About

Bastiaan Driehuys is a Professor of Radiology and Biomedical Engineering at Duke University. His research focuses on magnetic resonance imaging (MRI) with hyperpolarized 129Xe gas, a technology he has helped develop for high-resolution, non-invasive imaging of pulmonary function. His work is at the forefront of this novel imaging technique, which enhances MRI signals by a factor of 100,000 through laser-based hyperpolarization, enabling detailed visualization of lung health and disease. Driehuys's research program involves multiple NIH and industry-sponsored studies, and his team is working towards clinical implementation following recent FDA approval of the technology. His studies aim to improve early diagnosis and monitoring of interstitial and pulmonary vascular diseases, collaborating closely with pulmonary medicine colleagues. He provides research opportunities for Ph.D., Masters, medical, and undergraduate students, contributing significantly to the advancement of pulmonary MRI.

Research topics

• Medicine
• Internal medicine
• Radiology
• Physics
• Computer Science
• Nuclear medicine
• Nuclear magnetic resonance
• Thermodynamics
• Medical physics
• Physical therapy
• Chemistry
• Cardiology

Selected publications

• Correcting <scp> ¹²⁹ Xe </scp> Gas Exchange <scp>MRI</scp> for Incidental Gas‐Phase Excitation—Comparing Approaches, and Identifying Acceptable Thresholds for Reliable Quantification

  Magnetic Resonance in Medicine · 2026-04-20

  articleSenior authorCorresponding

  PURPOSE: Xe gas exchange MRI introduces contamination that can bias membrane uptake and red blood cell (RBC) transfer metrics. Here, we defined the level of gas-phase contamination that is acceptable, and compared the dual-echo (Hahn) and spectroscopy-informed (Willmering) frameworks to correct for it. METHODS: Xe MRI and spectroscopy. The dissolved-phase RF pulse was shortened from its optimal value to increase incidental gas contamination. The Hahn and Willmering corrections were implemented and refined using a grid search over a contamination-dominated ROI to determine any platform-specific scale and phase adjustments. The effects of contamination phase and scale were tested by synthetically reintroducing gas signal at phase offsets that maximized defect- and high intensity-like artifacts. Membrane/gas and RBC/gas defect and high percentages were evaluated against thresholds derived from a healthy reference cohort to determine acceptable gas contamination. RESULTS: Images acquired with the shortened pulse contained 24% ± 6% gas contamination. After applying empirically calibrated phase shifts (Hahn≈230°, Willmering≈49°) and no scale adjustment (optimal value≈1), both correction methods increased RBC SNR by ∼9% and reduced background noise. Phase- and scale-sweep experiments indicated that gas contamination becomes unacceptable once it exceeds ∼9% ± 3%, at which point membrane and RBC defect or high percentages surpass their thresholds. CONCLUSIONS: Gas-phase contamination beyond ∼9% requires correction. After a one-time empirical phase calibration, this can be equally well achieved by both the Hahn and Willmering methods.

  PublisherDOI

• Hyperpolarized 129Xe MRI Features Associated with Interstitial Lung Disease Identified Using an Interpretable Diagnostic Algorithm

  Academic Radiology · 2026-05-01

  articleOpen accessSenior author
  PublisherOA PDFDOI

• Optimized methods for robust calculation of chemical shifts in hyperpolarized 129Xe MR spectroscopy

  Proceedings on CD-ROM - International Society for Magnetic Resonance in Medicine. Scientific Meeting and Exhibition/Proceedings of the International Society for Magnetic Resonance in Medicine, Scientific Meeting and Exhibition · 2025-09-16

  articleSenior author

  Motivation: Hyperpolarized 129Xe RBC chemical shift is an increasingly important biomarker for lung disease, but the robustness of its calculation has not been investigated. Goal(s): The aim of this study is to identify a more accurate and optimal method of calculating RBC chemical shift in 129Xe MRS compared to standard practice. Approach: We investigated two approaches for the 0-ppm 129Xe gas-phase reference frequency: 1) the incidental gas signal arising during RBC excitation and 2) a dedicated gas excitation. Results: RBC shifts calculated using dedicated gas-phase references were found to be more repeatable and potentially more accurate than standard incidental references. Impact: Using dedicated 129Xe gas-phase excitation to calculate the 0-ppm reference frequency yielded more repeatable and accurate quantifications of the RBC shift than standard practice, enabling more robust characterization of lung disease via 129Xe MRS.

  PublisherDOI

• The emerging role of hyperpolarized ¹²⁹ Xe MRI in pulmonary hypertension

  Expert Review of Respiratory Medicine · 2025-07-02 · 1 citations

  editorialOpen access
  PublisherOA PDFDOI

• Design and Implementation of a Multi‐Center Trial of <scp> ¹²⁹ Xe </scp> Gas Exchange <scp>MRI</scp> and <scp>MRS</scp> to Evaluate Longitudinal Progression of <scp>COPD</scp>

  Journal of Magnetic Resonance Imaging · 2025-04-23 · 8 citations

  articleOpen access1st authorCorresponding

  ABSTRACT MR imaging holds the potential to enhance drug development efficiency by de‐risking early phase studies and increasing confidence in results. It can improve patient selection, increase repeatability, and provide greater sensitivity to change, thereby enabling smaller, faster clinical trials. For trials in the pulmonary space, hyperpolarized 129 Xe MRI is appealing because it provides 3‐dimensional imaging of pulmonary ventilation and gas exchange in a brief, non‐invasive exam. Metrics derived from 129 Xe MRI may be more sensitive to disease progression than conventional lung function assessments and may thus provide a valuable means to evaluate numerous novel pharmacologic and biologic therapies now in development. However, despite the acute need for better patient selection and for prognostic and monitoring biomarkers, 129 Xe MR imaging is not yet widely utilized in pulmonary drug development, partly because such trials must be conducted at multiple centers to enroll enough participants. Thus, incorporating 129 Xe MRI requires broader dissemination, harmonized image acquisition protocols, standardized dose delivery, visualization, and quantification. Multi‐site trials must also be able to operate across all major MRI vendor platforms and diverse software/hardware revisions. To this end, the 129 Xe MRI Clinical trials consortium has published a harmonized protocol describing four recommended acquisitions. Here we report on the first industry‐sponsored study to deploy this 129 Xe MRI/MRS protocol in a multi‐center, multi‐platform, multi‐national study to evaluate longitudinal progression of chronic obstructive pulmonary disease (COPD). We demonstrate the steps necessary to implement standardized 129 Xe‐MRI acquisition techniques across multiple sites and discuss the practices implemented, quality control approaches, and lessons learned for facilitating and accelerating the implementation of future trials that incorporate this technology. Level of Evidence: 5.

  PublisherOA PDFDOI

• Classification of COPD and ILD Subtypes Using 129Xe MRI/MRS with Unsupervised Cluster Analysis

  Proceedings on CD-ROM - International Society for Magnetic Resonance in Medicine. Scientific Meeting and Exhibition/Proceedings of the International Society for Magnetic Resonance in Medicine, Scientific Meeting and Exhibition · 2025-09-16

  articleSenior author

  Motivation: 129Xe MRI/MRS can assess distinct aspects of pulmonary gas exchange and hemodynamics. However, there is no gold standard against which these metrics can be validated. Goal(s): To evaluate whether unsupervised cluster analysis of 129Xe MRI/MRS metrics naturally reveal patterns known to be associated with certain disease groups. Approach: Eight 129Xe MRI/MRS features were subjected to k-means clustering with internal validation indices used to determine optimal cluster number. Results: The analysis identified four clusters that largely distinguished healthy, COPD, and ILD patient groups. Impact: This study offers a pathway for designing future prospective clinical trials that could validate non-invasive 129Xe MRI/MRS metrics of gas exchange by demonstrating that certain patterns distinguish between lung disease subtypes with high accuracy.

  PublisherDOI

• XeLHC: Harmonized Acquisition of Ventilation, Diffusion, and Gas Exchange 129Xe MRI At 3 Sites in the Lung Health Cohort

  Proceedings on CD-ROM - International Society for Magnetic Resonance in Medicine. Scientific Meeting and Exhibition/Proceedings of the International Society for Magnetic Resonance in Medicine, Scientific Meeting and Exhibition · 2025-09-16

  article

  Motivation: There is a limited understanding of multi-site comparability of Hyperpolarized 129Xe MRI (Xe-MRI). Goal(s): To better understand variability of Xe-MRI across sites, we examine preliminary results from ventilation, diffusion-weighted, and gas exchange Xe-MRI in 64 healthy young adults from the Lung Health Cohort. Approach: 3 of 5 designated sites have enrolled and imaged 64 participants following 129Xe MRI Clinical Trials Consortium recommended protocols. Images have been compared across sites to assess preliminary multi-site comparability of data. Results: Preliminary analysis indicates strong cross-site comparability in Ventilation and Gas Exchange Xe-MRI. Diffusion-weighted images appear to be less well harmonized. Impact: Trials using xenon MRI have been hampered by an incomplete understanding of normal variation across sites. In this preliminary analysis, we will examine the 3 major Xe-MRI contrasts (ventilation, diffusion-weighted, gas exchange) in 64 healthy volunteers imaged across three sites.

  PublisherDOI

• Same-session Repeatability of 129Xe MRI/MRS Measures of Gas Exchange in Idiopathic Pulmonary Fibrosis

  American Journal of Respiratory and Critical Care Medicine · 2025-05-01

  article

  Abstract Rationale. Patients with idiopathy pulmonary fibrosis (IPF) are known to have altered patterns of 129Xe gas exchange MRI and spectroscopy, including a reduced red blood cell to membrane ratio (RBC:M), reduced RBC chemical shift, and increased membrane high percent (MHP). On spectroscopy, the RBC:M is believed to reflect the efficiency of gas exchange between the interstitial membrane and the RBCs, while RBC chemical shift reflects pulmonary capillary oxygenation. On imaging, the MHP is believed to reflect interstitial membrane and thickening and has been associated with therapy response and poor outcomes in IPF. Previous work by Hahn et al. has characterized the repeatability of regional 129Xe gas exchange metrics in healthy subjects. Here we evaluate the repeatability of these measurements in IPF to differentiate normal measurement variation from meaningful changes reflecting disease progression or therapy response. Methods. Participants with diagnosed IPF (N=11, 1F/10M, age 70.55 ± 6.31yrs) underwent hyperpolarized 129Xe MRI. All underwent scans at study baseline using a 1L dose bag of xenon mixture, with a subset also undergoing scans at follow-up visits at 6 months (N=8) and one year (N=5) after baseline, for a total of 24 visits in all. At each visit, two sets of 129Xe MRI/MRS dynamic spectroscopy were performed with a brief break between each pair of scans. Repeatability was assessed using the Bland-Altman analysis and inter-class correlation coefficient (ICC). Results. RBC:M was most repeatable, with Bland-Altman limits of agreement (LOA) of [-0.062 – 0.057] and an ICC of 0.89. RBC chemical shift was also repeatable, with an LOA of [-0.500 – 0.435 ppm] and an ICC of 0.83. The MHP measurement had an LOA of [-23.850% – 23.250%] and an ICC of 0.7. There was negligible bias between the first and second scan across all three measurements. Conclusion. These results suggest that the RBC:M measurement is particularly repeatable in IPF, as well as the RBC chemical shift. The repeatability of the membrane high percent (MHP) was more modest, perhaps due in part to sensitivity to lung inflation volume. However, this marker exhibits large changes in interstitial lung diseases, is sensitive to both disease progression and therapy response, and has potential as a treatable trait in IPF. Studies using tailored dose bag volumes and consistent patient coaching are ongoing and expected to improve repeatability further. Overall, these findings provide a foundational assessment of repeatability of 129Xe MRI and spectroscopy in IPF.

  PublisherDOI

• Improved Workflow for Hyperpolarized 129Xe MRI through Automated Adjustments

  Proceedings on CD-ROM - International Society for Magnetic Resonance in Medicine. Scientific Meeting and Exhibition/Proceedings of the International Society for Magnetic Resonance in Medicine, Scientific Meeting and Exhibition · 2025-09-16

  article

  Motivation: Standard scanner adjustment routines are not generally applicable for hyperpolarized 129Xe lung MRI and thus manual input of required adjustment values by the technologist or researcher is typically required, which is both cumbersome and prone to error. Goal(s): Develop an automated adjustment-processing framework that makes the workflow for hyperpolarized 129Xe MRI analogous to that for routine proton MRI. Approach: An automated adjustment-processing framework was implemented via online custom adjustment calculations and pulse-sequence modifications to automatically read and set values from the custom calculations. Results: The automated adjustment-processing framework permitted 129Xe MRI with appropriate adjustment values and without manual interaction from the scanner operator. Impact: The proposed automated adjustment-processing framework results in a workflow for xenon lung MRI that mirrors that for routine clinical proton MRI, increasing the accessibility of hyperpolarized 129Xe lung MRI and facilitating its translation into clinical use.

  PublisherDOI

• Establishing standardized healthy reference distributions and values for multisite <scp> ¹²⁹ Xe </scp> gas exchange <scp>MRI</scp> / <scp>MR spectroscopy</scp> at 3 T across major scanner platforms

  Magnetic Resonance in Medicine · 2025-06-02 · 5 citations

  articleOpen accessSenior authorCorresponding

  Abstract Purpose To establish standardized reference distributions and values for 1‐point Dixon 129 Xe gas exchange MRI/MR spectroscopy (MRS) in a multicenter healthy cohort of 18–30‐year‐olds scanned on three major 3T platforms. These distributions and values enable consistent, quantitative analysis and interpretation of images and spectra. Methods Healthy participants from three centers—Duke University (Siemens), Cincinnati Children's Hospital (Philips), and the University of Iowa (General Electric)—underwent pulmonary function testing and a standardized 129 Xe MRI/MRS protocol with dissolved‐phase excitation at 208 ppm from the gas resonance. Raw data were converted to the ISMRMRD format, centrally processed, and used to construct reference values and distributions for each gas‐exchange compartment. The combined reference values were compared to site‐specific values and, with suitable transformation, to a cohort evaluated with 218 ppm excitation. Results The reference distributions were consistent across sites and platforms for both imaging and spectroscopy. Combined imaging reference means were ventilation = 0.71 ± 0.14, membrane (M)/gas = 0.97 ± 0.27 × 10 −2 , red blood cell (RBC)/gas = 0.48 ± 0.20 × 10 −2 , and RBC:M = 0.49 ± 0.11. Spectroscopic chemical shifts were 197.6 ± 0.3 and 218.2 ± 0.5 ppm for membrane and RBC, with an RBC oscillation amplitude of 10.0 ± 2.5%. Estimated global T 2 * values were 0.99 ± 0.04 ms (membrane) and 1.04 ± 0.03 ms (RBC). Quantitative maps showed no significant differences using site‐specific versus combined distributions. Scaled 208‐ppm distributions agreed well with single‐site distributions acquired with 218‐ppm excitation. Conclusions These reference distributions provide a robust benchmark for 3T 129 Xe gas exchange MRI/MRS, ensuring rigor, reproducibility, and cross‐center comparability.

  PublisherOA PDFDOI

Recent grants

• Early Detection of Changes in Pulmonary Gas Exchange by Hyperpolarized Xe MRI

  NIH · $6.0M · 2011–2027

• Xenon MRI in Pulmonary Hypertension

  NIH · $2.9M · 2021–2026

• NIH Grant P41RR005959

  NIH · $4.6M · 2013

• Using MRI To Visualize Regional Therapy Response In Idiopathic Pulmonary Fibrosis

  NIH · $5.7M · 2015–2026

• NIH Grant R01CA142842

  NIH · $1.5M · 2016

Frequent coauthors

• Mu He

  Nanjing University of Information Science and Technology

  87 shared

• Elianna Bier

  77 shared

• Rohan S. Virgincar

  71 shared

• David Mummy

  Duke University

  70 shared

• H. Page McAdams

  66 shared

• Scott H. Robertson

  64 shared

• Zackary I. Cleveland

  63 shared

• Ziyi Wang

  Hainan University

  62 shared

Labs

• Bastiaan Driehuys LabPI

  Not provided