Hepatocellular adenoma subtyping by qualitative MRI features and machine learning algorithm of integrated qualitative and quantitative features: a proof-of-concept study.

AIM: To evaluate hepatocellular adenoma (HCA) subtyping using qualitative magnetic resonance imaging (MRI) features and feasibility of differentiating HCA subtypes using machine learning (ML) of qualitative and quantitative MRI features with histopathology as the reference standard. MATERIALS AND METHODS: This retrospective study included 39 histopathologically subtyped HCAs (13 hepatocyte nuclear factor (HNF)-1-alpha mutated [HHCA], 11 inflammatory [IHCA], one beta-catenin-mutated [BHCA], and 14 unclassified [UHCA]) in 36 patients. HCA subtyping by two blinded radiologists using the proposed schema of qualitative MRI features and using the random forest algorithm was compared against histopathology. For quantitative features, 1,409 radiomic features were extracted after segmentation and reduced to 10 principle components. Support vector machine and logistic regression was applied to assess HCA subtyping. RESULTS: Qualitative MRI features with proposed flow chart yielded diagnostic accuracies of 87%, 82%, and 74% for HHCA, IHCA, and UHCA respectively. The ML algorithm based on qualitative MRI features showed AUCs (area under the receiver operating characteristic curve [ROC] curve) of 0.846, 0.642, and 0.766 for diagnosing HHCA, IHCA, and UHCA, respectively. Quantitative radiomic features from portal venous and hepatic venous phase MRI demonstrated AUCs of 0.83 and 0.82, with a sensitivity of 72% and a specificity of 85% in predicting HHCA subtype. CONCLUSIONS: The proposed schema of integrated qualitative MRI features with ML algorithm provided high accuracy for HCA subtyping while quantitative radiomic features provide value for diagnosis of HHCA. The key qualitative MRI features for differentiating HCA subtypes were concordant between the radiologists and the ML algorithm. These approaches appear promising to better inform clinical management for patients with HCA.

Can contrast-enhanced MRI with gadoxetic acid predict liver failure and other complications after major hepatic resection?

AIM: To determine whether a combination of clinical factors, the future liver remnant (FLR) ratio, and hepatic uptake of gadoxetic acid can be used to predict post-hepatectomy liver failure (PHLF) and other major complications (OMC). MATERIALS AND METHODS: Sixty-five consecutive patients who underwent pre-hepatectomy gadoxetic acid-enhanced magnetic resonance imaging (MRI) between October 2010 and December 2013 were included. The relative liver enhancement (RLE) of gadoxetic acid was calculated from regions of interest on MRI, and FLR ratios were obtained from computed tomography (CT). PHLF and OMC were defined by the International Study Group of Liver Surgery criteria and Clavien-Dindo grade of ≥3, respectively. Multivariate logistic regression modelling was performed to identify predictors of PHLF and OMC, including RLE, FLR ratio, age, sex, chemotherapy history, intra-operative blood loss, and intra-operative transfusion. RESULTS: Nine patients experienced PHLF and another nine patients experienced OMC. RLE was comparable to the FLR ratio in predicting PHLF (areas under the receiver operating characteristic [AUROC] curves, 0.665 and 0.705), but performed poorly in predicting OMCs (AUROCs, 0.556 and 0.702). Combining all clinical and imaging parameters as predictors yielded the best performing predictive models (AUROCs, 0.875 and 0.742 for PHLF and OMC, respectively). CONCLUSION: A model based on clinical parameters, the FLR ratio, and RLE of gadoxetic acid may improve pre-hepatectomy risk assessment.

MRI contrast agents for evaluating focal hepatic lesions.

Contrast-enhanced magnetic resonance imaging (MRI) has become an important tool in the detection and characterization of focal hepatic lesions especially when other investigations such as ultrasound (US) and computed tomography (CT) are inconclusive. The purpose of this pictorial review is to briefly summarize the properties of various MRI contrast agents used in hepatic imaging and to highlight their role in evaluation of focal hepatic lesions.

Effect of multislice CT technology on scanner productivity.

OBJECTIVE: In this study we analyzed the impact of multislice CT technology on scanner productivity in a tertiary care medical center. MATERIALS AND METHODS: We compared the productivity of two diagnostic CT scanners during the periods January 1 to August 31, 1999 (when both scanners had single-slice CT capability) and January 1 to August 31, 2000 (when one of these scanners was replaced with a multislice CT scanner). The scanners were used primarily for outpatients during the day shift and for inpatients during the evening shift; the demand for CT services was stable. For this analysis, we queried the hospital's radiology information system and identified the number of CT examinations performed during the two analysis periods. We also determined the examination mix, including proportion of enhanced and unenhanced examinations and the anatomic region examined, to ensure comparable patient populations. Statistical analysis was performed. RESULTS: The number of CT studies performed on the two scanners increased by 1772 (13.1%) from 13,548 (before multislice CT) to 15,320 (when multislice CT was available). The number of examinations enhanced with contrast media increased from 52% to 65%. Between 9:00 A.M. and 5:00 P.M., the number of CT examinations was similar on the single-slice scanners in the two periods (p > 0.05). However, in the period when multislice CT was available, the number of studies performed on the multislice scanner (5919) was 51.9% higher than those performed using the single-slice scanner (3896) (p < 0.0006). CONCLUSION: Using a multislice CT scanner leads to an increase in CT productivity, even though multislice studies are performed using more complicated protocols than are used on a single-slice CT scanner.

Hyperimmunoglobulinaemia E syndrome: pulmonary imaging features.

Hyperimmunoglobulinaemia E (hyper-IgE) syndrome (Job's syndrome) is a rare disease that presents with recurrent cutaneous and sinopulmonary infections, and which begins in infancy and is associated with extreme hyper-IgE. The pulmonary imaging features typically consist of recurrent alveolar lung infections, pneumatocoeles and, occasionally, pneumothorax. Various other causes of pneumatocoeles in children can be excluded on the basis of clinical history and other data. Computed tomography may add valuable information for the management of these patients.