Hepatocellular Carcinoma Automatic Diagnosis within CEUS and B-Mode Ultrasound Images Using Advanced Machine Learning Methods

Abstract

Hepatocellular Carcinoma (HCC) is the most common malignant liver tumor, being present in 70% of liver cancer cases. It usually evolves on the top of the cirrhotic parenchyma. The most reliable method for HCC diagnosis is the needle biopsy, which is an invasive, dangerous method. In our research, specific techniques for non-invasive, computerized HCC diagnosis are developed, by exploiting the information from ultrasound images. In this work, the possibility of performing the automatic diagnosis of HCC within B-mode ultrasound and Contrast-Enhanced Ultrasound (CEUS) images, using advanced machine learning methods based on Convolutional Neural Networks (CNN), was assessed. The recognition performance was evaluated separately on B-mode ultrasound images and on CEUS images, respectively, as well as on combined B-mode ultrasound and CEUS images. For this purpose, we considered the possibility of combining the input images directly, performing feature level fusion, then providing the resulted data at the entrances of representative CNN classifiers. In addition, several multimodal combined classifiers were experimented, resulted by the fusion, at classifier, respectively, at the decision levels of two different branches based on the same CNN architecture, as well as on different CNN architectures. Various combination methods, and also the dimensionality reduction method of Kernel Principal Component Analysis (KPCA), were involved in this process. These results were compared with those obtained on the same dataset, when employing advanced texture analysis techniques in conjunction with conventional classification methods and also with equivalent state-of-the-art approaches. An accuracy above 97% was achieved when our new methodology was applied.

Diagnostic evaluation and ablation treatments assessment in hepatocellular carcinoma

“…The microbubbles of the contrast agent produce harmonic echoes, which are detected by the transducer. This behavior is significantly different from that of the usual US waves reflected by the tissues [ 22 ].…”

“…Within B-mode US images, HCC appears, in more advanced evolution phases, as a well-defined region, of 3–5 cm in size, being hyperechogenic and often heterogeneous, due to the interleaving of fatty cells, necrosis, fibrosis and active growth tissue (Fig. 2 ) [ 22 ]. In CEUS images, HCC appears more highlighted, due to the dense and complex vessel structure that is specific to the malignant tumors (Fig.…”

“…Ultrasound (US) is an imaging tool is cheap, noninvasive, non-irradiating and, thus, repeatable, suitable for patient disease monitoring [22][23][24][25]. Contrast-Enhanced Ultrasound (CEUS) imaging is an improved ultrasound-based technology, assuming the injection into the blood of a specific contrast agent, consisting of gas filled microbubbles [22]. The contrast agent spreads through the human body, emphasizing the vessel structure in the region of interest [22].…”

“…Contrast-Enhanced Ultrasound (CEUS) imaging is an improved ultrasound-based technology, assuming the injection into the blood of a specific contrast agent, consisting of gas filled microbubbles [22]. The contrast agent spreads through the human body, emphasizing the vessel structure in the region of interest [22]. This technology leads to the highlighting of both large vessel flows, as well as of the microcirculation, being firstly implemented for hepatic tumor pathology, for abdominal emergencies and in order to recognize various tumor types [26][27][28][29][30][31][32][33][34][35][36].…”

“…The HCC tumors are usually hyperenhanced during the arterial phase, showing washout during the portal venous and delayed phases (Fig. 4) [22]. However, in many cases, within both B-mode US images and CEUS images, HCC is hardly distinguishable from the cirrhotic parenchyma.…”