A.
Subject Predisposition
For ex-vivo Virtual Biopsy and fusion imaging tests,
two hand-made phantoms,
one prepared with a chicken-breast and an olive with pit,
and another one prepared with four olives with pit,
were used.
The chicken-breast wanted to be a human breast tissue simulator,
while the olives with pit wanted to be solid breast mass simulators.
On the ex vivo phantoms,
both Virtual Biopsy and fusion imaging with 3DPan and 2D acquisitions were performed.
For in vivo real-time biopsy US Virtual Navigator guidance,
20 female patients with suspected breast lesions or for their follow-up (mean age = 36,
range = 22-49) underwent US examination and core biopsy.
The following core biopsy tools were used: BARD Magnum 15mm 14 gauge (Bard Biopsy Systems,
Tempe,
AZ,
USA) on 6 patients,
Hologic ATEC 12cm 9 Gauge with Suros Hologic suction unit (Hologic,
Inc.
Breast Biopsy Solutions,
Indianapolis,
IN,
USA) on 3 patients and Hospital Service Spring Biopsy needle PRECISA (Hospital Service,
Aprilia (Latina),
Italy) on 11 patients,
after signing a written informed consent (all the patients were already scheduled for breast core biopsy).
The subject was lying on the examination bed,
placing her arms above and behind her head,
in order to keep the breast as stable as possible and in order to easily reach the axilla for examination.
Fusion imaging between 3DPan acquisitions and 2D US scans was performed on 5 patients.
Tests were performed by 4 sonographers: 2 expert radiologists with 20 years of experience in breast imaging and 2 residents in radiology.
B.
Image acquisition and biopsy guidance
For all the examinations and core biopsy US guidance,
an Esaote MyLabTwice US system (Esaote S.p.A.,
Genova,
Italy),
equipped with Virtual Navigation option [20] Virtual Biopsy planning and real-time image fusion of 3D US with 2D US scans,
was employed.
Moreover,
Esaote LA923,LA523 and LA533 Linear Array Probes (LA923 and LA523 - Operating Bandwidth: 4-13 MHz; CFM-PW Frequencies: 4.5 - 5.6 - 6.3 – 7.1 MHz; LA533 - Operating Bandwidth: 3-13 MHz; CFM-PW Frequencies: 3.6 - 4.5 - 5.6 - 6.3 – 7.1 – 8.3 MHz) with different reusable tracking brackets with sensor mounted (Esaote Virtual Navigator dedicated support for LA923 and CIVCO 639-042 for LA533 - CIVCO Medical Solutions,
Kalona,
Iowa,
USA) were used.
LA923 probe has an array width of 105 mm and it was mainly used for a fast acquisition of large volumes.
LA533,
53 mm array width,
was mainly used for Virtual Biopsy core needle planning and guidance,
for small breast volumes acquisition and 2D US examination.
Moreover,
LA533 probe has a dual-possibility hand grip design,
pinch grip and palmar grip (appleprobe design),
in order to provide a neutral wrist position [21].
This resource represented an additional operator’s comfort option during long examinations and during particularly difficult to-be-biopsied patient’s breast areas.
Moreover,
the presence of a single tracking sensor on the probe enabled the usual probe handling in both pinch grip (Fig.
2) and palmar grip.
Virtual Navigator tracking of the core biopsy needle and the real-time fusion imaging between 3D and 2D US data on the US system was made possible by an electromagnetic tracking system,
consisting of a transmitter on a fixed position,
a small receiver mounted on the US probe through a dedicated support,
a small receiver properly mounted on the core biopsy device and the MCS,
which corrected possible subject’s movements,
applied on the examined target (in this case the patient’s sternum).
A twisting of the sensor cable and a blockage with plaster strips were made in order to maintain the MCS as much steady as possible.
The transmitter,
whose position is considered the origin of the reference space system and which is corrected by the data coming from the MCS,
was kept steady by a proper support,
while the position and the orientations of the US probe and of the core biopsy tool in the created 3D space were provided by the receiver units.
The electromagnetic field source tip was oriented to point the target,
the subject’s breast,
in order to address the highest intensity and the most homogeneous area of the created electromagnetic field on the US scanning area.
The magnetic field produced by Virtual Navigator electromagnetic tracking system is stronger at the transmitter site and it fades with distance from the transmitter: the magnetic field was lower than the Earth’s magnetic field at a distance of 78 cm from the transmitter,
therefore the MCS movement freedom was possible within 78 cm.
A non-metallic table was used to reduce as much as possible the interferences with the created electromagnetic field.
The MC precision test was already performed and described in a previously published study [22].
The same electromagnetic tracking system,
provided for the US probe,
was used also for the core biopsy instruments tracking.
The receiver support used for the tracking of the BARD Magnum 15 mm 14 gauge and for the Hospital Service Spring Biopsy needle PRECISA was a CIVCO VTrax Instrument Navigator (CIVCO Medical Solutions,
Kalona,
Iowa,
USA).
For the Hologic ATEC 12 cm 9 Gauge with Suros Hologic suction unit a disposable CIVCO 653-002,
Sterile ATEC vacuum-assisted breast biopsy tracking bracket was used (CIVCO Medical Solutions,
Kalona,
Iowa,
USA).
In the ex vivo tests,
the needle tracking was obtained using the CIVCO eTrax Needle Tip Tracking System (CIVCO Medical Solutions,
Kalona,
Iowa,
USA).
See Fig.
3.
All the subjects were anesthetized by a sovra-pectoral approach.
C.
3DPan,
Fusion Imaging Procedure and Core Biopsy execution
Before starting the Virtual Biopsy and 3DPan procedures,
a check of the accuracy of the electromagnetic field was performed: the same point coordinates were measured twice in two different spatial orientations by a dedicated registration pen with the electromagnetic sensor mounted in.
Accuracy lower than 0.2 cm was considered acceptable.
Virtual Biopsy,
enabled by the Virtual Navigator technology,
gave to the operator the possibility to plan the core biopsy needle path even before the insertion of the tool.
The core biopsy needle insertion was guided in plane and out of plane with proper graphical indications in both situations.
The Virtual Biopsy was used considering the single plane 2D US scan alone or considering also the fusion between the 3DPan acquisition and the 2D Us scan,
in order to enlarge the field of view and have three-dimensional view of the examined and biopsied area.
A particular visualization tool of the Virtual Biopsy,
the Intelligent Positioning system,
allowed to activate a sort of viewfinder at the level of the tip of the core biopsy needle,
in order to help the operator to reach the desired target.
The 3DPan tool,
based on the electromagnetic field positioning capabilities of Virtual Navigator technology and already employed in other clinical applications [23 ,
24],
enabled the gluing of different 3D US breast tissue volumes and the navigation within.
The operator had the possibility to use the large width array transducer (LA923) and to shift to the LA523 or LA533 probe with higher maneuverability for detailed analysis of the targets (possible lesions,
suspect echographic signs,
etc.) by simply changing the probe,
without any re-synchronization procedure between 3D and 2D views.
A thick layer of US gel (Aquasonic 100,
Parker Laboratories Inc,
Fairfield,
New Jersey,
USA) was used to ensure a complete coupling between the transducer and the examined subject’s skin,
to avoid black cones and dark areas on the US image and to prevent excessive pressure on the examined area,
in order not to change the breast tissue shape and position.
Custom color volumetric targets were placed on the acquired 2D scans directly or on the 3D US volume,
in order to identify the areas that have to be scanned and biopsied more precisely,
applying different tools for increased diagnostic confidence.
For in vivo tests,
patients underwent US guided core biopsies for suspicious breast lesions and also Elastosonography around the US-visible lesion,
Color Doppler or Power Doppler.
In ex-vivo tests,
core biopsies were performed using also Elastosonography.
3DPan reconstruction and gluing algorithm of different US volumes could work using two different processes: “Preview” made a 3D global reconstruction,
based only on the geometric and position information given by the probe position and orientation within the Virtual Navigator electromagnetic field,
while “Auto”,
in addition to the information coming from the tracking system,
performed a data analysis focused on tissue structure recognition,
in order to find the best matching among the volumes.
This could be particularly useful to compensate small movements,
due to breathing and/or little tissue compression caused by the US probe during scanning.
Major tissue deformation leads to a failure of the automatic gluing process.