Computer-assisted planning for percutaneous ethanol injection of hepatocellular carcinoma

ABSTRACT

Percutaneous ethanol injection of hepatocellular carcinoma is indicated for the treatment of small tumours superimposed on liver cirrhosis. It involves destroying the cancer cells of the liver by injecting ethanol using a needle placed in the centre of the tumour under ultrasound or computed tomography scanner. A precise pre-operative planning of such intervention is a major challenge for a successful intervention. It requires an accurate estimation of the injected dose and the three-dimensional tumour centre location. In this work, we propose a computer-assisted pre-operative planning based on a three-dimensional reconstruction of all involved anatomical structures in order to avoid damaging surrounding healthy tissues, dangerous probe trajectories and under-treatment of the tumour. This system consists of two main modules : A decision support module that helps physicians to make a percutaneous treatment decision and a therapeutic support one that deals with needle guidance planning once percutaneous treatment is chosen. The proposed system was evaluated on two clinical data sets and has led to promising results.

KEYWORDS:

• Computer-assisted pre-operative planning
• percutaneous ethanol injection
• hepatocellular carcinoma
• decision support system
• needle guidance
• 3D reconstruction

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes on contributors

Tarak Ben Saïd is graduated with an engineering degree in Computer Science from the National School of Computer Science (ENSI), University of Manouba (Tunisia) in 2005. He received his master degree in Computer Science in 2006. In 2011, he achieved his Ph.D. degree in Computer Science from the University of Manouba. He is currently an assistant professor in Computer Science at the National School of Electronics and Telecoms of Sfax (ENET'COM) and a research scientist at the Research Group Images and Forms (GRIFT) of CRISTAL laboratory of Tunisia. His research interests include 3D imaging, image processing and medical imaging.

Faten Chaieb graduated with an engineering degree in Computer Science from the National School of Computer Science (ENSI), University of Manouba (Tunisia) in 1998. She then went on to obtain her postgraduate degree (June 2001) and her Ph.D. degree (May 2009) in computer science also from National School of Computer Science (ENSI). She obtained her ‘habilitation degree’ (Habilitation universitaire) from the same University in June 2016. She started her career as a research engineer and member of the R & D team at the Centre for Studies and Research in Telecommunications (CERT), Tunisia (2000–2005). In fall 2005, she joined the National Institute of Applied Science and Technology (INSAT) as an assistant. From 2010 to 2018, she was Assistant Professor at INSAT, and in fall 2018 was appointed Associate Professor. Since 2000, she has been a researcher in the Images and Forms Research Group (GRIFT) at the CRISTAL laboratory, ENSI, in Tunisia. Her research interests include image processing, medical imaging, 2D/3D retrieval, shape analysis and 3D coding. More recently, she has also become interested in geometric deep learning applied to 3D human behavior understanding.

Faouzi Ghorbel is a graduate from Telecom Bretagne, France (1987). He holds a Ph.D. from the University of Rennes I (1990) and accreditation to supervise research (HDR) from the same university (1995). Formerly a professor at Telecom Lille I (1991 to 1998), he is Professor of National School of Computer Sciences of Tunis (ENSI) since 1998. He was the director of ENSI from 2014 to 2017. He is currently the head of CRISTAL Laboratory (Research Center in Image, Network, System Architecture and Multimedia) since 2007. He is also the president of the Tunisian Research Association of Sciences for Image (ARTS-PI) since 2004.

ORCID

Tarak Ben Saïd http://orcid.org/0000-0001-6206-2037