Improved detection of liver metastasis by registration and subtraction of dynamic contrast-enhanced magnetic resonance imaging

Dipartimento di Ingegneria Biomedica, Politecnico di Milano, Italy.
Conference proceedings: ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Conference 02/2007; 2007:59-62. DOI: 10.1109/IEMBS.2007.4352222
Source: PubMed


The purpose of this work was to validate a subtraction method for dynamic contrast-enhanced magnetic resonance volumes acquired at portal and delayed (3 hours) phases, to obtain a better depiction and delineation of hepatic secondary lesions. As the clinical readability of subtraction images depends on the degree of co-registration between volumes, we applied a 3D non-rigid registration procedure to realign portal and delayed-phase volumes before subtraction. Our method combines a rigid registration, based on Normalized Mutual Information maximization and a non-rigid registration, based on a multiresolution analysis performed through Wavelet Transform. The feasibility of the method was tested on thirty-five secondary lesions. Results showed that the proposed subtraction produces images of easy interpretation and improves diagnostic confidence. In particular, subtraction image of subcentimetric lesions was scored as diagnostic in the 89% of the cases.

10 Reads
  • [Show abstract] [Hide abstract]
    ABSTRACT: To compare contrast-enhanced subtraction magnetic resonance imaging (MRI) with contrast-enhanced standard MRI in assessing treatment response following loco-regional therapies for hepatocellular carcinoma (HCC). Institutional review board approval was obtained and informed consent was waived for this retrospective study. All patients were analysed from our institution's liver tumour database that had loco-regional HCC therapy and the following: (1) a contrast-enhanced MRI ≤6 weeks post-treatment, (2) an unenhanced T1-weighted high-signal treatment zone (TZ) ≥1 cm, (3) follow-up contrast-enhanced MRI performed ≥6 months post-treatment. Randomized standard and subtraction TZ datasets were independently assessed by three blinded radiology readers for either complete treatment necrosis or residual disease. The standard of reference (SOR) comprised a consensus read by two radiologists with knowledge of the follow-up MRI and all available clinical data. Statistical analyses were performed using receiver operating characteristics (ROC), t-test, and kappa statistic. Twenty-six patients (19 male and seven female patients; mean age 60 years, standard deviation 10.9 years, range 46-88 years) had a total of 45 corresponding HCCs and TZs. For ROC, the area under the curve (AUC) was 0.93 (subtraction protocol) versus 0.90 (standard protocol; p = 0.49). For the t-test, the mean reader confidence level was 4.4, 3.6, and 4.4 (subtraction protocol) versus 3, 3, and 3.7 (standard protocol; p ≤ 0.011). The kappa statistic for reader-to-SOR agreement was 0.83, 0.63, and 0.71 (subtraction protocol) versus 0.51, 0.36, and 0.64 (standard protocol). Subtraction MRI significantly improves the reader confidence level in the assessment of treatment response following loco-regional therapies for HCC.
    Clinical Radiology 01/2012; 67(7):649-55. DOI:10.1016/j.crad.2011.11.013 · 1.76 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Purpose: To evaluate the value of subtraction images when using MRI to assess liver tumors treated with percutaneous ablation. Materials and methods: Following percutaneous ablation of 35 liver tumors, two abdominal radiologists, blinded to outcomes, independently reviewed follow-up MRI examinations for tumoral enhancement suggestive of residual/recurrent tumor and rated their confidence level. After one year, the readers reviewed the same examinations with added subtraction images. Accuracy of the detection of residual/recurrent tumor and contrast-to-noise ratios (CNR; for tumoral enhancement-to-liver, tumoral enhancement-to-ablation zone, and ablation zone-to-liver) were calculated with and without subtraction images and compared using Wilcoxon signed rank test. Interobserver variability was computed using Kappa (κ) statistics. Results: Residual/recurrent tumor was present in 8 (23.5%) of 34 tumors. Accuracy of detecting residual/recurrent tumor with subtraction images and interobserver agreement (κ = 0.72, good) were better than accuracy of detecting residual/recurrent tumor and interobserver agreement (κ = 0.57, moderate) of enhanced MR images without subtraction. Mean CNR of subtraction images was significantly higher than that of enhanced MR images for tumoral enhancement-to-liver (0.2 ± 5 versus 11.6 ± 14.4, P = 0.03), tumoral enhancement-to-ablation zone (10.1 ± 12.5 versus 34.4 ± 29.4, P = 0.02), and ablation zone-to-liver (11.8 ± 13.3 versus 102.5 ± 238.4, P = 0.03). Conclusion: When using MRI, subtraction images help both detect and exclude residual/recurrent tumor following percutaneous liver ablations.
    Journal of Magnetic Resonance Imaging 02/2013; 37(2). DOI:10.1002/jmri.23827 · 3.21 Impact Factor