Computer-aided image analysis of focal hepatic lesions in ultrasonography: preliminary results

Purpose  To develop a computer-aided image analysis (CAIA) algorithm for analyzing US features of focal hepatic lesions and to correlate the feature values of CAIA with radiologists’ grading. Materials and methods  Two abdominal radiologists, blinded to the final diagnosis, independently evaluated sonographic images of 51 focal hepatic lesions in 47 patients: hemangiomas (n = 19), hepatic simple cysts or cystic lesions (n = 14), hepatocellular carcinoma (n = 11), metastases (n = 6), and focal fat deposition (n = 1). All images were graded using a 3- to 5-point scale, in terms of border (roundness, sharpness, and the presence of peripheral rim), texture (echogenicity, homogeneity, and internal artifact), posterior enhancement, and lesion conspicuity. Using a CAIA, texture and morphological parameters representing radiologists’ subjective evaluations were extracted. Correlations between the radiologists and the CAIA for assessing parameters in corresponding categories were computed by means of weighted κ statistics and Spearman correlation test. Results  A good agreement was achieved between CAIA and radiologists for grading echogenicity (weighted κ = 0.675) and the presence of hyper- or hypoechoic rim (weighted κ = 0.743). Several CAIA-derived features representing homogeneity of the lesions showed good correlations (correlation coefficient (γ) = 0.603∼0.641) with radiologists’ grading (P < 0.05). For internal artifact (γ = 0.469–0.490) and posterior enhancement (γ = −0.516) of the cyst and lesion conspicuity (γ = 0.410), a fair correlation between CAIA and radiologists was obtained (P < 0.05). However, parameters representing lesions’ border such as sharpness (γ = 0.252–0.299) and roundness (γ = −0.134–0.163) showed no significant correlation (P > 0.05). Conclusion  As a preliminary step in US computer-aided diagnosis for focal hepatic lesions, a CAIA algorithm was constructed with a good agreement and correlation with human observers in most US features. In addition, these features should be weighted highly when a computer-aided diagnosis for characterizing focal liver lesions on US is designed and developed.