Advanced Ultrasound Imaging

Molecular Photoacoustic Imaging of Prostate Cancer

PA
A sensitive, non-invasive method to detect localized prostate cancer, particularly for repetitive study in patients undergoing active surveillance, remains an unmet need. We propose a photoacoustic (PA) imaging approach by targeting the prostate-specific membrane antigen (PSMA), which is over-expressed in the vast majority of prostate cancers. Through in vivo demonstration on an experiment with mice model, spectroscopic PA imaging can visualize the targeted agent, YC-27, that binds on PSMA+ tumor.

Contact: Haichong “Kai” Zhang (hzhang61@jhu.edu>)
Collaborators: Dr. Emad M. Boctor, Dr. Martin G. Pomper, Dr. Sangeeta Ray
Sponsor: NIH R21, DoD CDMRP
Related publication:
H. K. Zhang, Y. Chen, J. Kang, A. Lisok, I. Minn, M. G. Pomper, and E. M. Boctor, “Prostate Specific Membrane Antigen (PSMA)-Targeted Photoacoustic Imaging of Prostate Cancer In Vivo”, in Journal of Biophotonics (accepted).

Real-time Photoacoustic Imaging

PA
PA
Channel data acquisition and the synchronization between laser firing and transducer signal acquisition are two fundamental hardware requirements for photoacoustic (PA) imaging. Unfortunately, neither is equipped by most clinical ultrasound scanners, which makes research platforms expensive. In this work, we focus on enabling clinical ultrasound scanners to implement PA imaging, without requiring synchronization between laser firing and PA signal acquisition. Two elements of synchronization, frequency information and phase information, are recovered through solving a nonlinear optimization problem and binary search. Compared to the ground truth, remaining errors in frequency and phase correction are 0.28% and 2.79%, respectively.

Contact: Yixuan Wu (yixuan_wu@jhu.edu)
Collaborators: Dr. Haichong K. Zhang, Dr. Emad M. Boctor
Related publication:
Y. Wu, H. K. Zhang, and E. M. Boctor “Enabling vendor independent photoacoustic imaging systems with asynchronous laser source” in Photons Plus Ultrasound: Imaging and Sensing 2018 (Vol. 10494, p. 104945M). International Society for Optics and Photonics. 2018

BeeSpaceMouse

beespacemouseHandheld 2D ultrasound is very useful for intra-operative imaging, but requires some reconstruction effort in order to create 3D US volumes, unless one is using large and expensive 3D US probes. Common probe tracking approaches involve either global tracking mechanisms (suffering from jitter and complexity) or local tracking algorithms (e.g. image-based, suffering from drift).

Our BeeSpaceMouse uses a combination of local sensors to reconstruct the probe trajectory with multiple degrees of freedom. These sensors can be very low-cost – e.g. optical mice and a Wii Remote – yet enable simple 3D US acquisition with no global tracking overhead. The resulting trajectories can be used as input for 3D US algorithms or constrain image-based approaches.

 

Bone Segmentation in Ultrasound Images

IboneSegmentationn this work, we developed an intuitive and computationally inexpensive bone segmentation approach. Prior knowledge about the appearance of bone in ultrasound images is exploited toward achieving robust and fast bone segmentation. Continuity and smoothness of the bone surface are incorporated in a cost function, which is globally minimized using dynamic programming.

The performance of this method is evaluated on ultrasound images collected from two male cadavers. The images are segmented in about half a second making the algorithm suitable for real-time applications.

 

Tracked Ultrasound: Bone structure detection and registration

Dr. Chen Lei, Dr. Sungmin Kim, Dr. Peter Kazanzides, Dr. Russell H. Taylor, Dr. Emad M. Boctor

FatUSUltrasound imaging has been proposed to eliminate invasive collection of bone points for registration in robotic orthopedic surgery. However, the variation of ultrasound speed in different tissues introduces significant errors in the localization of the bone surface. We present a speed of sound correction approach based on automated segmentation of bone and fat on ultrasound B-mode images. The bone structure is automatically detected by using a difference of Gaussian filter-based approach. The fat area on the ultrasound B-mode image is automatically segmented with a fast marching method. The depth of bone and fat are then used to correct the average speed of sound that is the default setting on the ultrasound machine.

View paper

 

Doppler Ultrasound: Echosure

Dr. Chen Lei, Dr. Jerry Prince, Dr. Emad M. Boctor,

EchoSureDoppler imaging provides information such as the flow direction and the relative velocity towards or away from the probe. Information can then be used to quantify the arterial stenosis. However, given that it provides velocity data that is dependent on the location and angle of measurement, repeat measurements to detect problems over time may require an expert to return to the same location. We therefore developed an image-guidance system [1] based on ultrasound B-mode images that enables an inexperienced user to position the ultrasound probe at the same site repeatedly in order to acquire a comparable time series of Doppler readings. The system utilizes a bioresorbable fiducial and complementing software composed of the fiducial detection, key points tracking, probe pose estimation, and graphical user interface (GUI) modules.

View paper

 

Prostate Cancer Imaging

Photoacoustic-based Brachytherapy Guidance

Photoacoustic Compounding/Robotic Tracking

Thermal Imaging