What’s New in Photon Counting CT?

August 12, 2026 · 8 articles · 11 research themes · covering August 05, 2026 – August 12, 2026

Overview

Across this week’s set of studies, photon-counting CT (PCCT) emerges as a platform for more reliable quantitative imaging—spanning from physics/engineering validation to clinical translation. Method-focused work shows that PCCT-derived quantitative outputs (virtual monoenergetic attenuation and effective atomic number) can be sensitive to acquisition settings such as tube voltage, image-quality level, and pitch, underscoring the need for protocol standardization. Complementing this, ultra-high-resolution PCCT investigations highlight how small-pixel acquisition and iterative reconstruction interact with “bucket effects,” guiding parameter choices that preserve image fidelity without unnecessary dose escalation.

On the computational and workflow side, multiple papers point to AI as a key enabler for turning PCCT’s spectral potential into routine practice. An end-to-end differentiable PCCT framework integrates differentiable material decomposition into the reconstruction chain, enabling joint optimization for quantitative spectral outputs rather than relying on fixed post-processing. In parallel, a neuroradiology feasibility study uses discriminative and generative AI to automate eligibility/routing decisions between PCCT and conventional energy-integrating detector CT, addressing the operational bottleneck of manual selection.

Clinically, the dominant theme is improved localization and characterization through spectral/quantitative capabilities. PCCT is evaluated for preoperative localization of abnormal parathyroid tissue in primary hyperparathyroidism, with the promise of better performance in ectopic or difficult-to-detect cases. Cardiovascular applications include photon-counting CT iodine maps for late iodine enhancement in myocardial scar assessment, where quantitative image-quality evaluation and reader confidence improve. Finally, translational biomarker work uses spectral photon-counting CT to visualize gold-labeled monocyte uptake in human atherosclerotic plaques ex vivo, supporting PCCT’s role in studying inflammatory trafficking and potentially informing future monitoring strategies.

---

Photon-Counting CT (PCCT) Technology & Quantitative Imaging

Imaging in France: 2026 Update.

This narrative update summarized recent advances in radiology research and clinical practice in France, with emphasis on dual-energy and photon-counting computed tomography and emerging artificial intelligence applications. The key finding is that French research groups are actively developing and translating photon-counting CT techniques and AI-driven workflows to improve diagnosis and interventional radiology outcomes. Scientifically and clinically, the review highlights where photon-counting CT and AI are converging to advance imaging performance and patient care.

Talabard MP, Greffier J, Calame P et al. · Canadian Association of Radiologists journal = Journal l’Association canadienne des radiologistes · (2026) · View on PubMed ↗

---

Acquisition Protocol Optimization (Dose, Resolution, Stability)

Small-pixel Acquisition Unmasks “Bucket Effects” of Iterative Reconstruction in Ultra-High-Resolution Photon-Counting CT.

This cadaveric study tested how small-pixel acquisition affects “bucket effects” and image quality under iterative reconstruction in ultra-high-resolution photon-counting CT (UHR PCCT) of the midface. The key finding was that using smaller detector pixels (UHR: 120×0.2 mm vs standard collimation: 144×0.4 mm) changes signal-to-noise behavior and subjective image quality, revealing reconstruction-related bucket-effect behavior. Scientifically, it informs how to optimize PCCT detector pixel size and reconstruction settings to improve ultra-high-resolution image fidelity without increasing dose.

Huflage H, Wech T, Patzer TS et al. · Academic radiology · (2026) · View on PubMed ↗ · Free PDF ↗

---

Image Quality & Quantitative Reproducibility (Phantom/Methodology)

Impact of acquisition parameters in photon-counting computed tomography on the stability of virtual monoenergetic attenuation values and effective atomic number: a phantom study.

This phantom study examined how photon-counting CT acquisition parameters affect the stability of virtual monoenergetic attenuation values and effective atomic number (Zeff) in abdominal organs. The key finding was that variations in tube voltage, image-quality level, and pitch significantly influence the reproducibility of VMI CT numbers and Zeff derived from PCCT. These results are important for standardizing PCCT protocols so that quantitative tissue characterization metrics remain reliable across different acquisition settings.

Feng T, Yang C, Yang J et al. · European journal of radiology · (2026) · View on PubMed ↗

---

Clinical Applications: Cardiovascular (Myocardial Scar & Atherosclerosis)

Photon-counting CT iodine maps for late iodine enhancement: a retrospective feasibility study of image quality and reader confidence.

This retrospective feasibility study assessed photon-counting CT (PCD-CT) iodine maps for late iodine enhancement (LIE) in patients undergoing coronary CT angiography for ischemic-type myocardial scar evaluation. The key finding was that PCD-CT-derived iodine maps enabled quantitative image-quality assessment and improved reader confidence for LIE imaging compared with the limitations of conventional late iodine enhancement CT contrast. Clinically, this supports photon-counting CT as a potentially more robust approach for myocardial scar characterization after coronary CTA, potentially reducing validation barriers for LIE protocols.

Hartung V, Gruschwitz P, Serfling JK et al. · European radiology experimental · (2026) · View on PubMed ↗ · Free PDF ↗

---

Clinical Applications: Parathyroid Imaging & Surgical Planning

Photon-Counting Detector 4D CT for Preoperative Localization in Primary Hyperparathyroidism.

This study evaluated photon-counting detector 4D CT (PCD-CT) for preoperative localization of abnormal parathyroid tissue in 36 patients with primary hyperparathyroidism undergoing parathyroidectomy. PCD-CT was assessed against available ultrasound and Tc-99m sestamibi scintigraphy (available in 34 and 33 patients, respectively) to determine its localization performance, including in cases where conventional imaging is limited by ectopic disease. If validated, PCD-CT could improve minimally invasive parathyroidectomy planning by more reliably localizing ectopic or difficult-to-detect parathyroid lesions than ultrasound or sestamibi alone.

Huls SJ, Burkett BJ, Yu L et al. · AJNR. American journal of neuroradiology · (2026) · View on PubMed ↗

---

AI for Radiology Workflows & Decision Support

Automatic Patient Eligibility for Photon-counting CT Using Discriminative and Generative AI Models in Neuroradiology.

This retrospective study developed an AI-based system to automatically determine patient eligibility for photon-counting CT (PCCT) versus conventional energy-integrating detector (EID) CT in neuroradiology. The key finding was that discriminative and generative AI models using neuroradiology CT request information could automate routing decisions to the most appropriate CT technology, addressing the impracticality of manual selection. If implemented clinically, this could improve workflow efficiency and ensure patients receive PCCT when it is most beneficial for spectral/quantitative neuroradiology needs.

Martín-Noguerol T, López-Úbeda P, Escartín J et al. · Academic radiology · (2026) · View on PubMed ↗

---

AI for PCCT Reconstruction/End-to-End Learning

End-to-End Differentiable Photon Counting CT.

This work proposed an end-to-end differentiable photon-counting CT framework that integrates differentiable material decomposition into the imaging chain. The key finding is that by making the maximum-likelihood estimation (MLE)-based material decomposition differentiable (via the implicit function theorem) and inserting it as a layer, the entire PCCT pipeline can be optimized jointly with upstream models using cross-domain learning. This is significant because it enables training and optimization that directly targets quantitative spectral outputs (material decomposition) rather than treating them as fixed post-processing steps.

Wang S, Yang Y, Lee J et al. · IEEE transactions on medical imaging · (2026) · View on PubMed ↗

---

Imaging Biomarkers & Molecular/Cellular Tracking

Tracking the infiltration of gold-labelled monocytes into live human atherosclerotic plaques: Spectral photon counting CT imaging analysis.

This study investigated monocyte infiltration into live human atherosclerotic plaques by labeling human peripheral-blood monocytes with gold nanoparticles and tracking their uptake using spectral photon-counting CT (SPCCT) imaging. The key finding was that gold-labeled monocytes could be visualized and quantified in excised carotid plaque samples using SPCCT, with uptake corroborated by microwave plasma atomic emission spectroscopy (MP-AES). This provides a human ex vivo imaging approach to study plaque biology and monocyte trafficking, supporting translational development of spectral photon-counting CT for monitoring inflammatory processes in atherosclerosis.

Holmes D, Moghiseh M, Healy J et al. · Atherosclerosis · (2026) · View on PubMed ↗

---

---

Generated automatically on August 12, 2026. Covers PubMed articles published August 05, 2026 – August 12, 2026. Summaries are AI-generated; always consult the original publication for clinical or research decisions.