A new 3D ultrasound imaging guidance method for prostate biopsy is a featured article in the IEEE Transactions on Biomedical Engineering, a leading journal in the biomedical technology sector. A figure from this research appears on the front page of the April 2017 edition. 

Screen shots of the GUI developed for biopsy guidance. (Top) Registered gland (yellow mesh) and target (yellow sphere) appeared on the monitor after registration. (Bottom) Needle trajectory hits the target while the target changes the colour to green.

The work, led by Dean Barratt, and carried out by Yipeng Hu from the UCL Centre for Medical Image Computing, in collaboration with colleagues from UCL Medical Physics & Biomedical Engineering and UCL Hospital NHS Trust, employs 3D ultrasound imaging, prostate motion modelling, and high-accuracy optical tracking of the probe and needle typically used for this procedure. 

Over 47,000 men are diagnosed with prostate cancer every year in the UK. The current standard for diagnosing the disease involves collecting tissue samples from the prostate using a biopsy needle guided by transrectal ultrasound (TRUS). However, since prostate tumours are rarely visible in TRUS images, biopsy is essentially performed in a “blind” manner - i.e. without knowing where tumours are - which compromises the diagnostic performance of this approach. 

Recent studies have shown MRI scans to be an effective way to identify prostate tumours that are then targeted during biopsy. In practice, computer algorithms can be used to overlay a graphical representation of MRI-identified tumours onto the TRUS images that provide visible targets. The biopsy needle can then be navigated to collect sample tissues directly from these locations.

3D/4D ultrasound transducers can provide rapid volumetric imaging of the entire prostate gland during US-guided biopsies. Although these transducers are now widely available from ultrasound equipment manufacturers, they can cause significant errors in electromagnetic position tracking devices that are increasingly used for 3D US probe tracking in commercial image-guided navigation systems. 

The new publication outlines the design, development, and validation of the use of high-accuracy optical tracking as a solution to this problem, which has been integrated into a complete guidance system.

Optical tracking is able to offer real-time feedback on the 3D position and orientation of the ultrasound probe and the predicated biopsy needle trajectory, with respect to the MR-identified tumour locations. Progress is calculated and presented as a graphical display on the PC which is kept up to date in real-time during the procedure. 

The on screen feedback enables doctors to orientate the probe and needle to the optimum trajectory to reach the target in the prostate gland. The validation results of the paper showed a significant increase in needle placement accuracy using the 3D guidance system compared to visual targeting of lesions using a standard ultrasound-guided biopsy techniques.

It is hoped these results will be able to offer increased accuracy during biopsy procedures for patients who have been pre-assessed with MRI. 

All software was developed using a combination of MATLAB (The Mathworks, Cambridge, MA, USA) and C++.