimental sources of converging empirical evidence, we find the property of a small subset of neurons or channels in memorizing minority group information. Inspired by these findings, we hypothesize that spurious memorization, concentrated within a small subset of neurons, plays a key role in driving imbalanced group performance. To further substantiate this hypothesis, we show that eliminating these unnecessary spurious memorization patterns via a novel framework during training can significantly affect the model performance on minority groups. Our experimental results across various architectures and benchmarks offer new insights on how neural networks encode core and spurious knowledge, laying the groundwork for future research in demystifying robustness to spurious correlation.

  PublisherOA PDFDOI

Recent grants

• Quantitative Multimodal Imaging Biomarkers for Combined Locoregional and Immunotherapy of Liver Cancer

  NIH · $6.1M · 2016–2027

• q4DE: A Biomarker for Image-Guided, Post-MI Hydrogel Therapy

  NIH · $6.3M · 2014–2025

• NIH Grant R01HL082640

  NIH · $7.0M · 2013

• NIH Grant R01EB000473

  NIH · $13.3M · 2013

• Image Guided Delivery of Bioresponsive Hydrogels

  NIH · $3.2M · 2017–2022

Frequent coauthors

• Lawrence H. Staib

  226 shared

• Rafaela Guisantes

  Administração Regional de Saúde de Lisboa e Vale do Tejo

  144 shared

• Song Zhang

  Fudan University

  144 shared

• Frédérique Frouin

  Institut d'Imagerie Biomédicale

  144 shared

• Nan Meng

  Institut Curie

  144 shared

• Garrett Oncoscore

  Applied Mathematics (United States)

  144 shared

• Caroline Malhaire

  Institut Curie

  144 shared

• William Yaxley

  University of Queensland

  144 shared

Awards & honors

• IEEE Fellow (2001)
• Inducted into American Institute for Medical and Biological…