Now, a team of researchers from Berlin has engineered a solution. In a study published in Communications Engineering, Thomas Gladytz, Kathleen Cantow, Erdmann Seeliger, Thoralf Niendorf and colleagues introduce MRI-PHYSIOL-NIRS: an integrated platform that unites magnetic resonance imaging with invasive ground-truth oxygenation probes and near-infrared spectroscopy (NIRS) to decipher, for the first time, the quantitative relationship between MRI-based measures of apparent renal oxygenation and true tissue oxygenation. The work represents a substantive technical advance toward entirely MRI-based measurement of kidney oxygenation in health and disease, with ultimate application in patients.

The clinical stakes are considerable. Standard biomarkers such as serum creatinine and estimated glomerular filtration rate reflect kidney damage only after significant functional loss has already occurred, and none of the current biomarkers reveal any data on renal oxygenation. Given the vital role of tissue oxygenation in kidney pathology, renal oxygenation has strong potential as an early diagnostic biomarker in AKI—where it may enable detection of injury before irreversible functional decline—as well as a prognostic biomarker in patients at increased risk of ischemic injury, those with early CKD or early diabetic kidney disease. What has been missing is a validated means to monitor renal oxygenation in vivo and use that information to characterize the state of the kidney along a spectrum from healthy to compromised.

MRI can probe tissue oxygenation non-invasively because deoxygenated hemoglobin is paramagnetic and alters MR signals. Accordingly, the transverse relaxation times T₂ and T₂* are sensitive to deoxyhemoglobin levels and are used as indirect markers of tissue oxygenation. However, the relationship of these metrics with true tissue pO₂ is confounded by several physiological factors, including renal blood and tubular volume changes that affect kidney size and signal interpretation. Clear mechanistic links between MRI metrics of apparent oxygenation and true tissue oxygenation have remained lacking, and these discrepancies are a major reason why clinical use of MRI for renal oximetry has been limited.

The gold standard for measuring tissue oxygenation in vivo is invasive probes that provide absolute measurements of tissue pO₂. These probes have two major drawbacks: they must penetrate the tissue, making them entirely unsuitable for use in humans, and they measure pO₂ only at the pinpoint site of insertion, creating sampling bias in an organ that displays considerable spatial heterogeneity even within its major compartments. NIRS can bridge the gaps between these pinpoint measurements and whole-organ MRI coverage, at a size scale intermediate between the two. Moreover, NIRS allows estimation of several physiological parameters that confound the MRI-pO₂ relationship, including oxygenated, deoxygenated and total hemoglobin per tissue volume, and the oxygen saturation of hemoglobin.

Integrating these three modalities within the same animal posed formidable biomedical engineering challenges. All NIRS and PHYSIOL electronics were placed outside the 5 Gauss line of the 9.4 Tesla small-bore MR system in dedicated racks, with flexible optical fibers and extension lines guiding signals to and from the kidney. To navigate the spatial constraints of the MRI bore, which offers a clearance of only about 60 millimeters in diameter, the researchers used aluminum-coated 90-degree reflective micro prisms at the fiber ends to guide light into and out of the kidney along the longitudinal axis of the bore. A customized animal holder was designed to fix and stabilize the probes, ensuring safe transport of the animal into the scanner. The MR compatibility of the fiber probe head was verified in phantom experiments and in vivo, with the probes appearing as compact signal voids used to monitor positioning.

The NIRS setup itself was tailored for the small rat kidney, which measures just over 20 millimeters along its long axis. Nine lasers spanning wavelengths from 660 to 1060 nanometers were sequentially switched on for 4.7 milliseconds each in every acquisition cycle, with light detected by ten avalanche photodiodes. Photon transport through the highly perfused renal tissue was modeled by full GPU-based Monte Carlo simulation rather than the commonly applied diffusion approximation, which is invalid at the source-detector separations below 8 millimeters imposed by rat kidney anatomy. Validation on eleven calibration phantoms with known optical properties yielded a mean absolute percentage error of just 4 percent for the integrated NIRS-derived absorption coefficients. For the PHYSIOL setup, the conventional brass reflector of the perivascular flow probe was replaced by one made of macor ceramics to ensure MRI compatibility, and a long catheter placed the pressure transducer outside the magnet bore.

The team validated the platform’s feasibility by examining intra-individual repeatability at baseline under physiological conditions in 25 rats, demonstrating high consistency, reliability and validity. Recordings of all parameters throughout the baseline period were stable, showing only small physiological fluctuations. The quantitative maps obtained for T₂*, T₂, tubular volume fraction, blood volume fraction and MRI-derived oxygen saturation demonstrated that the measured parameters are not uniform throughout the kidney but differ between the renal cortex and medulla. These baseline results were quantitatively consistent with data previously acquired using the modalities separately, indicating that the integration was successful.

To showcase the platform’s capabilities, the researchers applied two contrasting interventions while the living animals were inside the MR scanner. The first was transient occlusion of the suprarenal aorta, an acute pathophysiological challenge relevant to procedures such as kidney transplantation that risk severe intrarenal hypoxia and acute ischemia-reperfusion injury. Aortic occlusion resulted in significant changes in nearly all measured parameters: pO₂, T₂*, T₂, MRI-derived oxygen saturation, blood volume fraction, tubular volume fraction and kidney size all decreased, along with NIRS-derived oxygen saturation and hemoglobin measures, while cortical deoxyhemoglobin increased. Upon release of the occlusion, most parameters returned to pre-occlusion levels within four minutes. Critically, the analysis showed significant correlations between pO₂ and T₂* and T₂, indicating that changes in these MRI metrics during occlusion and recovery are primarily driven by tissue oxygenation.

The second intervention was hyperoxia, a widely used therapeutic measure achieved by increasing the fraction of inspired oxygen from 21 to 100 percent. Hyperoxia increased cortical and medullary pO₂ and MRI-derived T₂* and T₂, while blood volume fraction decreased. Again, significant correlations between pO₂ and the MRI metrics indicated that T₂* and T₂ changes were essentially driven by tissue oxygenation. However, the correlations of T₂* and T₂ changes with those of blood volume fraction, tubular volume fraction and kidney size revealed that these structural and hemodynamic factors have a considerable impact on the relationship between true and apparent oxygenation. The authors conclude that MRI-based quantification of tubular and blood volume fractions should routinely be performed together with kidney size for physiological interpretation of renal T₂* and T₂.

By linking MR oximetry to the ground-truth metric of tissue pO₂, the researchers are closer to the ultimate objective of applying non-invasive MRI of renal oxygenation to patients in clinical practice. Stand-alone MRI oximetry could be implemented as a tool for risk assessment and early diagnosis of AKI in patients with pre-existing conditions such as CKD, diabetes and old age who undergo procedures involving large volumes of x-ray contrast agents or renal ischemia-reperfusion injury. A case in point is kidney transplantation, where the oxygenation status of donor kidneys could become a key prognostic criterion in the Kidney Donor Profile Index and might even expand the pool of available kidneys through inclusion in the Expanded Donor Criteria. For prospective living donors, an oximetry-based prognostic estimate of the future of their remaining kidney may help decision-making, while oximetry of both the transplanted kidney and the donor’s remaining kidney may become a mainstay of post-transplantation monitoring.

The multimodal nature of the readouts, combined with simultaneous measurements at high temporal resolution and spatial mapping, marks a substantive advance over prior studies using single modalities. Previous attempts to analyze the quantitative relationship between renal T₂* and T₂ and tissue pO₂ relied primarily on comparisons of MRI data with invasive pO₂ measurements obtained in separate cohorts of animals, in the left versus the right kidney, or by sequential data acquired at different times. One earlier study utilized a dedicated integrated MRI-PHYSIOL setup to obtain simultaneous data from the same kidney in vivo but did not obtain data on the major confounders of the T₂*-pO₂ relationship. The MRI-PHYSIOL-NIRS platform now provides an unprecedented opportunity to decode the links between MRI metrics and physiological measures of renal tissue oxygen while avoiding confounding effects of inter-individual differences that arise when using sequential examinations with stand-alone modalities.

Looking forward, the researchers envision assembling a data dictionary of changes in renal oxygenation in response to specific perturbations, and how MRI oximetry measurements relate to PHYSIOL and NIRS measurements. The MRI data can be fed into machine learning algorithms such as generative adversarial networks, with NIRS and PHYSIOL data serving as ground truth. Such networks can be used for renal systems analysis to evaluate how accurately true renal oxygenation can be predicted from MRI measures alone. The next step is to extend MRI-PHYSIOL-NIRS analysis to pathological conditions, where the integrated platform could track oxygenation changes in various renal disease models. Longitudinal data could further train these networks to detect oxygenation changes before clinical signs appear, enabling early disease interception.

The platform also supports the 3R principles of animal research. If the three methods were performed as stand-alone experiments, substantially larger cohorts would be required to overcome biological variability among individual animals to achieve the same statistical power. MRI-PHYSIOL-NIRS can be used with longitudinal experimental designs where animals serve as their own controls, thus reducing animal numbers. Furthermore, by calibrating MRI readouts against ground-truth measurements of tissue oxygenation, the researchers generate a validated reference framework that will allow future studies to rely on MRI alone, without invasive probes, to refine pre-clinical kidney research.

While the platform was implemented in a small-bore 9.4 Tesla preclinical scanner rather than on a clinical human MR system, the high field strength was required to achieve sufficient sensitivity and spatial resolution for studies in small rodents. In clinical MRI, the lower intrinsic sensitivity at 1.5 to 3.0 Tesla is largely offset by the use of larger voxel sizes. Importantly, the use of a preclinical model enables invasive ground-truth measurements for calibration and validation of MRI-based readouts—procedures that would not be ethically feasible in human subjects. Although MRI-PHYSIOL-NIRS was customized to target the kidney, potential applications extend to other organs where changes in oxygenation have profound physiological consequences, and additional non-invasive modalities such as photoacoustic imaging or MR thermometry could enrich the methodological portfolio.

In current routine clinical practice, physicians are completely lacking any reliable non-invasive means to observe the oxygenation vulnerabilities that drive kidney pathology. The MRI-PHYSIOL-NIRS platform offers the tantalizing possibility of extending our understanding of renal pathophysiology from rodent models to humans and lays the foundation for future clinical applications to better diagnose and intercept kidney disease, where MRI oximetry could emerge as a standalone option empowered by artificial intelligence. This potential could have a significant impact on daily clinical practice, transforming how kidney disease is detected, monitored and treated across the spectrum from acute injury to chronic decline.

Subject of Research: MRI-PHYSIOL-NIRS: an integrated platform to study kidney oxygenation

Article Title: MRI-PHYSIOL-NIRS: an integrated platform to study kidney oxygenation

Article References: Gladytz, T., Cantow, K., Millward, J. M., Waiczies, S., Tasbihi, E., Grosenick, D., Seeliger, E., & Niendorf, T. (2026). MRI-PHYSIOL-NIRS: an integrated platform to study kidney oxygenation. Communications Engineering, 5(1), Article 176. https://doi.org/10.1038/s44172-026-00799-x

Image Credits: AI Generated

DOI: 10.1038/s44172-026-00799-x

Keywords: MRI-PHYSIOL-NIRS, integrated, platform, kidney, oxygenation, scientific research