Measuring Identification and Quantification Errors in Spectral CT Material Decomposition

Abstract

Material decomposition methods are used to identify and quantify multiple tissue components in spectral CT but there is no published method to quantify the misidentification of materials. This paper describes a new method for assessing misidentification and mis-quantification in spectral CT. We scanned a phantom containing gadolinium (1, 2, 4, 8 mg/mL), hydroxyapatite (54.3, 211.7, 808.5 mg/mL), water and vegetable oil using a MARS spectral scanner equipped with a poly-energetic X-ray source operated at 118 kVp and a CdTe Medipix3RX camera. Two imaging protocols were used; both with and without 0.375 mm external brass filter. A proprietary material decomposition method identified voxels as gadolinium, hydroxyapatite, lipid or water. Sensitivity and specificity information was used to evaluate material misidentification. Biological samples were also scanned. There were marked differences in identification and quantification between the two protocols even though spectral and linear correlation of gadolinium and hydroxyapatite in the reconstructed images was high and no qualitative segmentation differences in the material decomposed images were observed. At 8 mg/mL, gadolinium was correctly identified for both protocols, but concentration was underestimated by over half for the unfiltered protocol. At 1 mg/mL, gadolinium was misidentified in 38% of voxels for the filtered protocol and 58% of voxels for the unfiltered protocol. Hydroxyapatite was correctly identified at the two higher concentrations for both protocols, but mis-quantified for the unfiltered protocol. Gadolinium concentration as measured in the biological specimen showed a two-fold difference between protocols. In future, this methodology could be used to compare and optimize scanning protocols, image reconstruction methods, and methods for material differentiation in spectral CT.

Clinical Utility of Multi‐Energy Spectral Photon‐Counting Computed Tomography in Crystal Arthritis

“…Multi‐energy spectral photon‐counting CT (SPCCT) is a novel imaging technique which uses a standard polychromatic x‐ray source and photon‐counting detector that records the number and energy of transmitted photons in multiple energy bins (), providing a range of energy‐dependent Hounsfield units (). As x‐ray attenuation of each material is energy dependent, multi‐energy bin data allow specific identification and quantification of several materials simultaneously (). Our previous findings show that multi‐energy SPCCT can differentiate MSU, CPP, and HA in vitro ().…”

Spectral Photon-Counting Molecular Imaging for Quantification of Monoclonal Antibody-Conjugated Gold Nanoparticles Targeted to Lymphoma and Breast Cancer: AnIn VitroStudy

“…During the course of data analysis, we observed hidden K-edge phenomenon [50], which could be associated with low concentrations of AuNPs uptake by Raji and SKBR3; no enhancement of attenuation in energy bin 4 (Figures 2(c) and 2(d)) [50, 51]. However, our material decomposition algorithm [50] was able to recover this hidden information using the effective linear attenuation for each material (for each concentration and energy bin), which was estimated by taking the mean of respective regions in the reconstructed data, and demonstrates the uniqueness of spectral CT imaging [20].…”

“…The assessment of the reconstructed images involved measuring the linear attenuation for each material and converting linear attenuation into Hounsfield units (HU). Using an in-house programme that uses the calculated effective mass attenuation of the calibration vials [20, 39], material decomposition (MD) was applied to the energy images and quantification of the trastuzumab and rituximab was performed by measuring the amount of gold in the cell clumps.…”

“…As an emerging revolution in X-ray-based imaging, spectral molecular CT promises to complement the existing molecular imaging modalities and to be a potential tool for delivering personalised medicine [6–10]. Preclinical results with spectral molecular CT have been very encouraging and indicate that the specific identification and quantification of tissue types and nanoparticles is possible, such as imaging of vulnerable plaque [11, 12], soft tissue quantification [13], reduction in metal-related CT artefacts [14, 15], crystal-induced arthropathies [16, 17], quantitative imaging of excised osteoarthritic cartilage [18], and K-edge imaging of high-Z (atomic number) biomedical nanoparticles [19, 20].…”

Effect of Reconstruction Algorithm on the Identification of 3D Printing Polymers Based on Hyperspectral CT Technology Combined with Artificial Neural Network

“…Afterwards, Si-Mohamed Salim et al further demonstrated that spectral CT has the capability of splitting two blended heavy metal-based contrast agents by qualitative and quantitative analysis in K-edge imaging [23]. Aamir Younis Raja et al proposed the MARS MD algorithm quantifying material identification at multiple material concentrations by switching reconstructed energy bins into sparse material images [24]. As a result, spectral CT has already gained constant attention due to its capacity of tissue characterization, damage or lesion detection, and material decomposition.…”