MRI-based fat quantification of the liver: Is it time for commercially available products?

PURPOSE: (1) To assess the clinical applicability of commercially available solutions for MR-based quantification of the hepatic fat fraction (HFF) and (2) to compare their results with clinically established in-phase/oppose-phase (IP/OP) imaging as proposed by Dixon. METHODS: Twenty-eight patients underwent MRI examinations using multigradient-echo sequences including multi-peak modeling and T2∗ correction, IP/OP imaging and multi-echo spectroscopy with successive HFF evaluation. Histopathological examination yielded the fraction of adipose hepatocytes (fAH) and the presence of increased liver iron concentration (LIC). We correlated HFF with fAH, and assessed concordance correlations among the MR-based methods with the presence of increased LIC as a control parameter. We investigated the liver segmentation quality and overall workflow of the postprocessing solutions (Philips LiverHealth and Siemens LiverLab). RESULTS: IP/OP imaging yielded a very strong correlation (r=0.88) with fAH when excluding three cases with increased LIC. Multigradient echo imaging and multiecho spectroscopy quantifications yielded similar correlations (r=0.87…0.93) as IP/OP imaging but were insensitive to increased LIC. Visceral fat, kidney tissue and major vessels were included regularly in the segmentation. Spectroscopic fat quantification was sensitive to the inclusion of visceral fat. CONCLUSIONS: IP/OP imaging allows HFF quantification when ruling out hepatic siderosis, whereas dedicated multi-echo imaging sequences and spectroscopy show no bias for increased iron concentration. The segmentation quality and workflow of both postprocessing solutions need to be improved. Nevertheless, all solutions are able to bring MRI-based hepatic fat quantification into the clinical application. We therefore recommend commercial hepatic fat quantification tools for institutions specialised to abdominal imaging.

Analysis of Bone Scans in Various Tumor Entities Using a Deep-Learning-Based Artificial Neural Network Algorithm-Evaluation of Diagnostic Performance.

The bone scan index (BSI), initially introduced for metastatic prostate cancer, quantifies the osseous tumor load from planar bone scans. Following the basic idea of radiomics, this method incorporates specific deep-learning techniques (artificial neural network) in its development to provide automatic calculation, feature extraction, and diagnostic support. As its performance in tumor entities, not including prostate cancer, remains unclear, our aim was to obtain more data about this aspect. The results of BSI evaluation of bone scans from 951 consecutive patients with different tumors were retrospectively compared to clinical reports (bone metastases, yes/no). Statistical analysis included entity-specific receiver operating characteristics to determine optimized BSI cut-off values. In addition to prostate cancer (cut-off = 0.27%, sensitivity (SN) = 87%, specificity (SP) = 99%), the algorithm used provided comparable results for breast cancer (cut-off 0.18%, SN = 83%, SP = 87%) and colorectal cancer (cut-off = 0.10%, SN = 100%, SP = 90%). Worse performance was observed for lung cancer (cut-off = 0.06%, SN = 63%, SP = 70%) and renal cell carcinoma (cut-off = 0.30%, SN = 75%, SP = 84%). The algorithm did not perform satisfactorily in melanoma (SN = 60%). For most entities, a high negative predictive value (NPV ≥ 87.5%, melanoma 80%) was determined, whereas positive predictive value (PPV) was clinically not applicable. Automatically determined BSI showed good sensitivity and specificity in prostate cancer and various other entities. Particularly, the high NPV encourages applying BSI as a tool for computer-aided diagnostic in various tumor entities.