Related Experiment Video
Updated: Aug 21, 2026

Hybrid µCT-FMT imaging and image analysis
Published on: June 4, 2015
Photon-Counting CT With Low-keV Virtual Monoenergetic Images for Detecting Muscle Invasion in Bladder Cancer: A
Wei Wang1, Jiangkai He1, Jingyun Wu1
1Department of Radiology, Peking University First Hospital, No. 8, Xishiku Street, Xicheng District, Beijing 100034, China.
None:
Background: CT is limited for local staging of bladder cancer due to insufficient soft-tissue contrast. Photon-counting detector (PCD) CT with low-keV virtual monoenergetic images (VMIs) provides greater lesion contrast with acceptable image noise and high resolution and may aid assessment for muscle invasion. Objective: To compare the performance of low-keV VMIs and conventional polychromatic images for diagnosing muscle invasion on PCD CT in patients with bladder cancer. Methods: This prospective study included patients with pathologically confirmed bladder cancer who underwent preoperative PCD CT between August 2025 and March 2026. VMIs at 40, 50, 60, and 70 keV and T3D images (representing conventional polychromatic images) were reconstructed from portal venous-phase acquisitions. Two radiologists assessed objective and subjective image quality measures to identify an optimal VMI reconstruction. Three radiologists independently reviewed the optimal VMI and T3D reconstructions to assess muscle invasion using an algorithm based on bladder wall continuity at the lesion base, incorporating ancillary features (stalk, inner-layer enhancement, tumor size ≤1 cm) in equivocal cases. Diagnostic performance was assessed using transurethral resection or cystectomy as the reference. Results: The analysis included 110 patients (81 male, 29 female; median age, 63 years); 27 had muscle-invasive disease. Among VMI reconstructions, 40 keV was selected as the optimal energy level based on highest tumor SNR, tumor CNR, subjective tumor conspicuity, and subjective tumor-bladder wall interface delineation. Accuracy, sensitivity, and specificity for muscle invasion for reader 1 were 86.4%, 96.3%, and 83.1% for 40-keV VMIs and 66.4%, 96.3%, and 56.6% for T3D; for reader 2 were 87.3%, 88.9%, and 86.7% for 40-keV VMIs and 71.8%, 85.2%, and 67.5% for T3D; and for reader 3 were 80.9%, 88.9%, and 78.3% for 40-keV VMIs and 65.5%, 85.2%, and 59.0% for T3D. For all readers, accuracy and specificity were higher for 40-keV VMIs than for T3D (p<.001), whereas sensitivity was not significantly different (p>.99). Conclusion: In patients with bladder cancer undergoing PCD CT, diagnostic performance for muscle invasion was greater for 40-keV VMIs than for conventional polychromatic images. Clinical Impact: PCD CT with low-keV VMIs may have a complementary role for preoperative local staging of bladder cancer.

