Search

Early vessel wall thickening is seen as an indicator for the development of cerebrovascular disease. Quantification of wall thickening using conventional measurement methods is difficult owing to the relatively thin vessel wall thickness compared to the acquired MR voxel size. We hypothesize that a convolutional neural network (CNN), can incorporate spatial orientation, shape, and intensity distribution of the vessel wall in an accurate thickness estimation for subvoxel walls. MR imaging of 34 post-mortem specimens was performed using a 3D gradient echo protocol (isotropic acquired voxel size: 0.11 mm; acquisition time: 5h46m). Simulating clinically feasible resolutions, image patches were sampled at a clinically feasible isotropic voxel size of 0.8 mm (patch size: 113 voxels). Image patches were sampled centered around vessel wall voxels where the wall thickness of the center voxel was measured at the original resolution using a validated measurement method. The image patches were fed into our CNN, which consisted of five subsequent 3D convolutional layers, followed by two fully connected layers feeding into the linearly activated output layer. Our network can distinguish walls with a target thickness between 0.2-1.0 mm. In this range, the median offset between the target thickness and estimated thickness is 0.14 mm (interquartile range: 0.22 mm). For walls with a target thickness below and above half the voxel size (0.4 mm), the median offset is 0.17 mm and 0.10 mm, respectively. In conclusion, our results show that a CNN can accurately measure the thickness of subvoxel vessel walls, down to half the voxel size.

Photoacoustic (PA) or optoacoustic (OA) imaging combines the high (blood) contrast to light with the high-resolution of ultrasound. The method can visualize vascularization deep inside tissue. Of late there is interest in PA imaging of synovial joints which are expected to be associated with increased vascularization in the event of rheumatoid arthritis (RA). We here describe our approach in investigating the application of the PA technique in arthritis. We are developing a CT-geometry version PA finger imager, intended for early clinical assessment of the method. The imager uses two curved array ultrasound detectors each with 64 elements with central frequencies 1.5 and 7.5 MHz respectively, stacked above each other. Both cover approximately 180 degrees of the circle. Illumination is provided with a multiple of optical fiber bundles coupled to a laser-OPO system. Ultrasound imaging is also possible with the system, since the curved arrays are each provided with 12 or 8 ultrasound pulsers. We have investigated systematically imaging of finger vasculature in healthy volunteers using an earlier laboratory prototype. In this paper we present finger imaging results of a patient diagnosed with rheumatoid arthritis.

We present a 'hybrid' imaging system, which can image both optical absorption properties and acoustic transmission properties of an object in a two-dimensional slice using a computed tomography photoacoustic imager. The ultrasound transmission measurement method uses a strong absorber of light which is placed in the path of the light illuminating the sample. This acts as a source of ultrasound, whose interaction with the sample can be measured at the far-end of the sample using the same ultrasound detector used for photoacoustics. Such measurements are made at various angles around the sample in a computerized tomography approach. The ultrasound transients at the multi-element detector at all projections are analyzed for both times-of-arrival and amplitude. Using a fan-beam projection reconstruction algorithm we obtain hybrid images of optical absorption, speed-of-sound and acoustic attenuation. We validate the method on an appropriate phantom

Breast cancer dominates cancers in females. This burden on society and the room for improvements in the current practice of mammography have been stimuli for developing new modalities like photoacoustic mammography. At the University of Twente (UT), an instrument had been developed aimed at performing limited area scans on the human breast. This instrument is called the Twente Photoacoustic Mammoscope (PAM). The PAM is based on generating laser-induced ultrasound from absorbing structures in the breast. The heart of the instrument is a flat PVDF based detector matrix comprising 590 active elements. We show the performance characteristics of the ultrasound detector. The exciting source is an Nd:YAG laser operating at 1064 nm with 5 ns pulses. A study protocol was designed to explore the feasibility of using the PAM to detect cancer in the breasts of patients. The protocol was executed at the Medisch Spectrum Twente by using the mammoscope to obtain photoacoustic region-of-interest (ROI) images of the suspect/symptomatic breasts. We compare the photoacoustic images obtained with x-ray mammograms and ultrasound images. We show photoacoustic images of ROI in one case where we attribute high intensity regions to tumor vascularization.

The Twente Photoacoustic Mammoscope (PAM) is based on generating laser-induced ultrasound from absorbing structures in the breast. The heart of the instrument is a flat PVDF based detector matrix comprising 590 active elements. The exciting source is an Nd:YAG laser operating at 1064 nm with 5 ns pulses. The instrument is built around a hospital bed. A study protocol was designed to explore the feasibility of using the photoacoustic technique as embodied in PAM to detect cancer in the breasts of patients with suspect/symptomatic breasts. The protocol was approved by a Medical Ethics testing committee and the instrument approved for laser and electrical safety. The protocol was executed at the Medisch Spectrum Twente by using the mammoscope to obtain photoacoustic region-of-interest (ROI) images of the suspect/symptomatic breasts. We report on one case and compare the photoacoustic images obtained with x-ray mammograms and ultrasound images.

Performance studies of a clinical prototype for detecting tumors in the breast based on the photoacoustic effect are presented in terms of sensitivity, frequency response and resolution. Some imaging results on well characterized breast tissue phantoms with embedded tumor simulating inserts are also shown.

A novel photoacoustic breast phantom was developed using poly(vinyl alcohol) gel prepared by a simple technique that imparts optical scattering to the gel without the neccessity for scattering particles. Tumour simulating gel samples of suitable absorption coefficient were also prepared using a second technique, by adding appropriate quantities of dye at the time of formation; the samples were then cut into spheres. The optical absorption coefficient of the spheres was chosen as between 4 - 7 times that of breast tissue. A breast phantom with a thickness of 60 mm, embedded with such 'tumours' was developed for studying the applicability of photoacoustics in mammography. Time resolved photoacoustics, in a transmission mode, was used to image the inhomogeneities. Light excitation was from a liquid-light guide coupled Nd:YAG laser at 1064 nm, with a 5 ns pulse duration. The guide was mechanically scanned across the surface of the phantom, with the time-of-flight signals recorded using a PVDF based detector array. A modified delay and sum beamforming algorithm was used to reconstruct the photoacoustic sources. Results of these experiments are discussed.