Dataset Information

Automatic classification of focal liver lesions based on MRI and risk factors.

ABSTRACT:

Objectives

Accurate classification of focal liver lesions is an important part of liver disease diagnostics. In clinical practice, the lesion type is often determined from the abdominal MR examination, which includes T2-weighted and dynamic contrast enhanced (DCE) MR images. To date, only T2-weighted images are exploited for automatic classification of focal liver lesions. In this study additional MR sequences and risk factors are used for automatic classification to improve the results and to make a step forward to a clinically useful aid for radiologists.

Materials and methods

Clinical MRI data sets of 95 patients with in total 125 benign lesions (40 adenomas, 29 cysts and 56 hemangiomas) and 88 malignant lesions (30 hepatocellular carcinomas (HCC) and 58 metastases) were included in this study. Contrast curve, gray level histogram, and gray level co-occurrence matrix texture features were extracted from the DCE-MR and T2-weighted images. In addition, risk factors including the presence of steatosis, cirrhosis, and a known primary tumor were used as features. Fifty features with the highest ANOVA F-score were selected and fed to an extremely randomized trees classifier. The classifier evaluation was performed using the leave-one-out principle and receiver operating characteristic (ROC) curve analysis.

Results

The overall accuracy for the classification of the five major focal liver lesion types is 0.77. The sensitivity/specificity is 0.80/0.78, 0.93/0.93, 0.84/0.82, 0.73/0.56, and 0.62/0.77 for adenoma, cyst, hemangioma, HCC, and metastasis, respectively.

Conclusion

The proposed classification system using features derived from clinical DCE-MR and T2-weighted images, with additional risk factors is able to differentiate five common types of lesions and is a step forward to a clinically useful aid for focal liver lesion diagnosis.

SUBMITTER: Jansen MJA

PROVIDER: S-EPMC6522218 | biostudies-literature | 2019

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Automatic classification of focal liver lesions based on MRI and risk factors.

Jansen Mariëlle J A MJA Kuijf Hugo J HJ Veldhuis Wouter B WB Wessels Frank J FJ Viergever Max A MA Pluim Josien P W JPW

PloS one 20190516 5

<h4>Objectives</h4>Accurate classification of focal liver lesions is an important part of liver disease diagnostics. In clinical practice, the lesion type is often determined from the abdominal MR examination, which includes T2-weighted and dynamic contrast enhanced (DCE) MR images. To date, only T2-weighted images are exploited for automatic classification of focal liver lesions. In this study additional MR sequences and risk factors are used for automatic classification to improve the results ...[more]

PMID: 31095624

Similar Datasets

Project description:ObjectivesTo compare clinical success, procedure time, and complication rates between MRI-guided and CT-guided real-time biopsies of small focal liver lesions (FLL) < 20 mm.MethodsA comparison of a prospectively collected MRI-guided cohort (n = 30) to a retrospectively collected CT-guided cohort (n = 147) was performed, in which patients underwent real-time biopsies of small FLL < 20 mm in a freehand technique. In both groups, clinical and periprocedural data, including clinical success, procedure time, and complication rates (classified according to CIRSE guidelines), were analyzed. Wilcoxon rank sum test, Pearson's chi-squared test, and Fisher's exact test were used for statistical analysis. Additionally, propensity score matching (PSM) was performed using the following criteria for direct matching: age, gender, presence of liver cirrhosis, liver lobe, lesion diameter, and skin-to-target distance.ResultsThe median FLL diameter in the MRI-guided cohort was significantly smaller compared to CT guidance (p < 0.001; 11.0 mm vs. 16.3 mm), while the skin-to-target distance was significantly longer (p < 0.001; 90.0 mm vs. 74.0 mm). MRI-guided procedures revealed significantly higher clinical success compared to CT guidance (p = 0.021; 97% vs. 79%) as well as lower complication rates (p = 0.047; 0% vs. 13%). Total procedure time was significantly longer in the MRI-guided cohort (p < 0.001; 38 min vs. 28 min). After PSM (n = 24/n = 38), MRI-guided procedures still revealed significantly higher clinical success compared to CT guidance (p = 0.039; 96% vs. 74%).ConclusionDespite the longer procedure time, freehand biopsy of small FLL < 20 mm under MR guidance can be considered superior to CT guidance because of its high clinical success and low complication rates.Clinical relevance statementBiopsy of small liver lesions is challenging due to the size and conspicuity of the lesions on native images. MRI offers higher soft tissue contrast, which translates into a higher success of obtaining enough tissue material with MRI compared to CT-guided biopsies.Key points• Image-guided biopsy of small focal liver lesions (FLL) is challenging due to inadequate visualization, leading to sampling errors and false-negative biopsies. • MRI-guided real-time biopsy of FLL < 20 mm revealed significantly higher clinical success (p = 0.021; 97% vs. 79%) and lower complication rates (p = 0.047; 0% vs. 13%) compared to CT guidance. • Although the procedure time is longer, MRI-guided biopsy can be considered superior for small FLL < 20 mm.

Dataset Information

Automatic classification of focal liver lesions based on MRI and risk factors.

Objectives

Materials and methods

Results

Conclusion

Publications

Automatic classification of focal liver lesions based on MRI and risk factors.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets