Abdominal applications of ultrasoundfusion imaging technique: liver, kidney,and pancreasEuropean Society of Radiology (ESR)
Abstract
Fusion imaging allows exploitation of the strengths of all imaging modalities simultaneously, eliminating or minimizingthe weaknesses of every single modality. Ultrasound (US) fusion imaging provides benefits in real time from both thedynamic information and spatial resolution of the normal US and the high-contrast resolution and wider field of viewof the other imaging methods. US fusion imaging can also be associated with the use of different ultrasoundtechniques such as color Doppler US, elastography, and contrast-enhanced US (CEUS), for better localization andcharacterization of lesions. The present paper is focused on US fusion imaging technologies and clinical applicationsdescribing the possible use of this promising imaging technique in the liver, kidney, and pancreatic pathologies.
� Fusion imaging helps in the detection andlocalization of lesions with low conspicuity onstandard B-mode US.
� US fusion imaging can also be associated with theuse of different ultrasound techniques such as colorDoppler US, elastography, and contrast-enhancedUS (CEUS).
� The current principal use of US fusion imaging isduring hepatic interventional procedures. However,new applications in both intra- and extra-abdominalareas are emerging more and more.
IntroductionTaking advantage of various imaging techniques to im-prove both diagnosis and interventional procedures hasbecome a very common process and is an integral part ofthe work of the modern radiologist. Normally, the associ-ation process is a mental act that involves the integrationof information coming from multiple imaging methodssuch as ultrasound (US), computed tomography (CT),
magnetic resonance (MR), and positron emission tomog-raphy (PET).During interventional procedures, this process is re-
ferred to as “cognitive fusion” or “visual registration”and consists of the careful studying of an examinationacquired before the procedure, usually CT or MR, andthe subsequent use of US as a guide for the performanceof the procedure, after mental superimposition of thespatial information from the prior study [1]. However,this process can be difficult if the ideal US scanningplane is different to the classical orthogonal CT or MRimage. Moreover, breathing and displacement and de-formation of the abdominal structures due to pressurefrom the US probe can affect the process of mentalregistration [2].Thanks to the recent improvements of technology and
computing power, a real-time computerized fusion ofradiological images has been developed and imple-mented in modern high-end US machines, to allow syn-chronous association of US images with one or moreother cross-sectional studies such as CT, MR, or PET,which are instantly reconstructed in the correspondingplane.US guidance is still the guidance method of choice for
percutaneous interventional procedures, as it providesreal-time imaging, does not use ionizing radiation, is* Correspondence: [email protected]
easily accessible, and is cheap [3]. However, compared toCT and MR, it has lower contrast resolution and a nar-rower field of view and is affected by the presence of gasand fat. The use of real-time fusion imaging allows ex-ploitation of the strengths of all imaging modalities sim-ultaneously, eliminating or minimizing the weaknessesof every single modality. Therefore, US fusion imagingprovides benefits in real time from both the dynamic in-formation and spatial resolution of the normal US andthe high-contrast resolution and wide field of view ofthe other imaging methods. US fusion imaging can alsobe associated with advanced US imaging techniquessuch as color Doppler US, elastography, and contrast-enhanced US (CEUS), for better localization andcharacterization of lesions to be treated [4].
Fusion imaging technologyThere are several available spatial tracking methods forUS probes, including optical, image-based, and electro-magnetic tracking [1]. The electromagnetic tracking sys-tem is the one mostly used for percutaneousinterventional procedures. It comprises a magnetic fieldgenerator, located 20–30 cm from the patient, and a pos-ition sensor attached to the probe, or integrated into theneedle. When the position sensor is moved in the mag-netic field, an induced electric current is generated,allowing the system to recognize its 3D spatial positionand orientation.The image fusion procedure begins with the import-
ation of data from a previous CT/MR/PET exam. Nextis the planning phase, which consists of several steps tostudy the target lesions and the structures involved inthe procedure and to establish the spatial orientation ofthe patient with respect to the probe. To do this, bothanatomical landmarks and external markers can be used.
Using anatomical landmarks alone, synchronization ofthe US images with CT/MR must be manual and re-quires the identification of motionless anatomical struc-tures on the US (e.g., vessels, cysts, calcifications) thatare then manually matched on the tomographic exam[5]. If external markers are used, placed on the patient’sskin during the CT acquisition phase, the image coup-ling process will be automatic, faster, and more reliable.When image matching is complete, real-time US and
CT/MR/PET images are arranged side-by-side or overlaidon the US monitor, displaying the same plane and movingsynchronously together (Figs. 1, 2, 3 and 4). Thus, fusionimaging helps in the detection and localization of lesionswith low conspicuity on standard B-mode US [4]. It is alsopossible to indicate the desired needle route, which canthen be followed easily during the procedure.There are many applications of fusion imaging, but
given the relative novelty of the technology, most arestill under investigation and require additional clinicaltrials. The current principal use of US fusion imaging isduring hepatic interventional procedures. However, newapplications in both intra- and extra-abdominal areasare emerging more and more.
LiverUS is the method of choice for the interventional percu-taneous approach to hepatic lesions. Its advantages inthis area are manifold, both for diagnostic proceduressuch as biopsies and therapeutic procedures such as ab-lations. Firstly, it is a real-time method, allowing the liverto be constantly followed during respiratory movements.It allows identification of the most appropriate plane forneedle insertion; this does not have to be an axial planebut can be oriented at will according to the circum-stances. Furthermore, the use of color Doppler US
Fig. 1 CT-US fusion imaging. Treatment of very small hypervascular nodular recurrence of HCC adjacent to a previously ablated area
European Society of Radiology (ESR) Insights into Imaging (2019) 10:6 Page 2 of 6
allows us to identify major vascular structures (e.g., hep-atic arteries, portal vein branches, hepatic veins) whichshould preferably be avoided during needle insertion. Fi-nally, fused CEUS can be useful to increase the conspi-cuity of lesions to be treated and vascular structures tobe avoided [3].However, US has some weaknesses which can affect the
success of the procedure. Firstly, it is an operator-dependent method and has a lower contrast resolutionthan CT and MR. Air contained in the hollow organs or inthe biliary tract (in the case of pneumobilia) may limit theavailable acoustic window. Even air in the lung paren-chyma, bones, and calcification (e.g., gallstones) interferes
with the US. Furthermore, lesions and their margins arenot always clearly visible on the B-mode US, even afterthe administration of contrast medium, particularly in theinhomogeneous and cirrhotic livers. The most commoncauses of mistargeting during US-guided radiofrequencyablation (RFA) of hepatocellular carcinoma (HCC) areconfusion with cirrhotic nodules, poor conspicuity of thetarget lesion, and poor acoustic window [2]. Lesions lo-cated deep in the most distal sectors of the acoustic conecan be blurred and difficult to identify. Using fusion im-aging technologies, it is possible to place beside or overlayupon the US image images from modalities that do notsuffer from all these problems, such as CT and MR. In this
Fig. 3 CT-US fusion imaging. Biopsy of pancreatic neck carcinoma, hypodense on CT and hypoechoic on the US
Fig. 2 MR-US fusion imaging. Biopsy of very small liver nodule, hypointense on the late hepatobiliary phase of MR, with a final diagnosisof a metastatic lesion
European Society of Radiology (ESR) Insights into Imaging (2019) 10:6 Page 3 of 6
way, radiologists can exploit all the strengths of the differ-ent methods in a single session, increasing the safety,speed, and results of the procedure and improving theconfidence of the operator.Some hepatic tumors which can be visualized by
CT or MR cannot be seen on the US due to theirsmall size, their location, or their echogenicity. Inthese cases, fusion imaging has been proven to en-hance the conspicuity of HCC nodules and to in-crease the feasibility of percutaneous RFA of HCCsnot visible on the conventional US [4–6]. If HCCsare still not visible after fusion imaging, anatomiclandmarks surrounding the lesions can be used forcorrect needle placement [5]. Thus, with the use offusion imaging, a larger population can benefit fromUS-guided ablation procedures instead of undergoingCT-guided ablation or major surgery, which are moreharmful and expensive techniques.Fusion imaging can also reduce false-positive lesion
detection during US-guided RFA and consistently im-prove the detection of HCCs, especially when theseare smaller than 2 cm [7]. The ability of fusion im-aging to reduce false positives also applies to theevaluation of local tumor progression after RFA andTACE [8].
KidneyUS is the usual first-line imaging method for the assess-ment of the kidneys. Thanks to their retroperitoneal lo-cation, in the lumbar region below the rib cage, anexcellent acoustic window is generally available, with-out the interposition of air-containing structures orbones. Most renal lesions are incidental findings andare frequently asymptomatic. The main utility of US isthe precise discrimination of solid lesions from cystic
lesions. However, with only B-mode US, it can often bedifficult to distinguish between simple and complex orneoplastic cysts. Even in the case of solid renal lesions,there may be difficulties with the B-mode US in termsof detection and characterization. For these reasons, itis often necessary to exploit second-level methods suchas CT and MR for the study of renal lesions. CEUS isalso emerging as a useful technique to study cystic le-sions and their related septal vascularization, as well assolid lesions, during ablative procedures [9, 10]. Ininterventional procedures, fusion imaging can beextremely helpful in identifying the renal lesions to betreated, especially those with poor conspicuity onnormal B-mode US. The combination of CEUS with fu-sion imaging is effective in the classification of indeter-minate renal lesions and can also improve thecharacterization of cystic lesions [11]. Therefore, in thecase of tumors, the use of CEUS with fusion imaging al-lows minimization of the risk of treating benign lesionssurgically or the risk of missing cancer [12].When targeting renal lesions during a percutaneous
procedure, fusion imaging can help in recognizing themost appropriate part of the lesion to biopsy (especiallyfor cystic lesions) or the best position to place the elec-trodes for ablation. In the case of multiple lesions, fusionimaging allows us to distinguish the specific lesion to betreated with greater confidence, allowing the margins ofthe lesion to be more precisely distinguished. Further-more, the use of fusion imaging is valuable in determin-ing the correct path for the needle, avoiding harm tostructures such as the renal vessels, renal pelvis, adrenalglands, spleen, and colon.However, the use of fusion imaging in renal disease is
still under investigation in the literature and requiresfurther clinical studies.
Fig. 4 CT-US fusion imaging. Radiofrequency ablation planning of pancreatic neck carcinoma, hypodense on CT and hypoechoic on Doppler US(performed to highlight the peripancreatic vessels)
European Society of Radiology (ESR) Insights into Imaging (2019) 10:6 Page 4 of 6
PancreasWhen used by experienced operators, US allows thepancreas to be studied in excellent detail. As US is areal-time method, radiation-free, and can be performedat the bedside, it is an important aid for guiding pancre-atic percutaneous procedures such as ablation of lesionsor drain positioning [13]. However, since the pancreas isa retroperitoneal organ, it may be difficult to visualize itentirely if there is an interposition of hollow organs; thisis particularly true with respect to the tail, sometimescalled the “blind area” of the pancreas. CT and MR, onthe other hand, are superior to US in permitting fullvisualization of the pancreas, providing clearer demon-stration of its relationships to the delicate surroundingstructures, lying behind the colon and stomach and inclose contact with the duodenum, the portal vein andmajor arterial structures such as the aorta, the celiacaxis, and the superior mesenteric artery. For this reason,fusion imaging can be an extremely useful tool torecognize and avoid damaging these structures duringUS-guided percutaneous procedures. It has been shownthat performing drainage of pancreatic necrosis usingUS fusion imaging is superior to classic B-mode US interms of safety, efficiency, and hospitalization length andcosts [14]. US fusion imaging also allows bettervisualization of the “blind area” when it is not clearlyshown with normal B-mode US [15].Clinical indications for fusion imaging of the pancreas
can therefore be summarized as guidance for biopsy anddrainage and percutaneous treatment of pancreatic can-cer such as radiofrequency ablation or irreversibleelectroporation.Even in these circumstances, the use of fusion imaging
in pancreatic disease is not well-described in the litera-ture and requires further clinical studies to confirm itsvalidity.
Limitations of fusion imagingOne of the most challenging limitations affecting US fu-sion imaging is the risk of mistargeting a lesion. In hep-atic ablation, mistargeting normally occurs in about 2%of cases and is principally due to the small size of the le-sion or confusion with the surrounding pseudo lesionssuch as regenerative nodules. Lesions located in subcap-sular or subphrenic areas, as well as lesions with poorconspicuity, can also be missed [9].Another limitation of US fusion imaging is the need to
synchronize a static image from a CT or MR study withthe breathing motion and changing position of the pa-tient, in particular when approaching a subdiaphrag-matic organ such as the liver [10]. During the breathingcycle, the movement of the liver is complex and includestranslations and rotations. During breathing, the periph-eral regions of the liver move more widely than the
central ones, which are more fixed in position by thepresence of the hepatic pedicle. In the periphery of theliver, there are also fewer anatomical landmarks such asthe vessels. For this reason, registration error during fu-sion imaging especially affects the peripheral portions ofthe liver and patients with large respiratory movements[11]. Although retroperitoneal, registration errors mayalso affect the kidney, as it is a mobile organ subject tomovement with breathing. Registration errors occurwhen the respiratory phase of the reference examinationis different from that during image synchronization.Therefore, MR, which is normally performed in expir-ation (as opposed to CT, usually performed during in-spiration), is more comparable with the patient’sbreathing status during the interventional procedure andis therefore less associated with registration errors [12].The pancreas is a less mobile organ during breathing
and is therefore theoretically less affected by registrationerrors. However, visualization of the pancreas often re-quires that pressure be applied with the probe on theupper abdomen to displace the overlying hollow organs.This can change the relationships between abdominalstructures, which can affect the matching between theUS and CT/MR images.Finally, while fusion imaging is increasingly proving to
be a promising technology, further randomized clinicaltrials are needed to define its presumed superiority overcognitive fusion during image-guided procedures.
ConclusionUS fusion imaging is a relatively novel technique in theabdominal US panorama. Its ability to exploit all thestrengths of multiple imaging methods used together inreal time makes it a tool of great value when performinga percutaneous procedure. Increasing the confidence ofthe operator, it allows better visualization of the abdom-inal structures and more precise planning of needlepaths, avoiding delicate structures, minimizing radiationexposure, and so increasing safety and efficiency (anddecreasing cost) of these procedures.
AcknowledgementsThis paper was prepared by Mirko D’Onofrio (Member of the ESR UltrasoundSubcommittee) and Alessandro Beleù, Department of Radiology, G.B. RossiHospital – University of Verona, Verona, Italy – with the contribution fromSubcommittee Members Diana Gaitini, Jean-Michel Corréas and Adrian Brady(Chair of the ESR Quality, Safety and Standards Committee). The paper wassupported by the ESR Ultrasound Subcommittee led by Dirk Clevert (Chair ofthe ESR Ultrasound Subcommittee).It was approved by the ESR Executive Council on November 15, 2018.
Authors’ contributionsAll authors read and approved the final manuscript.
Publisher’s NoteSpringer Nature remains neutral with regard to jurisdictional claims in publishedmaps and institutional affiliations.
European Society of Radiology (ESR) Insights into Imaging (2019) 10:6 Page 5 of 6
Received: 12 December 2018 Accepted: 3 January 2019
References1. Maybody M, Stevenson C, Solomon SB (2013) Overview of navigation systems
in image-guided interventions. Tech Vasc Interv Radiol 16(3):136–143.2. Lee MW, Lim HK, Kim YJ et al (2011) Percutaneous sonographically guided
radio frequency ablation of hepatocellular carcinoma: causes ofmistargeting and factors affecting the feasibility of a second ablationsession. J Ultrasound Med 30(5):607–615.
3. Lorentzen T, Nolsøe CP, Ewertsen C et al (2015) EFSUMB guidelines oninterventional ultrasound (INVUS), part I. general aspects (long version).Ultraschall Med 36(5):E1-14.
4. Lee MW (2014) Fusion imaging of real-time ultrasonography with CT or MRIfor hepatic intervention. Ultrasonography 33(4):227–239.
5. Song KD, Lee MW, Rhim H, Cha DI, Chong Y, Lim HK (2013) Fusion imaging-guided radiofrequency ablation for hepatocellular carcinomas not visible onconventional ultrasound. AJR Am J Roentgenol 201(5):1141–1147.
6. Mauri G, Cova L, De Beni S et al (2015) Real-time US-CT/MRI image fusionfor guidance of thermal ablation of liver tumors undetectable with US:results in 295 cases. Cardiovasc Intervent Radiol 38(1):143–151.
7. Lee MW, Rhim H, Cha DI, Kim YJ, Lim HK (2013) Planning US forpercutaneous radiofrequency ablation of small hepatocellular carcinomas(1–3 cm): value of fusion imaging with conventional US and CT/MR images.J Vasc Interv Radiol 24(7):958–965.
8. Min JH, Lee MW, Rhim H et al (2014) Local tumour progression after loco-regional therapy of hepatocellular carcinomas: value of fusion imaging-guided radiofrequency ablation. Clin Radiol 69(3):286–293.
9. Meloni MF, Smolock A, Cantisani V et al (2015) Contrast enhancedultrasound in the evaluation and percutaneous treatment of hepatic andrenal tumors. Eur J Radiol 84(9):1666–1674.
10. Gulati M, King KG, Gill IS, Pham V, Grant E, Duddalwar VA (2015) Contrast-enhanced ultrasound (CEUS) of cystic and solid renal lesions: a review.Abdom Imaging 40(6):1982–1996.
11. Rübenthaler J, Paprottka KJ, Marcon J, Reiser M, Clevert DA (2016) MRI andcontrast enhanced ultrasound (CEUS) image fusion of renal lesions. ClinHemorheol Microcirc 64(3):457–466.
12. Helck A, D'Anastasi M, Notohamiprodjo M et al (2012) Multimodalityimaging using ultrasound image fusion in renal lesions. Clin HemorheolMicrocirc 50(1–2):79–89.
13. Qiu W, Sun X, Wei F, Wang G, Ye J, Lv G (2016) Clinical study of B-modeultrasound-guided retroperitoneal and abdominal catheter treatment ofsevere acute pancreatitis. Minerva Chir 71(1):25–30.
14. Zhang H, Chen GY, Xiao L et al (2018) Ultrasonic/CT image fusion guidancefacilitating percutaneous catheter drainage in treatment of acutepancreatitis complicated with infected walled-off necrosis. Pancreatology18(6):635–641.
15. Sumi H, Itoh A, Kawashima H et al (2014) Preliminary study on evaluation ofthe pancreatic tail observable limit of transabdominal ultrasonographyusing a position sensor and CT-fusion image. Eur J Radiol 83(8):1324–1331.
European Society of Radiology (ESR) Insights into Imaging (2019) 10:6 Page 6 of 6
