Junji Shiraishi, Presenter: License agreement, Riverain Medical License agreement, MEDIAN Technologies Research support, Riverain Medical Research support, Totoku Electric Co, Ltd

Katsutoshi Sugimoto, Abstract Co-Author: Nothing to Disclose

Fuminori Moriyasu MD, PhD, Abstract Co-Author: Nothing to Disclose

Naohisa Kamiyama PhD, Abstract Co-Author: Employee, Toshiba Corporation

Kunio Doi PhD, Abstract Co-Author: Shareholder, Hologic, Inc License agreement, Hologic, Inc License agreement, Deus Technologies, LLC License agreement, Riverain Medical License agreement, Mitsubishi Corporation License agreement, MEDIAN Technologies License agreement, General Electric Company License agreement, Toshiba Corporation Research support, Deus Technologies, LLC Research support, DuPont Research support, Elbit Medical Imaging Ltd Research support, Fuji Photo Film Co, Ltd Research support, General Electric Company Research support, Hitachi, Ltd Research support, Eastman Kodak Company Research support, Konica Minolta Group Research support, Mitaya Manufacturing Co, Ltd Research support, Mitsubishi Corporation Research support, Koninklijke Philips Electronics NV Research support, Hologic, Inc Research support, Riverain Medical Research support, Seiko Corporation Research support, Siemens AG Research support, 3M Company Research support, Toshiba Corporation

We developed a computer-aided diagnostic (CAD) scheme for classifying focal liver lesions into liver metastasis, hemangioma, and three histological differentiation types of hepatocellular carcinoma (HCC), by use of physicians' subjective pattern classifications and computerized image features in B-mode and micro flow imaging (MFI) of contrast-enhanced ultrasonography.

We used 214 cases in this study, of which 223 focal liver lesions consisted of 58 metastases, 53 hemangiomas and 112 HCCs: 36 well differentiated (w-HCC), 56 moderately differentiated (m-HCC), and 20 poorly differentiated (p-HCC). Pathologies of all cases were determined based on biopsy or surgical specimens. The ultrasound equipment used was SSA-790A (Toshiba Medical Systems). MFI was obtained with the contrast-enhanced low mechanical index (MI) pulse subtraction imaging at the fixed plane which included a distinctive cross section of the focal liver lesion. In the MFI, the inflow high signals in the plane, which were due to the vascular patterns and the contrast agent (Microbubble: Sonazoid), were accumulated following a flash scanning with a high MI ultrasound exposure. In our computerized scheme, single B-mode image and the original, vessel-enhanced and temporally subtracted frame images of MFI were used for extracting image features which were related to the vascularity of focal lesions, replenishment patterns and the flow of the contrast agent. Total of 17 image features were used for a cascade of 4 artificial neural networks in order to classify focal liver lesions.

Our preliminary results indicated that the sensitivities for classification of focal liver lesions were 92.5% for metastasis, 96.7% for hemangioma, and 98.7% for all HCCs. In addition, in terms of classification for histological differentiations, the sensitivities of 76 HCCs were 79.2% for w-HCC, 80.6% for m-HCC, and 87.5% for p-HCC. Therefore, overall accuracy for classifying focal live lesions into 5 categories was 87.7%.

The CAD scheme for classifying focal liver lesions by use of B-mode and contrast-enhanced ultrasonography has a potential to improve diagnostic accuracy in histological diagnosis of HCCs and other liver diseases.

The histological diagnosis of liver lesions can be estimated by use of the CAD scheme for contrast-enhanced ultrasonography.

Shiraishi, J, Sugimoto, K, Moriyasu, F, Kamiyama, N, Doi, K, Computer-aided Diagnosis for Classification of Focal Liver Lesions by Use of B-mode and Contrast-enhanced Ultrasonography.  Radiological Society of North America 2008 Scientific Assembly and Annual Meeting, February 18 - February 20, 2008 ,Chicago IL. http://archive.rsna.org/2008/6012802.html