Photon-counting CT iodine maps show promise in early thoracic imaging studies

Photon-counting detector CT can generate iodine maps from every scan without the dedicated dual-energy protocols that conventional CT requires, and a new scoping review has now mapped how this capability is being used across thoracic medicine. The review, published in European Radiology Cardiovascular and Thoracic, analyzed ten early clinical studies covering pulmonary embolism, chronic thromboembolic pulmonary hypertension, parenchymal lung disease, pleural empyema, and cardiac applications. Its central message is nuanced: the technology performs well against external reference standards, but its real advantage lies in workflow rather than raw diagnostic power.

The review team, led by Ana M. Rendón-Garavito, searched MEDLINE, Embase, and Web of Science through February 2026, following the Joanna Briggs Institute framework and PRISMA-ScR reporting guidelines. Two reviewers screened records independently, and from 112 initial database hits the process ultimately yielded ten studies, all published between 2022 and 2025. Every included investigation was a single-center university-hospital study: four from Switzerland, three from Germany, and one each from France, Austria, and the United States. Six were retrospective and four prospective. Together the cohorts enrolled 832 patients, of whom 722 underwent photon-counting CT examinations, with roughly 696 unique patients after accounting for documented overlap between study groups.

All ten studies used the first-generation NAEOTOM Alpha scanner from Siemens Healthineers, and six relied on prototype, pre-release, or non-commercial post-processing software. That single-vendor dependence matters, the authors caution, because reported performance cannot yet be separated from one specific detector-software combination. Iodine contrast concentrations ranged from 300 to 400 mgI/mL with injection volumes of 40 to 100 mL, and protocols varied substantially across sites, limiting how easily the findings transfer to other clinical settings.

The physics behind the technology explains why iodine maps come almost for free on photon-counting systems. Iodine’s K-edge sits at 33.2 keV, meaning the element’s strongest spectral signature falls in the low-energy range that conventional energy-integrating detectors weight least heavily, since their signal scales with photon energy. Photon-counting detectors record each photon’s energy directly, preserving that signal and maintaining iodine quantification precision at substantially lower radiation doses than conventional dual-energy methods. In conventional dual-energy CT, spectral acquisitions require predefined tube-voltage combinations that are incompatible with high-pitch protocols, so high-pitch CT pulmonary angiography and spectral reconstruction have been mutually exclusive. Photon-counting CT removes that constraint by acquiring spectral data at any pitch.

In acute pulmonary embolism, iodine maps were a secondary product of high-pitch CT pulmonary angiography rather than the primary diagnostic endpoint, precisely because those high-pitch protocols cannot generate maps from conventional detector data. Two prospective studies by the same group enrolled 219 patients in total, one using a non-randomized matched-pair design with 51 examinations on each scanner type and the other a randomized comparison of 58 photon-counting and 59 conventional examinations. Subjective map quality was rated excellent or good by three readers in 94.0, 96.0, and 88.0 percent of photon-counting examinations, with no map rated non-diagnostic in the first study. In the direct comparison, map quality scores did not differ significantly between photon-counting and conventional dual-energy derived maps for any reader. Adding the iodine map to anatomic images yielded incremental pulmonary embolism detection in two additional photon-counting patients and one additional conventional patient.

Radiation dose emerged as a consistent technical advantage in the pulmonary vascular domain. Photon-counting CT used 13 to 26 percent lower effective doses than energy-integrating dual-energy CT in acute pulmonary embolism: 1.33 plus or minus 0.47 mSv versus 1.80 plus or minus 0.82 mSv in one study and 2.2 versus 2.8 mSv in the other, both differences statistically significant. In chronic thromboembolic pulmonary hypertension, the photon-counting effective dose of 1.19 plus or minus 0.33 mSv was roughly 80 percent lower than the 6.34 plus or minus 1.68 mSv delivered by perfusion SPECT/CT. Notably, however, the non-diagnostic iodine map rate in one acute pulmonary embolism study was 5.2 percent, attributed by the authors to early arterial timing producing heterogeneous parenchymal iodine distribution.

Chronic thromboembolic pulmonary hypertension showed the most internally consistent evidence in the review. Two studies from the same Zurich cohort, with confirmed patient overlap, evaluated 83 patients and a 26-patient subset. Diagnostic accuracy of the iodine map against an interdisciplinary clinical reference reached 0.88 on consensus reading, with sensitivity of 0.94, specificity of 0.84, and interreader Cohen kappa of 0.74. Per-reader analysis yielded accuracy of 0.85 to 0.88. Quantitative extent of perfusion defects on the iodine maps correlated with mean pulmonary artery pressure at r = 0.49 and with pulmonary vascular resistance at r = 0.52 to 0.60. When judged against perfusion SPECT/CT consensus instead, accuracy dropped to 0.73, illustrating how much the choice of reference standard shapes reported performance.

Cardiac applications demonstrated that late-phase photon-counting acquisitions can characterize myocardial tissue in ways that previously required cardiac MRI. In suspected acute myocarditis, iodine map-derived extracellular volume correlated with late gadolinium enhancement burden at r = 0.82 and with cardiac MRI-measured extracellular volume at r = 0.95 in basal segments, 0.91 in mid-ventricular segments, and 0.96 globally, with a threshold of 26.9 percent yielding an area under the curve of 0.95 for myocarditis diagnosis. In cardiac amyloidosis, the dual-energy method showed a statistically significant 2.24 percent offset against cardiac MRI while the single-energy method did not, a pattern concordant with prior dual-energy CT studies in which dual-energy methods overestimated MRI-measured extracellular volume by 2 to 3 percent. In late iodine enhancement detection, per-patient accuracy of the iodine map reached 85.2 percent for both readers against cardiac MRI as the reference standard.

Beyond the vascular and cardiac domains, the review captured adjunct roles for iodine maps in parenchymal and pleural disease. In a retrospective study of 325 patients with 206 confirmed lesions, adding the iodine map to conventional reconstructions significantly increased diagnostic confidence across consolidations, ground-glass opacities, and reticular lesions, and lesion classification changed in 29 of 206 lesions when the map was available. In pleural empyema, combined virtual monoenergetic images and iodine map reading significantly improved diagnostic accuracy and reader confidence at high monoenergetic energies, with the highest accuracy of 96 percent reached at 40 keV combined with the iodine map, using histopathology as the external reference standard.

The review’s most sobering finding is what remains missing: no study reported patient-level clinical outcomes, meaning no evidence yet exists on whether iodine maps actually change clinical management. Diagnostic accuracy against an external reference was reported in five studies, incremental diagnostic value in three, and quantitative measurements derived from the map in six, but none evaluated impact on patients themselves. The authors conclude that the distinguishing photon-counting contribution is the workflow advantage of generating iodine maps from every acquisition, including high-pitch protocols incompatible with conventional dual-energy, rather than superior accuracy. They call for prospective, multicenter studies with pre-specified clinical endpoints, multivendor validation as additional photon-counting platforms reach clinical use, direct comparisons against conventional dual-energy CT at matched dose and reconstruction settings, and standardized reconstruction methodologies before evidence-based clinical adoption can proceed.

Subject of Research: Clinical applications of iodine maps generated by photon-counting detector CT in thoracic imaging

Article Title: Iodine maps from photon-counting detector CT in thoracic imaging: a scoping review

Article References: Rendón-Garavito, A. M., Martínez Safar, I. E., Martínez D, D. A., Rendón-Garavito, C., Santiago-Villegas, F., Díaz F, M. P., De La Ossa Díaz, J. P., Caicedo Montaño, C. A., Rivera Bernal, A. L., Silva, D., & Costa da Silva, A. C. (2026). Iodine maps from photon-counting detector CT in thoracic imaging: a scoping review. European Radiology Cardiovascular and Thoracic, 1(1), Article 11. https://doi.org/10.1007/s44449-026-00006-1

Image Credits: AI Generated

DOI: 10.1007/s44449-026-00006-1

Keywords: photon-counting detector CT, iodine maps, thoracic imaging, scoping review, pulmonary embolism, chronic thromboembolic pulmonary hypertension, myocardial extracellular volume, radiation dose, dual-energy CT, diagnostic accuracy, Iodine, maps