Application of ultrasound fusion imaging technique for unilateral percutaneous vertebroplasty in treatment of osteoporotic thoracolumbar compression fracture

Abstract

OBJECTIVE:

To evaluate application of a computed tomography (CT)-ultrasound fusion imaging technique to unilateral percutaneous vertebroplasty (PVP) for treating patients with osteoporotic thoracolumbar compression fracture.

METHODS:

Fourteen patients with osteoporotic thoracolumbar compression fractures were included, randomly divided into CT-ultrasound fusion imaging (n = 7) and traditional X-ray fluoroscopy groups (n = 7). Patients in the first group underwent unilateral PVP using real-time CT-ultrasound fusion imaging. A body surface locator was placed on the side contralateral to the scheduled puncture site (2–3 cm from the spinous process). Patient CT image information was recorded in the ultrasound system for registration during real-time ultrasound and CT fusion imaging, and one-click automatic registration was then performed. The puncture point and target point at which the puncture needle arrived were determined on CT images, with the puncture being performed under ultrasound guidance. Patients in the second group underwent X-ray fluoroscopy-guided PVP. Bone cement injection was injected under monitoring using a C-arm X-ray system. Patients’ X-ray exposure and puncture times were recorded and compared between the two groups.

RESULTS:

The average puncture times in the CT-ultrasound fusion imaging and traditional X-ray fluoroscopy groups were 2.50±0.31 min (without exposing patients and operators to radiation) and 5.00±0.65 min (with the same duration of radiation exposure), respectively. The average times for bone cement injection were 3.29±0.81 min and 3.50±0.86 min, respectively. The mean visual analog scale (VAS) scores were 2.10±0.11 and 2.20±0.21, respectively. The bone cement was evenly distributed without cement leakage in patients in the CT-ultrasound fusion imaging group, but a poor distribution of bone cement and bone cement leakage were found in one patient in the traditional X-ray fluoroscopy group.

CONCLUSIONS:

Real-time CT-ultrasound fusion imaging is easy to perform, and provides precise localization of the puncture point, path, and target point. The selected puncture path was reasonable, and the needle had reached the target point accurately, which increased the success rate of puncture without radiation exposure.

Keywords

Ultrasound fusion imaging technique unilateral percutaneous vertebroplasty osteoporotic thoracolumbar compression fracture

1 Introduction

Osteoporosis has varying degrees of impact on the quality of life of older adults. Currently, osteoporotic vertebral compression fractures in older adults are the most common type of fracture seen in the field of spinal surgery [1]. Percutaneous vertebroplasty (PVP) has been increasingly applied for the treatment of osteoporotic vertebral compression fractures, achieving satisfactory clinical results [2]. The puncture is the key technique in PVP, directly affecting its efficacy. Of all the types of orthopedic surgery, PVP results in the highest radiation exposure to the surgeon [3, 4]. Therefore, researchers and orthopedic surgeons have provided many suggestions for reducing the intraoperative radiation exposure, such as wearing protective lead clothes, placing radiation shielding devices in the operating room, and reducing the number of intraoperative x-rays [5].

Currently, the two aspects most under consideration for reducing the intraoperative radiation exposure [6] are: 1) performing of unilateral PVP instead of traditional bilateral PVP, and 2) the use of robots for performing the surgical procedures. Many studies have compared the efficacy of unilateral and bilateral PVP [7 –10], and it is generally considered that PVP via a unilateral pedicular approach can effectively reduce radiation exposure time. The unilateral PVP approach can also reduce the risk of cement leakage into the epidural and intervertebral spaces through the puncture site [7 –10]. However, the radiation dose to the operator is still high. More importantly, the safety of PVP performed with reduced use of intraoperative X-ray fluoroscopy has been challenged. With the rapid development of computer-assisted navigation technology, robotic technology, and microsurgery technology, an increasing number of medical robots are being used in clinical practice [11, 12]. However, medical robots cannot currently perform difficult surgical procedures such as PVP, where the unilateral puncture generally requires an increased outer inclination angle, which may result in the puncture needle entering the spinal canal and thereby increasing the risk of damage to the nerve root. Therefore, reduction of the intraoperative radiation exposure to both the surgeon and patient has become a particularly urgent problem that needs to be solved.

Ultrasound has been used to guide percutaneous interventional procedures because of its advantages of real-time imaging, easy accessibility, lack of exposure to ionizing radiation, and lower cost, while the additional use of color Doppler ultrasound allows identification of vascular structures [13]. However, ultrasound is an operator-dependent method, and its use can affect the success of the procedure. Fusion of ultrasound images with other imaging modalities, such as CT and MRI, allows radiologists to exploit all the strengths of the different methods, eliminating or minimizing the weaknesses of a single modality. Fusion imaging uses electromagnetic position-tracking technology and a series of hardware and software techniques to achieve simultaneous display of real-time ultrasound images together with images from different modalities or images acquired at different times. The complementary advantages of two different imaging modes may then be incorporated. A study investigating the effectiveness of ultrasound-guided injections of the sacroiliac and facet joints showed that the fusion of ultrasound and CT images was accurate [14], and the techniques have also been used for education purposes [15]. However, little is known about the use of fusion imaging for guiding the puncture during PVP. Therefore, with the intent of reducing the exposure time to radiation, we used a real-time ultrasound fusion imaging technique to guide the puncture process during unilateral PVP, and verified the localization accuracy using the ultrasound fusion imaging.

2 Subjects and methods

2.1 Subjects

Fourteen patients with osteoporotic thoracolumbar compression fractures who were admitted to our hospital between April 2019 and August 2019 were included in this study. All patients were examined with CT and were randomly allocated to CT-ultrasound fusion imaging or traditional X-ray fluoroscopy groups, with seven patients in each group. Patients in the CT-ultrasound fusion imaging group underwent unilateral PVP with real-time CT-ultrasound fusion imaging, while patients in the traditional X-ray fluoroscopy group underwent X-ray fluoroscopy-guided PVP. The baseline characteristics of the patients in each group are presented in Table 1.

Table 1
Baseline characteristics of the patients in the two groups

Groups Gender Mean age Fracture site VSA score

Male Female T11 T10 L2

CT-ultrasound fusion imaging group (n = 7) 2 5 70±3.5 (range 62–82 years) 2 1 4 8.5

Traditional X-ray fluoroscopy group (n = 7) 1 6 71±3.2 (range 65–81 years) 2 1 4 8.3

Groups	Gender	Mean age	Fracture site	VSA score
CT-ultrasound fusion imaging group (n = 7)	2	5	70±3.5 (range 62–82 years)	2	1	4	8.5
Traditional X-ray fluoroscopy group (n = 7)	1	6	71±3.2 (range 65–81 years)	2	1	4	8.3

The inclusion criteria included: osteoporotic vertebral compression fracture and intact posterior wall of the vertebral body confirmed by imaging examinations; conservative treatments such as bed rest and analgesic use were ineffective; absence of non-healing wounds and skin ulcer around the surgical field; fresh fractures or non-union fractures confirmed by MRI or bone scan; and no invasion of myeloma in the posterior wall of the vertebral body.

The exclusion criteria included: unstable fractures; a bone fragment protruding posteriorly and occupying the spinal canal; posterior ligament complex injury; more than 2/3 loss of the anterior vertebral height; significant kyphotic deformity; and multiple vertebral metastases or lesions with destruction of the posterior wall of the vertebral body.

2.2 Fusion imaging technique

All patients were placed in the supine position (the same position used for the percutaneous puncture) and a body surface locator was positioned on the opposite side of the puncture site, 2–3 cm from the spinous process (Fig. 1). A CT scan was then acquired, making sure that the four marker points of the body surface locator were within the scanning range, and the images were saved to a CD.

Fig.1

Placement of the body surface locator.

Fusion imaging was performed with an APLIO i900 ultrasound system (Canon Medical Systems, Otawara, Japan). The main hardware of the system included a magnetic field generator and two sensing lines, with one end of a sensing line being attached to the main body of the system, and the other end being attached to a 1–8 MHz convex array probe. One end of the other sensing line was attached to the puncture needle, and its other end was attached to an automatic tracker (Figs. 2 and 3). The main software packages used included Smart Fusion, Smart Navigation, Auto Registration, and Smart Sensor 3D. The preparation steps before the operation are shown in Figs. 4 and 5.

Fig.2

A: The magnetic field generator and inductorium. B: The magnetic field generator of the image fusion system.

Fig.3

Fusion and navigation components.

Fig.4

Pre-operative preparation: establishment of patient information 1: Place the CD with patient information in CD drive; 2: choose Optica Media in the Location; 3: Select patient information that needs to be imported; 4: click Load to HDDo.

Fig.5

Pre-operative preparation: patient information establishment 1: Click on the “Exam” on the touch screen to enter the new patient interface; 2: Search for case information with image data according to ID or name; 3: Choose adherence criteria, such as MSK; 4: click Start.

2.3 Preparation of fusion imaging

The effective distance between the transmitter and the magnetic sensor is shown in Fig. 6A. If the distance from the transmitter to the magnetic sensor is within this range, the puncture site can be accurately localized. However, if metal objects or other magnetic materials are present in the vicinity of the ultrasound system, even if the distance is within this range, the accuracy of the localization can be low; therefore, the use of magnetic materials in the surrounding environment should be avoided. The transmitter should be oriented with the side with the lift-off function facing the application site of the transducer. Otherwise, the puncture guide line may not match the position of the puncture needle (Fig. 6B). CT image data were imported into the ultrasound system (Fig. 7) and patient specific data were created by matching the patients’ names with their input image data.

Fig.6

A: Effective distance from the transmitter to the magnetic sensor; B: The transmitter with the lifting function side facing the application site of the transducer.

Fig.7

CT image data is imported into the ultrasound system.

2.4 Fusing of CT images with ultrasound images

The ultrasound probe was modulated with the left and right sides being mutually adjusted and the fusion imaging technique was started. Corresponding image data were selected, with anatomical markers of tendons, vessels, and bones being used. After determination of the ultrasound image markers, the X-, Y-, Z-axis and zoom tools were used to find the images on the CT scans matching the ultrasound images. If the patient was scanned laying face downwards during the CT examination, the CT scan images were rotated 180° around the Y axis, making the images match with the position in which the ultrasound examination was performed. Once markers at the same level were located for the ultrasound examination and CT scans, the automatic fusion button was activated, and the instant image fusion process was begun. The degree of fusion was verified in two planes to achieve full fusion (Fig. 8). Manual modulation was necessary to reduce the deviation between the CT and ultrasound images and achieve optimal fusion.

Fig.8

(A, B) An instant fusion image is generated once the automatic fusion button is activated. The degree of fusion is verified from the anteroposterior and lateral views. (C) The X-, Y-, Z-axis and zoom tools were used to find the same image on the CT scans to allow full image fusion.

2.5 Fusion imaging for localization and puncture

The patient was asked to lie in the prone position with both hands on either side of the head and the hips were fixed with a restraining strap. After conventional disinfection, the ultrasonic probe was covered with a sterile cover and its external surface was coated with sterile coupling agent. Using the CT scans, the insertion site for the puncture needle was determined in the outer edge of the pedicle of the vertebral arch, and the target point reached by the puncture needle was determined according to the puncture path (Fig. 9). A 5-mL syringe containing 5 mL 1% lidocaine was attached to a long trocar and a percutaneous puncture was made according to the needle entry point under real-time ultrasound monitoring. Lidocaine was injected on the ideal bone surface and the needle was then removed. After the lidocaine injection, a 5-mm long transverse incision was made at the injection site using a scalpel. An 11 G or 13 G puncture needle (COOK needle Murphy II) was selected according to the size of the pedicle, and was percutaneously inserted via the predesigned entry and target points.

Fig.9

On CT scans, the entry point (red circles) was determined on the outer edge of the pedicle, and the target point (blue circles) was designated as the center of the vertebral body according to the puncture path.

Once the target point was reached, the ultrasound probe was rotated into the long-axis view (sagittal CT) to confirm that the needle had reached the target point (Fig. 10A, B). The puncture was then terminated. A C-arm X-ray system was used to further verify that the needle had reached the target point; in the anteroposterior view, the needle should reach the spinous root of the vertebral body when in the correct position, while on the lateral view it should reach the anterior 1/3 part of the vertebral body (Fig. 10 C).

Fig.10

(A, B) The puncture needle reaching the target point on the sagittal CT view; (C) Confirmation of the needle reaching the target point on anteroposterior and lateral views obtained from a C-arm X-ray system.

2.6 Bone cement injection

After confirmation of the puncture needle reaching the target point, the bone cement was stirred. Bone cement in the wetting stage was injected into a special injection sleeve, and when it reached the drawing stage, it was slowly injected into the fractured vertebral body through a working sleeve under the perspective of the C-arm X-ray system. If bone cement exceeded the median line on the anteroposterior projection (the posterior 1/4 part of the vertebral body), the injection was terminated (Fig. 11A). Injection of bone cement was also terminated if it leaked into the intervertebral space or paravertebral soft tissues. Following the procedure, the puncture point was wrapped with a sterile dressing. When the uninjected bone cement had heat-hardened, the patient was turned over to a surgical flat bed. After lying flat for 24 hours, the patient was asked to get out of bed for activities. CT scans were performed again to check the distribution of the bone cement (Fig. 11B).

Fig.11

A: The bone cement injection meets the surgical requirement; B: Distribution of bone cement in the vertebral body.

2.7 Outcome measures

The correct position of the puncture needle was verified using a C-arm X-ray system. The patient’s X-ray exposure time and puncture time were then recorded and compared between the two groups.

3 Results

In the CT-ultrasound fusion imaging group, image registration and matching were completed within 30 minutes in one 76-year-old female patient, and within 2 minutes in the other six patients. The puncture time was between 2 and 3 min, with an average of 2.50±0.31 min, with neither patients nor operators being exposed to ionizing radiation during this procedure. An accurate puncture was achieved in all patients. The bone cement injection took between 2 and 4 min with an average of 3.29±0.81 min. The mean visual analog scale (VAS) score was 2.10±0.11. The CT images showed that the bone cement was evenly distributed without cement leakage.

In the traditional X-ray fluoroscopy group, the puncture time was between 4 and 6 min, with an average time of 5.00±0.65 min, with the radiation exposure time being the same as the puncture time. The bone cement injection took 2.5–5 min, with an average duration of 3.50±0.86 min. The mean VAS score was 2.20±0.21. CT images showed that bone cement was poorly distributed in one patient, with the bone cement having leaked outside the vertebral body margins, although the patient had no clinical symptoms (Table 2).

Table 2
Comparison of the average time durations of the puncture and bone cement injection procedures between the two groups

Groups The average puncture time The average time of bone cement injection

CT-ultrasound fusion imaging group 2.50 ± 0.31 3.29 ± 0.81

Traditional X-ray fluoroscopy group 5.00 ± 0.65 3.50 ± 0.86

F value 7.638 0.348

P value 0.000 0.740

Groups	The average puncture time	The average time of bone cement injection
CT-ultrasound fusion imaging group	2.50 ± 0.31	3.29 ± 0.81
Traditional X-ray fluoroscopy group	5.00 ± 0.65	3.50 ± 0.86
F value	7.638	0.348
P value	0.000	0.740

4 Discussion

Fusion imaging is a technique using real time ultrasound guidance, with the entire localization and puncture procedures requiring no X-ray fluoroscopy, which effectively reduces the radiation exposure time. Ultrasound images fused with CT/MRI images can compensate for the disadvantages of ultrasound images in terms of bone- and gas-filled structures, yet still present the advantages of real-time imaging and dynamic display of blood flow inherent to ultrasound imaging. The fusion imaging of ultrasound with CT/MRI images provides obvious advantages for improving medical imaging evaluation, especially in interventional procedures [16]. Fusion imaging techniques have been used for liver cancer and prostate cancer [17 –19], and for a variety of image-guided procedures, physical examinations, and prenatal diagnoses [20]. However, little is known about the use of fusion imaging in musculoskeletal (MSK) sports medicine injuries.

In the current study, a new Aplio i-series ultrasound system (APLIO i900) produced by Canon Medical Systems was used for the fusion imaging. This ultrasound system uses a matrix target line model to obtain fusion accuracy, and the theoretical accuracy is 0.185 mm, which is the diameter of the target line. The fusion imaging technique allows surgeons to correctly locate the puncture point and the target point, and perform operations in real time. During simulation of the actual scanning, six motion patterns were applied to the probe for correction and registration, including translation of the probe along the direction perpendicular to the scanning plane, translation of the probe along the direction parallel to the scanning plane, translation of the probe along the direction of the acoustic beam emitted from the transducer, swinging of the probe along the direction parallel to the scanning plane, swinging of the probe along the direction perpendicular to the scanning plane, and rotation of the probe along the direction of the acoustic beam emitted from the transducer. Three images are acquired in each direction to ensure that the images are fully fused.

The puncture is the most important technique in PVP. Whether or not a precise puncture is truly achieved can directly affect the distribution of the bone cement and treatment efficacy after surgery. For punctures performed under traditional X-ray fluoroscopy, the doctors and patients both receive radiation exposure, and the accuracy of the position reached by the puncture needle cannot be observed. Furthermore, the needle angle and path often need to be repeatedly adjusted, which may cause trauma to the patient and increase the radiation exposure to the doctor and patient. In the current study, we included patients with osteoporotic compression fractures and used fusion imaging for localization and guidance of the puncture procedure during PVP. The fusion imaging provided the following advantages. 1) A CT image is used to determine the puncture entry point, path, and target point, and to select the pedicles at their largest diameter, which facilitates intuitive determination of the safe and rational application of the puncture, ensuring that the puncture entry point is accurate, the puncture path is safe, and the target point is reasonable. 2) Fusion of ultrasound and CT images allows surgeons to perform the puncture in real time. The application of the fusion imaging technique with magnetic navigation realizes real-time monitoring of the procedures under ultrasound guidance, which is safe and accurate.

In the current study, accurate punctures into the target were performed in all seven patients, the duration of the radiation exposure was reduced, and the procedure time for the puncture was shortened. 3) Accurate puncturing of the target point realizes a reasonable distribution of bone cement, which improves the treatment efficacy and reduces the occurrence of bone cement leakage. Therefore, the fusion imaging technique has a wide range of potential applications in musculoskeletal diseases. The fusion imaging technique allows determination of the optimal target point; for patients with tumor, the target point is selected directly at the bone destruction zone, while for patients with fracture, the target point is selected as the safe area closest to the fracture. Fusion imaging for PVP provides the advantages of accurate localization and greater accuracy during surgery.

However, attention should be paid to the following issues when performing fusion imaging. 1) During preoperative CT scanning, the patient should be placed in the same position that will be used during surgery; otherwise, an accurate image fusion cannot be achieved. 2) The skin surface locator should be properly placed within the scanning range and should not affect the operation. In the present study, the image fusion required 30 minutes in one patient in the CT-ultrasound fusion imaging group, with the reason for this being that the surface locator was placed 5 cm away from the operation area, which caused incomplete information collection during the CT scanning, resulting in failure of the automatic fusion and the necessity to adopt manual fusion. Generally, the skin surface locator should be placed on the opposite side of the puncture site, 2–3 cm away from the spinous process. If the distribution of bone cement injected via the unilateral approach is not good and the bilateral puncture approach is used, then the surface locator should be removed. 3) Fusion imaging applies a magnetic field effect, which is a contraindication in patients with pacemakers. Furthermore, metal objects may affect the fusion effect, and should be removed prior to image fusion. 4) Before puncture, puncture depth should be accurately measured and the appropriate needle length selected. As the puncture needle needs to be equipped with a magnetic sensor, the needle length should be 2 cm longer than the expected length, and entering of the correct needle length data into the ultrasound machine can help achieve accurate positioning. 5) During the puncture, the puncture needle should be advanced strictly according to the designed puncture path. To ensure that the puncture needle reaches the target point, the needle position should be verified in multiple directions by moving the ultrasound probe. After confirmation of the needle reaching the target point on the sagittal CT view, a C-arm X-ray system was used to further verify the correct needle position, before injection of the bone cement.

This study has some limitations worthy of note. The number of patients included in the current study is small, and a further study with a larger sample size is needed to confirm the operation times. Furthermore, although fusion imaging can achieve accurate localization of the puncture site and provide guidance for the puncture, the technique cannot monitor the injection of bone cement, although monitoring of the bone cement injection under ultrasound could be achieved through detection of the distribution of bone cement in the vertebral body by its heat dissipation properties, allowing radiation exposure to be further reduced. Fusion imaging should be performed by ultrasound physicians who are highly experienced with musculoskeletal ultrasound, and the use of fusion imaging in sports-related injuries requires further investigation.

Conflict of interest

The authors declare that they have no conflict of interest.

References

Yang

H.L.

, Li

, Wang

G.L.

, Meng

, Percutaneous kyphoplasty for bone cement leakage. Zhonghua Jiepou Yu Linchuang Zazhi 20(5) (2015), 377–378.

Kallmes

D.F.

, Roy

S.S.

, Piccolo

R.G.

, et al., Radiation dose to the operator during vertebroplasty: Prospective comparison of the use of 1-cc syringes versus an injection device. AJNR Am J Neuroradiol 24(6) (2003), 1257–1260.

Kruger

, Faciszewski

, Radiation dose reduction to medical staff during vertebroplasty: A review of techniques and methods to mitigate occupational dose. Spine (Phila Pa 1976) 28(14) (2003), 1608–1613.

Belkoff

S.M.

, Molloy

, Temperature measurement during polymerization of polymethylmethacrylate cement used for vertebroplasty. Spine (Phila Pa 1976) 28(14) (2003), 1555–1559.

Yang

X.M.

, Wu

T.L.

, Xu

H.G.

, et al., Modified unilateral transpedicular percutaneous vertebroplasty for treatment of osteoporotic vertebral compression fractures. Orthop Surg. 3(4) (2011), 247–252.

Jang

J.S.

, Lee

S.H.

, Jung

S.K.

, Pulmonary embolism of polymethylmethacrylate after percutaneous vertebroplasty: A report of three cases. Spine (Phila Pa 1976) 27(19) (2002), E416–418.

Theocharopoulos

, Perisinakis

, Damilakis

, et al., Occupational exposure from common fluoroscopic projections used in orthopaedic surgery. J Bone Joint Surg Am 85-A(9) (2003), 1698–1703.

Tohmeh

A.G.

, Mathis

J.M.

, Fenton

D.C.

, et al., Biomechanical efficacy of unipedicular versus bipedicular vertebroplasty for the management of osteoporotic compression fractures. Spine (Phila Pa 1976) 24(17) (1999), 1772–6.

Steinmann

, Tingey

C.T.

, Cruz

, Dai

, Biomechanical comparison of unipedicular versus bipedicular kyphoplasty. Spine (Phila Pa 1976) 30(2) (2005), 201–5.

10.

Chen

, Li

, Xie

, et al., Comparison of unipedicular and bipedicular kyphoplasty on the stiffness and biomechanical balance of compression fractured vertebrae. Eur Spine J. 20(8) (2011), 1272–80.

11.

Chung

H.J.

, Chung

K.J.

, Yoon

H.S.

, Kwon

I.H.

, Comparative study of balloon kyphoplasty with unilateral versus bilateral approach in osteoporotic vertebral compression fractures. Int Orthop. 32(6) (2008), 817–20.

12.

Hoppe

, Elfiky

, Keel

M.J.

, et al., Lavage prior to vertebral augmentation reduces the risk for cement leakage. Eur Spine J. 25(11) (2016), 3463–3469.

13.

K.L.

, Zhang

Y.D.

, Zhuang

J.T.

, et al., Subinguinal microsurgical varicocelectomy with intraoperative microvascular Doppler ultrasound leads to the pain-free outcome after surgery. J Xray Sci Technol 25(5) (2017), 839–846.

14.

Hall

M.M.

, Mautner

, Evolution of musculoskeletal and non-musculoskeletal sports ultrasound. Br J Sports Med. 49(3) (2015), 139–40.

15.

Gui

X.G.

, Ru

X.L.

, Jiang

Z.H.

, Song

B.S.

, [Analysis of perioperative complications of percutaneous kyphoplasty for osteoporotic vertebral compression fracture]. Zhongguo Gu Shang. 26(3) (2013), 205–9.

16.

Grelat

, Le Van

and Fahed

, et al., Rare complication of percutaneous technique: intradural cement leakage and its surgical treatment. World Neurosurg. 118 (2018), 97.

17.

Valerio

, Donaldson

, Emberton

, et al., Detection of clinically significant prostate cancer using magnetic resonance imaging-ultrasound fusion targeted biopsy: A systematic review. Eur Urol. 68(1) (2015), 8–19.

18.

Minami

, Minami

, Chishina

, et al., Combination guidance of contrast-enhanced US and fusion imaging in radiofrequency ablation for hepatocellular carcinoma with poor conspicuity on contrast-enhanced US/fusion imaging. Oncology. 87(Suppl 1) (2014), 55–62.

19.

Sandulescu

D.L.

, Dumitrescu

, Rogoveanu

, Saftoiu

, Hybrid ultrasound imaging techniques (fusion imaging). World J Gastroenterol. 17(1) (2011), 49–52.

20.

Finnoff

J.T.

, Berkoff

, Brennan

, et al., American Medical Society for Sports Medicine (AMSSM) recommended sports ultrasound curriculum for sports medicine fellowships. PM R 7(2) (2015), e1–e11.