RESUMEN
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.
Asunto(s)
Carcinoma Hepatocelular , Neoplasias Hepáticas , Carcinoma Hepatocelular/diagnóstico por imagen , Medios de Contraste , Humanos , Neoplasias Hepáticas/diagnóstico por imagen , Aprendizaje Automático , UltrasonografíaRESUMEN
Location of small gastric or colorectal tumors during a laparoscopic procedure is often imprecise and can be misleading. There is a real need for a compatible and straightforward tool that can be used intraoperatively to help the surgeon in this regard. We emphasize in the present work on the fabrication of a new and innovative inductive proximity switch architecture, fully compatible with laparoscopic surgery and with direct application in precise localisation of bowel tumors. An electromagnetic detection probe optimized for laparoscopic surgery and preconditioned for sterilisation was designed and constructed. Various metallic markers designed to be attached to the gastrointestinal mucosa were used for detection by the probe, from standard endoscopic and laparoscopic haemostatic clips to other custom made tags. Experiments were performed in dry and wet-lab experimental laboratory environment using ex-vivo segments of calf's small bowel and colonic surgical specimens from human patients. The dry-lab detection range varied considerably depending on the metallic component of the tags, from 0.5â¯mm for the endoscopic hemostatic clip to 3.5â¯mm for the 0.9â¯mm thickness stainless-steel custom tags. The latter was actually detectable from the serosal side of the fresh colonic surgical specimens in 85% of the attempts if the scanned area was less than 150â¯cm2 and less than 2â¯mm of fat was interposed between the probe and the bowel. The newly designed system has the potential to discover metallic tags attached to the bowel mucosa for precise intraoperative laparoscopic location of digestive tumors. Further work is in progress to increase the sensitivity and detection range of the system in order to make it fully compatible with the clinical use.