Abstract
This review is a brief overview of the current status and the potential role of robotics in interventional radiology. Literature published in the last decades, with an emphasis on the last 5 years, was reviewed and the technical developments in robotics and navigational systems using CT-, MR- and US-image guidance were analyzed. Potential benefits and disadvantages of their current and future use were evaluated. The role of fusion imaging modalities and artificial intelligence was analyzed in both percutaneous and endovascular procedures. A few hundred articles describing results of single or several systems were included in our analysis.
Keywords: endovascular, interventional radiology, navigation oncology, percutaneous, robotics, robotic systems
Introduction
Interventional radiology (IR) is a relatively young medical specialty, which developed following an important shift in medicine, away from traditional open procedures and towards minimally invasive therapies.
The IR field is characterized by the application of state-of-the-art technological developments. These advances have enabled real-time, minimal invasive imaging-guided treatments and increasing investments from the industry in IR.
The role of IR is increasingly aimed at performing complex procedures that may be difficult due to poor visualization, proximity to non-target organs, small size lesions or complex vascular anatomies, for example resulting from tortuous or diseased vessels, which may be narrowed or occluded from atherosclerosis or thrombosis. 1
For this reason, navigation tools, including robotic guidance systems, have the potential to simplify these procedures, improving the targeting of lesions, reducing the procedural time and radiation dose and potentially improving the outcomes. 2
The use of robotics in IR is an exciting evolution following extensive efforts in medical technology and engineering in the recent years.
The last decades have seen a great surge in development of robotic systems for use in healthcare.3–7
IR robotic systems can be used for both percutaneous and endovascular procedures. Different kind of imaging modality guidance can be used with integrated IR robotic systems, such as computed tomography (CT), magnetic resonance imaging (MRI), ultrasound (US), and fluoroscopy. IR robots, in addition receiving input by single modality imaging, can receive guidance through fused multimodality imaging. Some devices are completely integrated and are able to perform all steps of image capture, registration and fusion, and can adequately insert a needle with imaging guidance in semiautonomous autonomous operations.8,9
Other prototypes of medical robots are patient mounted and can be either CT, US, or MR compatible with multiple degrees of freedom (DOF) for device actuation.
Endovascular robotic systems represent another type of IR robotic device, capable of ensuring precise and stable catheter navigation and, moreover, reducing radiation exposure for the staff during the procedures. Endovascular robotic systems can be used in both for small caliber arteries as well as large caliber arteries. Current endovascular robotic systems consist of two main components, a patient-side mechanical device and the operator control station 10 (Table 1).
Table 1.
Features of Different IR Robotic Systems Currently Used.
| Imaging Modality | Type of Procedure | Percutaneous IR Robot | Endovascular IR Robot |
|---|---|---|---|
|
|
|
|
Widely used to define a robot's motion capabilities, the number of DOF of a robot represents the total number of independent displacements or aspects of a movement. First generation robots in IR assisted the orientation of a percutaneous needle with 2 DOF, after which robots used gained up to 6 DOF, amplifying the possibilities. 3
This review provides an overview of the current status and the potential role of robotics in IR.
Methods
A literature search of articles was performed using PubMed, including articles published in English between January 2002 and July 2022, focusing on reviews and original articles on robotics in IR and oncology.
The search strategy included the following MESH terms “robots, robotics, interventional radiology, new prospective, oncology, interventional oncology.”
No publishing date limitations were imposed. Careful consideration was made in an effort to reduce the overlapping populations in case of multiple publications from the same center. Abstracts were screened and full-text articles were obtained. The total number of articles screened was 130 and 89 were selected.
Navigational Systems
To improve ablative procedures in IR for oncology, there have been recent developments in assistance in trajectory planning and placement of the instruments with the use of navigational devices and image registration.1–3 Navigational systems include tracking systems and robotic systems. Among tracking navigation systems there are electromagnetic, optical, and laser navigation systems.
Using the tracking navigation systems, commonly used during fusion imaging, fiducial markers placed on the instruments enable the capture of the movements of the instrument, which are displayed on a screen in real-time. To register the position of the patient, other markers are placed on the skin during pre-interventional imaging.
These markers are used to link the position of the patient and the position of the instrument, in order for the trace to be visualized and planned. 4
Robot-assisted systems do not trace the instrument, but rather deliver active guidance for the placement of the probe or electrode. These robotic systems deliver active guidance probe placement, where the robotic arm moves towards the targeted position.1–3 The target is defined prior to the intervention using data on the access point, angle and depth of the probe. 4 After the probe is placed, the correct position is then confirmed by control imaging.
Tables 2, 3, and 4 summarize the main commercially available tracking and robotic navigations systems.
Table 2.
List of Main Commercially Available Tracking Navigation Systems.
| Literature Reference (Author, Name) | Device (Robot Name, Company Name, Country) | Field, Subfield (Technique, Image Guidance) | Features | Results Obtained |
|---|---|---|---|---|
| Wallach et al. (2014) 81 | CAS-One IR (CAScination AG) | CT | Stereotactic navigation system for ablation procedures. The system makes use of an optical tracking system | Studied in phantom study and compared with freehand needle placement, obtained higher accuracy of needle insertion and three times the speed of manual mode. |
| Durand et al. (2017) 82 | IMACTIS (Imactis SAS) | CT | Electromagnetic navigation system | The results of a phantom study on out-of-plane punctures with this navigation system have shown a gain in time and accuracy compared to a standard CT-guided intervention. |
| Long et al. (2019) 46 | TiRobot system (TINAVI Medical Technology Co., Ltd) | CT | Optical tracking system | In-vivo study of 91 patients, in which 66 patients were treated with percutaneous screw fixation assisted by the TiRobot robotic navigation and localization system. The operation-related trauma was reduced as well as blood loss and operation duration, compared with the control group treated with classical technique. |
Table 3.
Main Commercially Available Percutaneous Robotic Navigation Systems.
| Literature Reference (Author, Name) | Device (Robot Name, Company Name, Country) | Field, Subfield (Technique, Image Guidance) |
Features | Results Obtained |
|---|---|---|---|---|
| Levy et al. (2021) 22 | XACT ACE (XACT Robotics) | CT | 5 DOF | In-vivo study of 32 percutaneous abdominal and pelvic biopsies showed that CT-guided procedures with this robotic device have 100% success rate with higher accuracy of needle placement during biopsies (<2 mm from the target) compared to standard techniques. |
| Lonjon et al. (2016) 83 | Rosa (Zimmer Biomet) | CT | 6 DOF; has the ability to include multiple different arms—each of which is able to accommodate drill guide sleeves | Trial of 20 spinal surgery patients, of whom 10 were operated on with freehand technique and 10 with robotic technique. The results obtained showed a higher precision of pedicle screw placement by robotic-assisted technique, compared to the classical technique. |
| Beyer et al. (2016) 12 | Maxio (Perfint Healthcare) | CT | 5 DOF | Retrospective single-center study of 64 in vivo patients who
were to perform thermoablations of liver tumors. 30 patients were treated with manual guidance while 34 with robotic guidance technique. The results obtained demonstrate greater accuracy in needle placement with robotic technique (needle deviation 1.6 mm vs 3.3 mm with manual technique). In addition, both dose and procedure time are lower with robotic technique. |
| Czerny et al. (2015) 19 | iSYS (iSYS Medizintechnik GmbH) | CT | These systems use a wired joystick control, so that the radiologist can move the positioning unit in two axes (the total elongation of each axis is ± 2 cm) and rotate in two directions (the total degrees in each direction is ± 2 cm and ± 35°) to adjust the position and angles of the needle. | Kirshner wire placements in 20 cadavers, all successfully performed. Mean planning time with Syngo iGuide was 4:16 min, mean positioning time of iSYS1 was 3:35 min, and mean placement time of the K wires was 2:22 min. Mean total intervention time was 10:13 min for pedicle. |
| Engstrand et al. (2016) 84 | iSYS (iSYS Medizintechnik GmbH) | CT | Pointing device attached to a 7 DOF multifunctional passive
support arm mounted on a carbon plate placed under the
patient. This is made possible by a standard 3-pin holder and connection to a standard neuronavigation system. The main component of iSYS1 is a 4-axis robotic positioning unit (RPU). |
Studied in microwave ablation in 20 patients. The median number of antenna readjustments required was zero. No major complications were related to either the procedure or the use of high-frequency jet ventilation. The mean total patient radiation dose was 957.5 ± 556.5 mGy × cm, but medical personnel were not exposed to irradiation. |
| De Baère et al. (2022) 29 | EPIONE robotic system (Quantum Surgical) | CT | 6 DOF | Study of 21 patients for percutaneous thermal ablation (radiofrequency RFA or microwave MWA) of liver tumors. They showed that the robotic-assisted thermal ablation was feasible for 22/23 lesions (95.7%) and in 19/20 patients (95.0%), 70.8% of tumors requiring no adjustment, no adverse events were observed and rate of local tumor control was 83.3% for patients and 85.7% for tumors, at 6 months. |
| Anzidei et al. (2015) 15 | ROBIO™ EX (Perfint Healthcare Pvt. Ltd) | CT | In vivo study of 100 patients referred for CT-guided lung biopsy were randomly assigned to group A (robot-assisted procedure) or group B (conventional procedure) revealed that the robotic assistance was associated with an high diagnostic accuracy and a reduction of procedure duration and radiation dose. | |
| Welleweerd et al. (2020) 36 | KUKA MED 7 R800 (KUKA GmbH) | US | Phantom study of the efficacy of robotic 3D US breast acquisition and echo-guided breast biopsies. Although the presence of the radiologist was necessary to control the procedure, the system studied here showed that the use of the robot allows for optimal precision in the positioning of the needle in the target lesion. | |
| Knott et al. (2019) 33 | RAST), (VortxRx; HistoSonics, Inc) | US | In porcine model, a non-invasive and non-thermal focused ultrasound therapy based on histotripsy, studied renal ablation, demonstrating a complete histologic destruction of the target renal tissue while sparing the urothelium. | |
| Lim et al. (2019) 35 | Transrectal ultrasonography (TRUS) probe manipulator | US | 4 DOF | Studied in comprehensive tests, including 2 bench tests, 1 imaging test, 2 in vitro targeting tests, and an IRB-approved clinical trial on 5 patients. Preclinical tests showed that image-based needle targeting can be accomplished with accuracy on the order of 1 mm. Prostate biopsy can be accomplished with minimal TRUS pressure on the gland and submillimetric prostate deformations. All 5 clinical cases were successful with an average procedure time of 13 minutes and high targeting accuracy. |
| Melzer et al. (2005) 85 | INNOMOTION (Innomedic GmbH) | MRI | 6 DOF | All interventions of injection at sympathetic chain, sciatic nerve, and coeliac plexus were successfully completed in 16 patients, without any major adverse events except for minor side effects of increased sweating in one patient after two procedures. |
| Tilak et al. (2015) 34 | MAGNETOM (Siemens AG) | MRI | In vivo study of 88 cases with data available for image-based
accuracy analysis. Between the manual and robotic groups, there
was no statistically significant difference in this diagnosis
per subject. The robotic individual biopsy sampling time was shorter than the manual approach's (P = 0.030). Between the robotic and the manual groups, the total duration of the procedure was not statistically different (P = 0.40). |
Table 4.
Main Commercially Available Endovascular Robotic Navigation Systems.
| Literature Reference (Author, Name) | Device (Robot Name, Company Name, Country) | Field, Subfield (Technique, Image Guidance) | Features | Results Obtained |
|---|---|---|---|---|
| Rolls et al. (2014) 63 | Magellan (Hansen Medical) | Endovascular | 7 DOF | In vitro studies demonstrated potential for increased
performance catheter stability during cannulation, as well as
faster attainment of learning curve plateaus. Bilateral uterine artery embolization in a small study of 5 women, demonstrated that the ability to manoeuver using this system greatly facilitated the navigation in the internal iliac arteries, achieving a successful outcome in all five cases. |
| Mahmud et al. (2016) 50 | CorPath 200 (Corindus Vascular Robotics) | Endovascular | The RAPID (Robotic-Assisted Peripheral Intervention for
peripheral arterial Disease) study was a prospective single-arm,
single-center, open-label, non-randomized study of
robotic-assisted peripheral vascular
interventions. Peripheral arterial revascularization was performed in 20 patients and achieved success in 100% of the cases. |
Robotic Systems
As mentioned, the use of robotics in IR is applied in both percutaneous interventions as well as in endovascular procedures. 11 Specifically in IR, where the use of ionizing radiation is often integral, one of the main potential advantages of robotics is the reduction of radiation for the operator while working from a remote console, while on the patients’ side there are the benefits of reduced procedure and radiation times. Several literature reviews demonstrate how robotics provide improved results in the sense of increased precision and reduced complications related to the procedure and the patients’ recovery, while main disadvantages are related to the level of current technology in providing sensitive feedbacks, as summarized in Table 5.5–7
Table 5.
Schematic Overview of Main Advantages and Challenges of Robotics in IR.
| Related Area | Advantages | Challenges |
|---|---|---|
| Procedure | Improved precision & control; Reduced time | Technical challenges in e.g. depth perception, pressure feedback; Long pre-op/op time |
| Patient & Recovery | Reduced radiation exposure time; Fewer complications Reduced mortality | Complications due to technical challenges, e.g. tissue injury |
| Provider | Remote console access; No radiation exposure | Training costs, need for skilled staff |
Percutaneous Applications of Robotics System
Proper visualization of the target using different image guidance modalities, such as fluoroscopy, US, CT, cone-beam CT (CBCT) and MRI, is paramount to the success of the procedure.
Percutaneous needle interventions include minimally invasive procedures such as biopsies, tumor ablations and infiltrations, which can be done using different imaging guidance modalities. 5
Percutaneous CT- or CBCT-guided interventions can be used effectively for image-guided biopsy and tumor ablation.
Navigation software and robotic assistance can offer a customized option in a technically challenged biopsy or ablation target. Early phantom and clinical experience with robotic navigation systems suggest procedural accuracy, reduced procedure time, and reduced exposure to radiation compared to hand-held techniques are advantages both for patient and for operator.12–15
The main advantage of percutaneous robot CT-guided interventions is the prevention of unnecessary radiation exposure while performing precise image-guided interventions.11,16–18
Currently, two main percutaneous CT robotic systems are in use, namely the Maxio (Perfint Healthcare) and iSYS (iSYS Medizintechnik GmbH). 4 Studies have confirmed improved insertion time and accuracy, while reporting mixed results on whether a decrease in patient radiation dose was significant.4,12,13,19–22
Levy et al. evaluated the accuracy, the duration of procedure and the radiation dose using a robotic device (XACT Robotics, Ltd) in 32 percutaneous abdominal and pelvic biopsies, demonstrating that the advantage of using this device allowed not only pre-planning of the trajectory, but also re-adjustment of the trajectory without operator intervention. They demonstrated that an accurate needle targeting with <2 mm error can be achieved in patients when using a CT-guided robotic system. 22
Furthermore, it was demonstrated that the XACT robotic device is able to perform precision actions minimizing the risk of needle displacement despite respiratory movements of the patient, achieving an average accuracy of 1.7 mm from tip to target with an average procedure time from skin to target of <8.5 minutes in lung ablations. 22
In lung ablations, this allows for a more timely diagnosis and treatment for patients, reducing the “watchful waiting” period.
Thanks to its high resolution, absence of radiational risk and its capability of monitoring, investigating tissue properties and localizing lesions, MRI is of growing interest in the IR field. Navigation software and robotic assistance has been successfully used during real-time MRI procedures with various clinical applications, ranging from cryotherapy in renal cancer to transrectal prostate biopsies.23,24 The main limits of MRI-robot guided interventions are the limited space to access the patient during the procedure and the fact that the high-strength magnetic field impedes the usage of conventional, metal-based materials for the robotic devices. 16
Among the imaging techniques, US-based interventions are very accessible for its non-radiation risk and relative low cost. However, it is an operator-dependent examination with poor reproducibility. 25 To overcome this limitation, robotics also started to expand in this area with the first robot with using US-imaging, the B-Rob I, introduced by Kettenbach et al. 25 This system allows the real-time tracking of lesions and automatization of procedures with high precision and reliability.
The problem of using US-guided robots mainly concerns the positioning of the US probe and the pressure exerted by the mechanical arm to obtain reproducible images. Therefore, in a phantom study in 2018, Berger et al. 22 developed a table-mounted robotic arm model that would guarantee stability and simultaneously enable real-time US-imaging and needle placement in the target lesion. 26
For cryoablation in abdominal area, Wu et al. designed and studied in 2013 a prototype coil-mounted robot that would allow the cryoablation needle to be positioned in real-time with MR images without time-consuming manual insertion. 27
The possibility of control by the radiologist from a remote workstation makes it possible to assess the needle trajectory in real-time and ensure precise positioning.
More recently, in 2020, He et al. presented an innovative robot-integrated system under development. Their model handles an almost fully automated needle insertion under the guidance of the radiologist who directs the needle to target with real-time MR images. The real innovation lies in the possibility to target more than one lesion at a different localization at the same time, and this is enabled by the narrow size of the prototype mounted directly on the patient. This innovative device may help to perform percutaneous ablation in hepatocellular carcinoma (HCC) with multiple lesions and needle access points, while exploiting the advantages of MR images. 28 The results obtained are considered promising and further studies will be required to improve these models.
Organ-specific percutaneous robotic applications
Many phantom tests, animal experiments and in vivo studies have been employed to test robotic systems in different organ-specific fields of application (Table 6).
Table 6.
List of Robotic Systems According to Specific Organ Application.
| Organ | Ref. | System | Features | Clinical Use |
|---|---|---|---|---|
| Liver | 1) Beyer et al.
14
2) de Baère et al. 29 |
1) Maxio (Perfint Healthcare) 2) EPIONE robotic system (Quantum Surgical) |
1) 5 DOF 2) 6 DOF |
1) Microwave thermoablation 2) Thermal ablation (radiofrequency or microwave) |
| Lung | Anzidei et al. 15 | ROBIO™ EX (Perfint Healthcare) | CT System | Biopsies |
| Kidney | Knott et al. 33 | RAST (VortxRx; HistoSonics, Inc) | US System | Non-thermal histotripsy-based focused ultrasound |
| Prostate | 1) Lim et al.
35
2) Tilak et al. 34 |
1) TRUS probe manipulator 2) MAGNETOM (Siemens AG) |
1) 4 DOF, US System 2) MRI system 3 |
2) TRUS scan and biopsy 2) A 3-T wide-bore MRI biopsy |
| Breast | 1) Welleweerd et al.
36
2) Fischer et al. 24 |
1) KUKA MED 7 R800 (KUKA GmbH) 2) ROBITOM II (Friedrich-Schiller University of Jena) |
1) US System 2) MRI-System |
1) 3D US breast acquisition and echo-guided breast
biopsies 2) Biopsies and therapeutic interventions through MRI |
| Brain and Spine | 1) Lonjon et al.
83
2) Long et al. 46 3) Czerny et al. 19 |
1. Rosa (Zimmer Biomet) 2. TINAVI (Medical Technology Co.) 3. iSYS (iSYS Medizintechnik GmbH) |
1) 6 DOF 2) Optical tracking system 3) Two axes (the total elongation of each axis is ±2 cm) and two directions (the total degrees in each direction is ±2 cm) with a total degrees in each direction is ±35° |
1) Spinal surgery, pedicle screw placement 2) Percutaneous screw fixation 3) Kirshner wire placement |
Liver
The application of modern, robot-assisted CT-based navigation systems is extending to the application in percutaneous liver treatments, including both thermal and non-thermal percutaneous ablations and percutaneous biopsies in order to allow a precise placement of a needle or antenna into the target lesion.
These systems could potentially reduce the radiation dose for the patient and interventionalist as well as to a lower number of needle replacements and as a consequence to a lower complication rate. 12
Mbalisike et al. tested a robotic guidance system, Maxio (Perfint Healthcare) as a guide during microwave thermoablation and compared it with the manual approaches, which showed an improvement in the accuracy of targeting the tumor, a reduction of patient dose and an increase of procedural performance during ablation. 13
Beyer et al. compared the conventional CT-guided manual irreversible electroporation (IRE) of malignant liver tumors and a robot-assisted approach using as robotic system Maxio (Perfint Healthcare), showing a reduction of the procedural time and a significantly lower dose-length product under robotic assistance. Furthermore, the procedural accuracy, measured as the deviation of the IRE probes with respect to a defined reference probe, was significantly higher using robotic guidance. 14
de Baère et al. evaluated the feasibility and safety of EPIONE robotic system (Quantum Surgical) for percutaneous needle insertion during thermal ablation (radiofrequency or microwave) of liver tumors. They showed that robotic-assisted thermal ablation was feasible for 22/23 lesions (95.7%) and 19/20 patients (95.0%), in which 70.8% of tumors no adjustment was required, and no adverse events were observed. The rate of local tumor control was 83.3% for patients and 85.7% for tumors at 6 months. 29
Lung
Image-guided percutaneous needle biopsy has proven to be a safe and effective technique for the diagnosis of many lung diseases.
Thermoablation treatments are increasingly used in the treatment of lung tumors, particularly in slowly progressing oligometastatic tumors, where radiotherapy is not feasible.
CT and CBCT are the most widely used imaging guidance modalities for lung biopsies and thermoablative treatments. 30
Dedicated interventional robotic systems which operate under imaging guidance also became available recently in order to increase the diagnostic accuracy and reduce the duration of CT- or CBCT-guided biopsies. 15
Anzidei et al. evaluated the performance of a robotic system (ROBIO™ EX, Perfint Healthcare) for CT-guided lung biopsy in comparison to the conventional manual technique and revealed that the robotic assistance is associated with an high diagnostic accuracy and a reduction of procedure duration and radiation dose. 15
Kidney
Minimally invasive treatments are proven options for small renal lesions, offering the advantage of limiting the invasiveness, mortality and preservation of renal function. 31
One of the most challenging aspects is tumor proximity to vital local structures, such as bowel loops or ureter, which requires special attention in order to avoid major complications. 32
Robotic systems are also being applied in minimally invasive kidney procedures.
Knott et al. 33 demonstrated the feasibility of robotically assisted sonic therapy (RAST) (VortxRx; HistoSonics, Inc), a non-invasive and non-thermal focused US-therapy based on histotripsy and renal ablation in a live porcine model, and observed a complete histologic destruction of the target renal tissue while sparing the urothelium. In contrast to thermal ablation, in which its use can be limited for tumors involving the renal hilum, histotripsy, which is based on cavitation, depends on the rapid expansion and collapse of a bubble cloud to disrupt targeted tissue via focused US in a purely mechanical method of tissue destruction.
Prostate
Prostate cancer is the most common malignancy in men and the second leading cause of cancer deaths in the Unites States. 34 MRI-guided biopsy was originally proposed as an alternative to transrectal ultrasound-guided biopsy after repeated negative outcomes in diagnosis despite rising prostate-specific antigen (PSA) values. 34
Robotic devices have been introduced to overcome some of the challenges of access to MRI-guided in-bore biopsies, as well as an effort to improve the accuracy of needle placement with the aim of better diagnostic performance.
Several MRI-compatible robotic devices have been developed and applied in the clinic with the aim of improving transrectal, transperineal and transgluteal biopsies under MRI guidance. 34
In their in vivo study, Tilak et al. studied robotic biopsies in a MAGNETOM (Siemens AG), a 3-T wide-bore MRI used to perform all biopsy procedures. 34 Remote control of the robotic arm by the radiologist allows for needle insertion via MRI guidance, using 3D Slicer software to align the needle insertion hole with its transperineal positioning trajectory. Each trajectory adjustment shift is monitored by a T2-weighted axial MR image that verifies the correct trajectory of the needle. 34
Another use for MRI-guided robotic-assisted procedures in urological field relates to transperineally implantation of gold seeds. Robotic-assisted seeds’ placement is guided by MR images, thus reducing the risk of not targeting the tumor. 24 The area of the tumor is directly visualized through MRI and the robot is set to place the seed with the real-time image guidance. In this way the robot minimizes the risk of not positioning the seeds correctly, which may occur with manual US-guided positioning. 24
The most common biopsy method is freehand TRUS guided. Lim et al. 35 in 2019 evaluated a robot-assisted approach for TRUS-guided prostate biopsy; the robot they used is a TRUS probe manipulator that moves the TRUS probe with 4 DOF, faithfully replicating manual movement. They developed software capable of rendering the three-dimensional image and created a prostate coordinate plane to define the plane to be biopsied. Several comprehensive tests were performed, including a clinical trial on 5 patients. Success was achieved in all 5 cases, with millimetric lesion targeting accuracy, demonstrating just how reliable and precise a robot-assisted prostate biopsy can be.
Breast
Breast cancer is the second most commonly diagnosed cancer in the world and the most common malignancy among women. Because the incidence rate is so high, early diagnosis is crucial in cancer management.
Different imaging methods are used to detect breast cancer, including mammography, US, and MRI. Mammography is the most common imaging modality in clinical practice. If a lesion is detected, a tissue sample taken by a biopsy to confirm malignancy. In most cases, the biopsy procedure is performed under US guidance. 36 The needle insertion can be hindered by tissue margins and displacement of the lesion due to the forces exerted during needle insertion. For these reasons, the robotic assistance can be helpful in these challenges, thanks to a precise and stable manipulation.
Welleweerd et al. in 2019 36 presented an end-effector for robotic 3D US breast acquisition and US-guided breast biopsies. In this study, the KUKA MED 7 R800 (KUKA GmbH) was used as the robotic arm, which is capable of locating the breast, acquiring and reconstructing the volume in three dimensions, identifying the target and guiding the needle in all directions, except for insertion which is regulated by the radiologist who determines the correct depth of the needle and stops it. The study was done with prototypes, not in vivo, to assess the correct positioning of the needle, determining the accuracy of the biopsy. In each case, an accuracy of 0.3 ± 1.5 mm inside and 0.1 ± 0.36 mm outside the US plane was achieved when positioning the needle. The system achieved a statistically insignificant Euclidean distance error between the needle trajectory and the true target. Although the presence of the radiologist was therefore necessary to control the procedure, the system studied here showed that the use of the robot allows for optimal precision in the positioning of the needle in the target lesion. 36
Although mammography is the gold standard for breast screening, MRI is extremely sensitive in breast cancer, with sensitivity rates of around 90% and is therefore indicated in high risk patients. 36
The benefits of using MRI in conjunction with minimally invasive biopsies have also been studied in the breast cancer interventions.
Fischer et al., who designed and implemented a MRI mammography-compatible device for breast biopsy and cryoablation, the ROBITOM II (Friedrich-Schiller University of Jena). The robot allows for biopsies and therapeutic interventions through MRI localization in a better way than US- or mammography-guided procedures. Particularly with the approval of new minimally invasive treatment options such as cryoablation, radiofrequency ablation (RFA) and microwave thermotherapy, the need to develop robots capable of guiding treatment instruments with MRI images has grown.37–39
The additional advantage of MRI guidance for interventional procedures on the breast concerns the ability to directly visualize temperature changes within the lesion treated with ablation, in which ablative treatment can be adjusted in itinere. 40
Brain and spine
The ROSA (robotic surgical assistant; Zimmer Biomet) is a robotic system developed for use in spinal (e.g. vertebroplasties) and brain surgery. 27 This robot has been tested in performing stereotactic needle biopsies for diffuse intrinsic pontine glioma (DPIG), for the positioning of pedicle screws, as well as circumferential arthrodesis, demonstrating extreme accuracy in performing such procedures.41–44 Hunsche et al. demonstrated a significant reduction in exposure dose in lead localization during ROSA robot-guided deep brain stimulation. 45
Other CT-guided robotic systems in orthopedic IR are the TiRobot system (TINAVI Medical Technology Co., Ltd), which improves the accuracy in the placement of pedicle screws (like the ROSA system) compared with the traditional manual technique.46,47
Li et al. studied in cadaveric samples a body mounted robotic system for MRI-guided lumbar spine injections and showed that it is able to provide more accurate and reproducible cannula placements compared to the freehand procedure, as well as a reduction of the number of insertion attempts. 48
Endovascular applications of robotics
Around the mid-2000s, several robotic systems were developed to address the problems associated with manual endovascular catheterization, in an effort to ensure precise and stable catheter navigation and, above all, to reduce radiation exposure during the various procedures. 49
Main drawbacks of robotics in endovascular procedures are the lack of tactile and force feedback, the extensive healthcare costs for both the purchase as well as the maintenance of robotic systems, the additional training of medical staff, and the fact that they cannot be used in emergency endovascular procedures due to increased pre-intervention set-up times. 1
Large caliber arteries robot systems
The first generation of FDA-approved endovascular robotic systems is the Sensei X system (Hansen Medical), which consists of a workstation, a remote catheter manipulator located next to the patient, and a catheter (Artisan Extend Control Catheter). 19 This catheter, which has a diameter of 14 Fr, provides a certain stability however is not suitable for safe navigation in small vasculature.19,50–52 The interventional radiologist, located at the remote workstation, uses a multidirectional joystick to transmit directional information to the control station at the patient's side, moving the catheter during the procedure. 53 The Sensei robotic system was initially used in cardiac electrophysiology for the collection of data for treatment of atrial fibrillation and atrial flutter with RFA, and was later also used for endovascular aneurysms repair (EVAR) and for the treatment of post-transplant pulmonary stenosis.54–56 Currently, the Sensei system is mostly used in interventional cardiology. Further FDA-approved endovascular robotic systems in this field include the Niobe System (Stereotaxis) and the Amigo remote catheter system (RCS) (Catheter Robotics).49,57,58 The robotic systems mentioned are not compatible with the use of catheters with smaller diameters in order to be able to pass through small-caliber vessels. 58
Small caliber arteries robot systems
Further developments have been made in an effort to navigate in small vasculature and through peripheral arterial systems. 59 These next generation systems include the Magellan robotic system (Hansen Medical), the CorPath 200 (Corindus Vascular Robotics) and its successor the CorPath GRX (Corindus Vascular Robotics). The Magellan system is designed along the lines of the Sensei system. Unlike the Sensei system, the catheters used by the Magellan are smaller and increasingly flexible. 53 In particular, 3 catheters have been approved by the FDA, with maximum diameters of 10, 9, and 6 Fr, lengths that can vary from 50 to 102 cm, with 7 DOF and equipped with 180 and 90 degrees of multidirectional bending.59–62 There are several studies in the literature which have shown that the Magellan system represents a new approach for endovascular procedures, both diagnostic and therapeutic, as it can provide greater freedom of movement, especially in complex and tortuous vascular anatomy.63,64 In 2014 Rolls et al., 63 using the Magellan robotic system, performed bilateral uterine artery embolizations in a small study of 5 women. A catheter with a maximum diameter of 9 Fr (NorthStar) was used, and it was demonstrated that the ability to manoeuver greatly facilitated the navigation in the internal iliac arteries, achieving a successful outcome in all 5 cases. 63 In 2015, Rao et al. 64 evaluated the possibility of using the Magellan robotic system for the treatment of liver cancer through transarterial chemoembolization. In this study, 7 cases of transarterial chemoembolization using Doxorubicin spheres were performed, with a 100% success rate in the absence of major complications. Furthermore, the ability of the catheter to cannulate hepatic arterial vessels, the time required to perform each procedure and the radiation exposure time were evaluated. Exposure time during the entire procedure was assessed using a radiation monitoring system called RaySafe, for all healthcare personnel involved and the patient, demonstrating an average 80% reduction in radiation exposure for the operator compared to non-robotic-assisted techniques. The average time for each procedure was 1 hour and 42 minutes, while the time from cannulation to the start of embolization ranged from 25 to 88 minutes. In all 7 cases, the robotic catheter was able to cannulate the celiac tripod and the superior mesenteric artery; in 85% of cases the common hepatic artery; in 71% the right/left hepatic arteries and in 42% of cases the distal vessels. 64 Lastly, the CorPath 200 system has recently received FDA approval for vascular procedures while it had initially been solely approved for percutaneous coronary interventions. 50 In the RAPID study (robotic-assisted peripheral intervention for peripheral arterial disease), this system was used for peripheral arterial revascularization, without complications in all 20 patients enrolled. 50 Among the advantages attributed to this system is its compatibility with many guidewires and catheters. 53
Tethered versus non-tethered
Endovascular interventions are traditionally performed with tethered catheters, manually guided by the physician, visible under fluoroscopy and with limited flexibility and range of motion. 49 Tethered catheters therefore rely heavily on the interventional radiologist’s experience. Consequently, the main shortcomings are technical difficulties in terms of precision, safety and lack of force detection. Furthermore, one of the main limitations of fluoroscopy is that the operator maneuvers without a 3D visual setup. 49 With robotics certainly, the concepts of minimal flexibility and limited range of motion have been overcome, however, there are still challenges ahead. Among these, one is due to the fact that such procedures are often performed in vasculature that is diseased, tortuous, thrombotic and therefore poorly navigable, with a high rate of major complications both peri- and post-procedural. For this reason, next-generation endovascular robotic devices, capable of overcoming such limitations and achieving better results, could be untethered microdevices, meaning that these are injected wireless devices that are not manually guided in a direct manner like catheters. An exciting achievement in the field of engineering is the designs of intelligent soft-body robots that can navigate inside small vessels with high precision, however, these devices are still in the early stages of development. 1
Fusion Imaging
Fusion image navigation systems enable the use and the advantages of all imaging modes simultaneously, reducing to a minimum the disadvanteges of each individual mode. 65
There are many applications of fusion imaging, but given the relative novelty of the technology, the current principal use of US-fusion imaging is for oncologic percutaneous procedures, such as biopsies and thermal ablations. 65
In this field of application, fusion imaging allows to precisely localize the target tumor and to guide antenna placement. 66
The first step of imaging fusion the importation of data from a previous CT, MRI or PET-exams.
The second step is the spatial alignment of the imaging dataset, in which anatomical landmarks and external markers can be used. Imaging registration can be carried manually by the operator, automatically based on matching common anatomical landmarks, or semi-automatically using a combination of both techniques. Appropriate alignment of anatomical landmarks is fundamental to ensure the proper targeting. 2
When an appropriate alignment is complete, real-time US and CT/MRI/PET-images are overlaid on the US monitor, displaying the same plane and moving synchronously together. 65
US guidance provides a real-time visualization of needle placement, does not use ionizing radiation and is easily accessible. However, the limitations of US are represented by deep lesions and/or large patients with a none complete visualization of target lesion. 2
The CT and MR images are not a real-time images, but offer superior three-dimensional visualization of the needle, electrode and target.
FDG-PET/CT fusion is a tool used to characterize malignancy based on tumor metabolic activity. 67
Fontana et al. evaluated the feasibility of performing percutaneous biopsy of lung lesions guided by fusion PET/CT-CBCT, comparing patients which underwent to CBCT-guided lung biopsy and patients which underwent PET/CT-CBCT fusion-guided lung biopsy, observing comparable results in term of technical and clinical success but with better results in term of definitive histology and immunohistochemical and molecular biology analysis for the PET/CT-CBCT fusion-guided lung biopsy patients. 67
The association of the advantages of different imaging techniques by fusion imaging improve the performance during the IR procedures.
For better localization and characterization of lesions, different US techniques can be used, such as color Doppler US, elastography, and CEUS. 65
Cone-Beam CT (CBCT) and Image Guidance Softwares
CBCT allows volumetric data acquisition with a single rotation of an x-ray source and a flat-panel detector, mounted at the ends of the C-arm in the angio-suite. 68
CBCT-derived volumetric data can be merged with pre-procedural cross-sectional images and/or combined with dedicated software for needle trajectory planning and ablation volume prediction.
Navigation software such as Xperguide (Philips Allura Xper FD20, Philips Healthcare) can be used for needle trajectory planning by monitoring needle progression, using merged fluoroscopic images with the preliminary CBCT dataset.
XperCT software (Philips Allura Xper FD20, Philips Healthcare) can be used to predict ablation volume. Intra-procedural CBCT showing the antenna can be merged with pre-procedural CT/MR/PET images. A 'virtual antenna' can be located exactly, after which the preferred power and ablation time is set, while an ablation volume is automatically produced at its tip, based on the manufacturer’s data 69 (Figure 1).
Figure 1.
Fusion imaging application in a HCC nodule. Contrast-enhanced CT (CECT) on (A) arterial, (B) portal, and (C) delayed phase demonstrated the presence of a HCC nodule in patient previously submitted to thermal ablation in the 4a segment. Fusion imaging between intra-procedural US and pre-procedural CECT images (D) and (E). Fusion imaging between intra-procedural CBCT and pre-procedural CECT with predicted volume of ablation (XperCT software) on axial (F), coronal (G), and sagittal scan (H). One month follow-up CECT on arterial (I), venous (J) and delayed (K) phase demonstrate a complete response.
Ierardi et al. evaluated the efficacy of percutaneous microwave ablation for lung malignancies comparing procedures performed using CBCT with and without volume prediction software (VPS), and found residual disease in only 1 patient (10%) in the group with VPS and 3 patients (30%) in the group without VPS. 70
Monfardini et al. evaluated the efficacy of real-time US and CBCT for the percutaneous ablation of renal tumors and found a primary technique efficacy of 100%, with no recurrence at the follow-up ranging from 8 to 26 months. 68
In vascular procedures, CBCT can be used in addiction to automated tumor feeder detection (AFD) softwares, such as Emboguide (Philips Healthcare) which can enhance the efficacy of reach the target vessels both during emergency settings and non-emergency settings.
The AFD software procedure comprises three steps, in which the first step is the manual identification and segmentation of a region of interest (ROI), the second is the manual identification of tip of catheter and finally the last step is the automated identification of feeding arteries. 71
The final 3D roadmap, containing the segmented ROI and feeding arteries and the paths from catheter to vessels, has been overlaid onto the live fluoroscopy images.
Many studies have been demonstrated the efficacy in term of visualization of artery feeders and in term of reduction of radiation dose of transcatheter arterial chemoembolization (TACE) using AFD softwares.71–73
Artificial Intelligence
The term artificial intelligence (AI) commonly applies to computational technologies that mimic the intellectual processes typical of human cognitive function, such as reasoning, learning and problem solving. Machine learning (ML) is a sub-category of AI that includes the development of algorithms to analyze data to identifiy complex patterns. A subfield of ML is deep learning (DL), which involves the training artificial neural networks (ANNs) to perform computational tasks.
Radiomics is the main field of application of AI. The radiomics process provides the extraction of quantitative features from radiological images and the combination of these with clinical features and genetic/molecular models to improve the prognostic evaluation.
Many studies show the role of radiomics in terms of prognostic evaluation of recurrence after MWA of liver, lung and kidney diseases.74–79
Among the fields of application of AI, augmented reality (AR) or virtual reality (VR) can be supportive tools during the IR procedures. Through advanced 3D rendering and spatial image manipulation, AR and VR allow practitioners to conceptualize difficult anatomy, increase realism in procedural planning (compared to standard 2D images) and improve procedural skills in a previously simulated environment. 80
Discussion
Navigational systems include tracking systems and robotic systems. Among tracking navigation systems there are electromagnetic, optical and laser navigation systems. 4
The introduction of robotics in IR is considered a huge and exciting step forward in medicine and surgery, with potential great benefits especially in oncology
CT-guided robotic systems currently in use provide as a main advantage the reduction of radiation exposure. The main advantage of percutaneous robotic CT-guided interventions is the prevention of unnecessary radiation exposure while performing precise image-guided interventions.
Currently, two percutaneous CT robotic systems represent the most used systems, the Maxio (Perfint Healthcare) and iSYS (iSYS Medizintechnik GmbH). 4 Some studies have confirmed improved insertion time and accuracy and a decrease in patient radiation dose using robotic assistance-guidance.4,12,13,19–21
Thanks to its high resolution, absence of radiation risk and its capability of monitoring, investigating tissue properties and localizing lesions, MRI is growing in interest also in robotic guided procedures with various clinical applications that range from cryotherapy to transrectal prostate biopsies.23,24
The main limits of MRI-robot guided intervention are the limited space to access to the patient during the procedure, the fact that the high-strength magnetic field impedes the application of conventional, metal-based materials for the robotic systems and the long acquisition times. 16
Robotics in US-guided procedures has proved to be very helpful in increasing reliability and reproducibility. In order to overcome the limitation of ultrasound as an operator-dependent examination, the robotic assistance guidance guarantees reproducibility with real-time tracking of lesions, with the major applications currently in breast and prostate biopsies.25,35,36
The use of robots in various endovascular procedures, both for diagnostic and therapeutic purposes, has demonstrated that it can overcome some of the limitations that the interventional radiologist has to face in traditional manual procedures, for example the limited DOF in movements and excessive radiation exposure during guided procedures. 49
Despite the success obtained in the studies in the literature,19,50,59–64,51–58 which demonstrate numerous advantages over traditional manual procedures including shorter procedure times, large DOF in movements, greater precision, fewer peri- and post-procedural complications and less exposure to radiation for both operator and patient, robotic systems still struggle to be incorporated into common endovascular procedures. This can be attributed to their main limitations such as high costs, the need of training courses for operators, but more importantly the fact that such systems require quite long preparation times before each procedure, which make the devices unsuitable in emergency procedures. 1
Multimodal image fusion navigation has also demonstrated a crucial role for both percutaneous and endovascular procedures. 65
Regarding percutaneous procedures, many studies have shown the possibility to improve the localization of target, the correct placement of the needle and the prediction of ablation volumes using association of the advantages of different imaging techniques by fusion imaging, including US, CT, CBCT, RM and CT-PET and specifical softwares such us XperGuide or XperCT.66–70 In endovascular procedures, CBCT can be used in addiction to automated tumor feeder detection (AFD) softwares, such as Emboguide (Philips Healthcare) which can enhance the efficacy of reach the target vessels both during emergency setting and non-emergency setting. 71
Many studies have demonstrated efficacy in terms of visualization of artery feeders and reduction of radiation dose in TACE using AFD softwares.71–73
AI and specifically radiomics, with the extraction of quantitative features from radiological images and the combination of these with clinical and genetic/molecular models, have been improving the prognostic assessment prior to perform both percutaneous and endovascular procedures.74–79
In conclusion, robotic-guidance in IR has demonstrated important benefits for its use, with wide applications in endovascular and percutaneous procedures using CT-, MR-, US-imaging and fusion imaging. Despite demonstrating main advantages, several practical limitations have not yet been overcome to ensure their widespread application. Advances in medical engineering and technology and increased application in IR have enabled important developments and recent FDA approvals of devices are an exciting step forward. Future developments to address main limitations associated with each application could ensure broader use in the IR field.
Abbreviations
- DOF:
degrees of freedom
- DIPG:
diffuse intrinsic pontine glioma
- EVAR:
endovascular aneurysms repair
- HIFU:
high-intensity focused ultrasound
- IR:
interventional radiology
Footnotes
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding: The author(s) received no financial support for the research, authorship, and/or publication of this article.
Ethical Statement: Our study did not require an ethical board approval because it did not contain human or animal trials.
ORCID iDs: C. Lanza https://orcid.org/0000-0002-8286-1562
L. Sciacqua https://orcid.org/0000-0003-1774-0729
References
- 1.Gunduz S, Albadawi H, Oklu R. Robotic devices for minimally invasive endovascular interventions: A new dawn for interventional radiology. Adv Intell Syst. 2021;3(2):2000181. doi: 10.1002/aisy.202000181. [DOI] [Google Scholar]
- 2.Chehab M, Brinjikji W, Copelan A, Venkatesan A. Navigational tools for interventional radiology and interventional oncology applications. Semin Intervent Radiol. 2015;32(4):416‐427. doi: 10.1055/s-0035-1564705. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Tacher V, de Baere T. Robotic assistance in interventional radiology: Dream or reality? Eur Radiol. 2020;30(2):925‐926. doi: 10.1007/s00330-019-06541-w. [DOI] [PubMed] [Google Scholar]
- 4.Fuhrmann I, Probst U, Wiggermann P, Beyer L. Navigation systems for treatment planning and execution of percutaneous irreversible electroporation. Technol Cancer Res Treat. 2018;17:1533033818791792. 10.1177/1533033818791792 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Rueda MA, Riga CT, Hamady MS. Robotics in interventional radiology: past, present, and future. Arab J Interv Radiol. 2018;2(02):56‐63. doi: 10.4103/AJIR.AJIR_25_18. [DOI] [Google Scholar]
- 6.Kassamali RH, Ladak B. The role of robotics in interventional radiology: current status. Quant Imaging Med Surg. 2015;5(3):340‐343. doi: 10.3978/j.issn.2223-4292.2015.03.15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Unger M, Berger J, Melzer A. Robot-assisted image-guided interventions. Front Robot AI. 2021;8, doi: 10.3389/frobt.2021.664622. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Yanof J, Haaga J, Klahr P, et al. CT-integrated robot for interventional procedures: preliminary experiment and computer–human interfaces. Comput Aided Surg. 2001;6(6):352‐359. doi: 10.1002/igs.10022. [DOI] [PubMed] [Google Scholar]
- 9.Cleary K, Melzer A, Watson V, Kronreif G, Stoianovici D. Interventional robotic systems: applications and technology state-of-the-art. Minim Invasive Ther Allied Technol. 2006;15(2):101‐113. doi: 10.1080/13645700600674179. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Beaman CB, Kaneko N, Meyers PM, Tateshima S. A review of robotic interventional neuroradiology. Am J Neuroradiol. 2021;42(5):808‐814. doi: 10.3174/ajnr.A6976. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Solomon SB, Patriciu A, Bohlman ME, Kavoussi LR, Stoianovici D. Robotically driven interventions: a method of using CT fluoroscopy without radiation exposure to the physician. Radiology. 2002;225(1):277‐282. doi: 10.1148/radiol.2251011133. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Beyer LP, Pregler B, Niessen C, et al. Robot-assisted microwave thermoablation of liver tumors: a single-center experience. Int J Comput Assist Radiol Surg. 2016;11(2):253‐259. doi: 10.1007/s11548-015-1286-y. [DOI] [PubMed] [Google Scholar]
- 13.Mbalisike EC, Vogl TJ, Zangos S, Eichler K, Balakrishnan P, Paul J. Image-guided microwave thermoablation of hepatic tumours using novel robotic guidance: an early experience. Eur Radiol. 2015;25(2):454‐462. doi: 10.1007/s00330-014-3398-0. [DOI] [PubMed] [Google Scholar]
- 14.Beyer LP, Pregler B, Michalik K, et al. Evaluation of a robotic system for irreversible electroporation (IRE) of malignant liver tumors: initial results. Int J Comput Assist Radiol Surg. 2017;12(5):803‐809. doi: 10.1007/s11548-016-1485-1. [DOI] [PubMed] [Google Scholar]
- 15.Anzidei M, Argirò R, Porfiri A, et al. Preliminary clinical experience with a dedicated interventional robotic system for CT-guided biopsies of lung lesions: a comparison with the conventional manual technique. Eur Radiol. 2015;25(5):1310‐1316. doi: 10.1007/s00330-014-3508-z. [DOI] [PubMed] [Google Scholar]
- 16.Loser MH, Navab N. A new robotic system for visually controlled percutaneous interventions under CT fluoroscopy. 2000:887–896. doi: 10.1007/978-3-540-40899-4_92. [DOI]
- 17.Patriciu A, Awad M, Solomon SB, et al. Robotic assisted radio-frequency ablation of liver tumors—randomized patient study. Med Image Comput Comput Assist Interv. 2005;8(Pt 2):526‐533. doi: 10.1007/11566489_65 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Stoffner R, Augschöll C, Widmann G, Böhler D, Bale R. Accuracy and feasibility of frameless stereotactic and robot-assisted CT-based puncture in interventional radiology: a comparative phantom study. RöFo—Fortschritte auf dem Gebiet der Röntgenstrahlen und der Bildgeb Verfahren. 2009;181(09):851‐858. doi: 10.1055/s-0028-1109380. [DOI] [PubMed] [Google Scholar]
- 19.Czerny C, Eichler K, Croissant Y, et al. Combining C-arm CT with a new remote operated positioning and guidance system for guidance of minimally invasive spine interventions. J Neurointerv Surg. 2015;7(4):303‐308. doi: 10.1136/neurintsurg-2013-011034. [DOI] [PubMed] [Google Scholar]
- 20.Abdullah BJJ, Yeong CH, Goh KL, et al. Robotic-assisted thermal ablation of liver tumours. Eur Radiol. 2015;25(1):246‐257. doi: 10.1007/s00330-014-3391-7. [DOI] [PubMed] [Google Scholar]
- 21.Minchev G, Kronreif G, Martínez-Moreno M, et al. A novel miniature robotic guidance device for stereotactic neurosurgical interventions: preliminary experience with the iSYS1 robot. J Neurosurg. 2017;126(3):985‐996. doi: 10.3171/2016.1.JNS152005. [DOI] [PubMed] [Google Scholar]
- 22.Levy S, Goldberg SN, Roth I, et al. Clinical evaluation of a robotic system for precise CT-guided percutaneous procedures. Abdom Radiol. 2021;46(10):5007‐5016. doi: 10.1007/s00261-021-03175-9. [DOI] [PubMed] [Google Scholar]
- 23.Hata N, Song SE, Olubiyi O, et al. Body-mounted robotic instrument guide for image-guided cryotherapy of renal cancer. Med Phys. 2016;43(2):843‐853. doi: 10.1118/1.4939875. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Hata N, Moreira P, Fischer G. Robotics in MRI-guided interventions. In: Topics in magnetic resonance imaging. Vol 27. Lippincott Williams and Wilkins; 2018:19‐23. doi: 10.1097/RMR.0000000000000159. [DOI] [PubMed] [Google Scholar]
- 25.Kettenbach J, Kronreif G, Figl M, et al. Robot-assisted biopsy using ultrasound guidance: initial results from in vitro tests. Eur Radiol. 2005;15(4):765‐771. doi: 10.1007/s00330-004-2487-x. [DOI] [PubMed] [Google Scholar]
- 26.Berger J, Unger M, Landgraf L, Bieck R, Neumuth T, Melzer A. Assessment of natural user interactions for robot-assisted interventions. Curr Dir Biomed Eng. 2018;4(1):165‐168. doi: 10.1515/cdbme-2018-0041. [DOI] [Google Scholar]
- 27.Wu FY, Torabi M, Yamada A, et al. An MRI coil-mounted multi-probe robotic positioner for cryoablation. In: Proceedings of the ASME Design Engineering Technical Conference. Vol 6 A. American Society of Mechanical Engineers; 2013. doi: 10.1115/DETC2013-13132. [DOI]
- 28.He Z, Kwok KW, Dong Z, et al. Design of a percutaneous MRI-guided needle robot with soft fluid-driven actuator. IEEE Robot Autom Lett. 2020;5(2):2100‐2107. doi: 10.1109/LRA.2020.2969929. [DOI] [Google Scholar]
- 29.de Baère T, Roux C, Deschamps F, Tselikas L, Guiu B. Evaluation of a new CT-guided robotic system for percutaneous needle insertion for thermal ablation of liver tumors: a prospective pilot study. Cardiovasc Intervent Radiol. 2022. doi: 10.1007/s00270-022-03267-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Carriero S, Della Pepa G, Monfardini L, et al. Role of fusion imaging in image-guided thermal ablations. Diagnostics. 2021;11(3):549. doi: 10.3390/diagnostics11030549. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Capitanio U, Montorsi F. Renal cancer. Lancet. 2016;387(10021):894‐906. doi: 10.1016/S0140-6736(15)00046-X. [DOI] [PubMed] [Google Scholar]
- 32.Higgins LJ, Hong K. Renal ablation techniques: state of the art. Am J Roentgenol. 2015;205(4):735‐741. doi: 10.2214/AJR.15.14752. [DOI] [PubMed] [Google Scholar]
- 33.Knott EA, Swietlik JF, Longo KC, et al. Robotically-assisted sonic therapy for renal ablation in a live porcine model: initial preclinical results. J Vasc Interv Radiol. 2019;30(8):1293‐1302. doi: 10.1016/j.jvir.2019.01.023. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Tilak G, Tuncali K, Song SE, et al. 3T MR-guided in-bore transperineal prostate biopsy: a comparison of robotic and manual needle-guidance templates. J Magn Reson Imaging. 2015;42(1):63‐71. doi: 10.1002/jmri.24770. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Lim S, Jun C, Chang D, Petrisor D, Han M, Stoianovici D. Robotic transrectal ultrasound guided prostate biopsy. IEEE Trans Biomed Eng. 2019;66(9):2527‐2537. doi: 10.1109/TBME.2019.2891240. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Welleweerd MK, Siepel FJ, Groenhuis V, Veltman J, Stramigioli S. Design of an end-effector for robot-assisted ultrasound-guided breast biopsies. Int J Comput Assist Radiol Surg. 2020;15(4):681‐690. doi: 10.1007/s11548-020-02122-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Agnese DM, Burak WE. Ablative approaches to the minimally invasive treatment of breast cancer. Cancer J. 2005;11(1):77‐82. doi: 10.1097/00130404-200501000-00012. [DOI] [PubMed] [Google Scholar]
- 38.Earashi M, Noguchi M, Motoyoshi A, Fujii H. Radiofrequency ablation therapy for small breast cancer followed by immediate surgical resection or delayed mammotome excision. Breast Cancer. 2007;14(1):39‐47. doi: 10.2325/jbcs.14.39. [DOI] [PubMed] [Google Scholar]
- 39.Desai JP, Gullapalli R. Image-guided robotic interventions for breast biopsy and ablative therapies: challenges and opportunities. Women’s Heal. 2009;5(3):229‐233. doi: 10.2217/WHE.09.1. [DOI] [PubMed] [Google Scholar]
- 40.Scheffler K. Fast frequency mapping with balanced SSFP: Theory and application to proton-resonance frequency shift thermometry. Magn Reson Med. 2004;51(6):1205‐1211. doi: 10.1002/mrm.20081. [DOI] [PubMed] [Google Scholar]
- 41.Lefranc M, Peltier J. Accuracy of thoracolumbar transpedicular and vertebral body percutaneous screw placement: coupling the Rosa® spine robot with intraoperative flat-panel CT guidance—a cadaver study. J Robot Surg. 2015;9(4):331‐338. doi: 10.1007/s11701-015-0536-x. [DOI] [PubMed] [Google Scholar]
- 42.Chenin L, Peltier J, Lefranc M. Minimally invasive transforaminal lumbar interbody fusion with the ROSATM spine robot and intraoperative flat-panel CT guidance. Acta Neurochir (Wien). 2016;158(6):1125‐1128. doi: 10.1007/s00701-016-2799-z. [DOI] [PubMed] [Google Scholar]
- 43.Carai A, Mastronuzzi A, De Benedictis A, et al. Robot-assisted stereotactic biopsy of diffuse intrinsic pontine glioma: a single-center experience. World Neurosurg. 2017;101:584‐588. doi: 10.1016/j.wneu.2017.02.088. [DOI] [PubMed] [Google Scholar]
- 44.Vadera S, Chan A, Lo T, et al. Frameless stereotactic robot-assisted subthalamic nucleus deep brain stimulation: case report. World Neurosurg. 2017;97(762):e11‐762.e14. doi: 10.1016/j.wneu.2015.11.009. [DOI] [PubMed] [Google Scholar]
- 45.Hunsche S, Sauner D, Majdoub FE, et al. Intensity-based 2D 3D registration for lead localization in robot guided deep brain stimulation. Phys Med Biol. 2017;62(6):2417‐2426. doi: 10.1088/1361-6560/aa5ecd. [DOI] [PubMed] [Google Scholar]
- 46.Long T, Li K, Gao J, et al. Comparative study of percutaneous sacroiliac screw with or without Ti robot assistance for treating pelvic posterior ring fractures. Orthop Surg. 2019;11(3):386‐396. doi: 10.1111/os.12461. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Du J, Gao L, Huang D, et al. Radiological and clinical differences between Tinavi orthopedic robot and O-arm navigation system in thoracolumbar screw implantation for reconstruction of spinal stability. Med Sci Monit. 2020;26:e924770. doi: 10.12659/MSM.924770. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Li G, Patel NA, Melzer A, Sharma K, Iordachita I, Cleary K. MRI-guided lumbar spinal injections with body-mounted robotic system: cadaver studies. Minim Invasive Ther Allied Technol. 2022;31(2):297‐305. doi: 10.1080/13645706.2020.1799017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Rafii-Tari H, Payne CJ, Yang GZ. Current and emerging robot-assisted endovascular catheterization technologies: A review. Ann Biomed Eng. 2014;42(4):697‐715. doi: 10.1007/s10439-013-0946-8. [DOI] [PubMed] [Google Scholar]
- 50.Mahmud E, Schmid F, Kalmar P, et al. Feasibility and safety of robotic peripheral vascular interventions results of the RAPID trial. 2016. [DOI] [PubMed]
- 51.Sajja KC, Sweid A, Al Saiegh F, et al. Endovascular robotic: feasibility and proof of principle for diagnostic cerebral angiography and carotid artery stenting. J Neurointerv Surg. 2020;12(4):345‐349. doi: 10.1136/neurintsurg-2019-015763. [DOI] [PubMed] [Google Scholar]
- 52.Pereira M. First-in-human, robotic-assisted neuroendovascular intervention. J NeuroIntervent Surg. 2020;12(4):338‐340. doi: 10.1136/neurintsurg-2019-015671. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53.Menaker SA, Shah SS, Snelling BM, Sur S, Starke RM, Peterson EC. Current applications and future perspectives of robotics in cerebrovascular and endovascular neurosurgery. J Neurointerv Surg. 2018;10(1):78‐82. doi: 10.1136/neurintsurg-2017-013284. [DOI] [PubMed] [Google Scholar]
- 54.Ahmidi N, Hager GD, Ishii L, Fichtinger G, Gallia GL, Ishii M. Surgical task and skill classification from eye tracking and tool motion in minimally invasive surgery. Med Image Comput Comput Interv MICCAI. Int Conf Med Image Comput Comput Interv. 2010;13(Pt 3):295‐302. doi: 10.1007/978-3-642-15711-0_37. [DOI] [PubMed] [Google Scholar]
- 55.Alderliesten T, Konings MK, Niessen WJ. Modeling friction, intrinsic curvature, and rotation of guide wires for simulation of minimally invasive vascular interventions. IEEE Trans Biomed Eng. 2007;54(1):29‐38. doi: 10.1109/TBME.2006.886659. [DOI] [PubMed] [Google Scholar]
- 56.Allaqaband S, Solis J, Kazemi S, Bajwa T. Endovascular treatment of peripheral vascular disease. Curr Probl Cardiol. 2006;31(11):711‐760. doi: 10.1016/j.cpcardiol.2006.08.002. [DOI] [PubMed] [Google Scholar]
- 57.Da L, Zhang D, Wang T. Overview of the vascular interventional robot. Int J Med Robot Comput Assist Surg. 2008;4(4):289‐294. doi: 10.1002/rcs.212. [DOI] [PubMed] [Google Scholar]
- 58.Fu Y, Liu H, Huang W, Wang S, Liang Z. Steerable catheters in minimally invasive vascular surgery. Int J Med Robot Comput Assist Surg. 2009;5(4):381‐391. doi: 10.1002/rcs.282. [DOI] [PubMed] [Google Scholar]
- 59.Argall BD, Chernova S, Veloso M, Browning B. A survey of robot learning from demonstration. Rob Auton Syst. 2009;57(5):469‐483. doi: 10.1016/j.robot.2008.10.024. [DOI] [Google Scholar]
- 60.Baert SAM, Viergever MA, Niessen WJ. Guide-wire tracking during endovascular interventions. IEEE Trans Med Imaging. 2003;22(8):965‐972. doi: 10.1109/TMI.2003.815904. [DOI] [PubMed] [Google Scholar]
- 61.Bicknell CD, Cheshire NJW, Riga C V, et al. Treatment of complex aneurysmal disease with fenestrated and branched stent grafts. Eur J Vasc Endovasc Surg. 2009;37(2):175‐181. doi: 10.1016/j.ejvs.2008.11.008. [DOI] [PubMed] [Google Scholar]
- 62.Bismuth J, Kashef E, Cheshire N, Lumsden AB. Feasibility and safety of remote endovascular catheter navigation in a porcine model. J Endovasc Ther. 2011;18(2):243-249. [DOI] [PubMed]
- 63.Rolls AE, Riga C V, Bicknell CD, Regan L, Cheshire NJ, Hamady MS. Robot-assisted uterine artery embolization: a first-in-woman safety evaluation of the magellan system. J Vasc Interv Radiol. 2014;25(12):1841‐1848. doi: 10.1016/j.jvir.2014.05.022. [DOI] [PubMed] [Google Scholar]
- 64.Rao S. Robot-assisted transarterial chemoembolization for hepatocellular carcinoma: initial evaluation of safety, feasibility, success and outcomes using the Magellan system. J Vasc Interv Radiol. 2015;26(2):S12. doi: 10.1016/j.jvir.2014.12.039. [DOI] [Google Scholar]
- 65.Abdominal applications of ultrasound fusion imaging technique: liver, kidney, and pancreas. Insights Imaging. 2019;10(1):6. doi: 10.1186/s13244-019-0692-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 66.Krücker J, Xu S, Venkatesan A, et al. Clinical utility of real-time fusion guidance for biopsy and ablation. J Vasc Interv Radiol. 2011;22(4):515‐524. doi: 10.1016/j.jvir.2010.10.033. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 67.Fontana F, Piacentino F, Ierardi AM, et al. Comparison between CBCT and fusion PET/CT-CBCT guidance for lung biopsies. Cardiovasc Intervent Radiol. 2021;44(1):73‐79. doi: 10.1007/s00270-020-02613-3. [DOI] [PubMed] [Google Scholar]
- 68.Monfardini L, Gennaro N, Orsi F, et al. Real-time US/cone-beam CT fusion imaging for percutaneous ablation of small renal tumours: A technical note. Eur Radiol. 2021;31(10):7523‐7528. doi: 10.1007/s00330-021-07930-w. [DOI] [PubMed] [Google Scholar]
- 69.Floridi C, Muollo A, Fontana F, et al. C-arm cone-beam computed tomography needle path overlay for percutaneous biopsy of pulmonary nodules. Radiol Med. 2014;119(11):820‐827. doi: 10.1007/s11547-014-0406-z. [DOI] [PubMed] [Google Scholar]
- 70.Ierardi AM, Grillo P, Bonanno MC, et al. Prediction of ablation volume in percutaneous lung microwave ablation: a single centre retrospective study. Tomography. 2022;8(5):2475‐2485. doi: 10.3390/tomography8050206. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 71.Chiaradia M, Izamis M-L, Radaelli A, et al. Sensitivity and reproducibility of automated feeding artery detection software during transarterial chemoembolization of hepatocellular carcinoma. J Vasc Interv Radiol. 2018;29(3):425‐431. doi: 10.1016/j.jvir.2017.10.025. [DOI] [PubMed] [Google Scholar]
- 72.Abdelsalam H, Emara DM, Hassouna EM. The efficacy of TACE; how can automated feeder software help? Egypt J Radiol Nucl Med. 2022;53(1):43. doi: 10.1186/s43055-022-00720-4. [DOI] [Google Scholar]
- 73.Miyayama S, Yamashiro M, Ikeda R, et al. Efficacy of superselective conventional transarterial chemoembolization using guidance software for hepatocellular carcinoma within three lesions smaller than 3 cm. Cancers (Basel). 2021;13(24):6370. doi: 10.3390/cancers13246370. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 74.Shan Q, Hu H, Feng S, et al. CT-based peritumoral radiomics signatures to predict early recurrence in hepatocellular carcinoma after curative tumor resection or ablation. Cancer Imaging. 2019;19(1):11. doi: 10.1186/s40644-019-0197-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 75.Park SY, Kim CK, Park BK. Dual-energy CT in assessing therapeutic response to radiofrequency ablation of renal cell carcinomas. Eur J Radiol. 2014;83(2):e73‐e79. doi: 10.1016/j.ejrad.2013.11.022. [DOI] [PubMed] [Google Scholar]
- 76.Wu J, Ding W, Wang Y, et al. Radiomics analysis of ultrasound to predict recurrence of hepatocellular carcinoma after microwave ablation. Int J Hyperth. 2022;39(1):595‐604. doi: 10.1080/02656736.2022.2062463. [DOI] [PubMed] [Google Scholar]
- 77.Taghavi M, Staal FC, Simões R, et al. CT radiomics models are unable to predict new liver metastasis after successful thermal ablation of colorectal liver metastases. Acta Radiol. 2021;64(1): 028418512110604. doi: 10.1177/02841851211060437. [DOI] [PubMed] [Google Scholar]
- 78.Petukhova-Greenstein A, Zeevi T, Yang J, et al. MR imaging biomarkers for the prediction of outcome after radiofrequency ablation of hepatocellular carcinoma: qualitative and quantitative assessments of the liver imaging reporting and data system and radiomic features. J Vasc Interv Radiol. 2022;33(7):814‐824.e3. doi: 10.1016/j.jvir.2022.04.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 79.Zhang G, Yang H, Zhu X, et al. A CT-based radiomics nomogram to predict complete ablation of pulmonary malignancy: a multicenter study. Front Oncol. 2022;12:841678. doi: 10.3389/fonc.2022.841678. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 80.Gurgitano M, Angileri SA, Rodà GM, et al. Interventional radiology ex-machina: impact of artificial intelligence on practice. Radiol Med. 2021;126(7):998‐1006. doi: 10.1007/s11547-021-01351-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 81.Wallach D, Toporek G, Weber S, Bale R, Widmann G. Comparison of freehand-navigated and aiming device-navigated targeting of liver lesions. Int J Med Robot Comput Assist Surg. 2014;10(1):35‐43. doi: 10.1002/rcs.1505. [DOI] [PubMed] [Google Scholar]
- 82.Durand P, Moreau-Gaudry A, Silvent A-S, et al. Computer assisted electromagnetic navigation improves accuracy in computed tomography guided interventions: a prospective randomized clinical trial. PLoS One. 2017;12(3):e0173751. doi: 10.1371/journal.pone.0173751. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 83.Lonjon N, Chan-Seng E, Costalat V, Bonnafoux B, Vassal M, Boetto J. Robot-assisted spine surgery: feasibility study through a prospective case-matched analysis. Eur Spine J. 2016;25(3):947‐955. doi: 10.1007/s00586-015-3758-8. [DOI] [PubMed] [Google Scholar]
- 84.Engstrand J, Toporek G, Harbut P, Jonas E, Nilsson H, Freedman J. Stereotactic CT-guided percutaneous microwave ablation of liver tumors with the use of high-frequency jet ventilation: an accuracy and procedural safety study. Am J Roentgenol. 2017;208(1):193‐200. doi: 10.2214/AJR.15.15803. [DOI] [PubMed] [Google Scholar]
- 85.Melzer A, Gutmann B, Remmele T, et al. INNOMOTION for percutaneous image-guided interventions. IEEE Eng Med Biol Mag. 2008;27(3):66‐73. doi: 10.1109/EMB.2007.910274. [DOI] [PubMed] [Google Scholar]

