Search

Cardiac arrhythmias are caused by irregular activation in the heart. When pharmaceutical therapy fails, severe arrythmias are often treated with radiofrequency ablation. Existing methods to map electrical activation in the heart have poor spatial resolution or are invasive. We proposed Acoustoelectric Cardiac Imaging (ACI) to achieve non-invasive mapping of cardiac electrical activities with mm resolution. In this study, we used simulations to develop a reconstruction method for generating current density map from multi-channel ACI signals, and applied it to in-vivo ACI data from a swine heart with a resolution less than 5mm. With 5 leads at SNR from 55 to 80, the five differential signals could be combined in simulation to reconstruct the source current density map, with a correlation coefficient ≈ 0.74 for the x-component. For in-vivo data, we find that singular values decomposition provides a criterion for reconstruction.

Lipids play a vital role in brain function; Alzheimer's disease and aging are both associated with changes in lipid composition. Current methods for imaging and monitoring lipid deposits within the brain have poor spatial resolution, sensitivity, and/or penetration depth. Photoacoustic imaging (PA/PAI) allows for deep (>5mm) mapping of optical absorption_properties of biological tissue.; however, water and lipid have relatively weak optical absorption properties in the typical near-infrared window used for PAI (700-1000nm). This study investigates PAI in the short-wave infrared (SWIR) range (>1200nm) as a method to dramatically enhance lipid and water contrast. Due to multiple strong absorption peaks within this optical window, a three wavelength unmixing technique labeled Tri-λ Concentration Mapping (CM) is shown to have the ability to quantify lipid and water content several millimeters into tissue. Values of lipid and water concentration are calculated at voxels within sections of marbled bovine muscular tissue. Investigated regions of muscular meat show a bimodal distribution about 75% and 50% predicted water concentration. Calculated lipid regions at depths of ~5mm display a more precise distribution centered around a lipid concentration of 57.25%. This technique demonstrates the ability of SWIR PAI and Tri-λ CM to identify and characterize lipid content within tissue at depths greater than 5mm. The results of this study show promise of the technique for use in more advanced studies, such as tracking Alzheimer's disease progression within mice.

Transcranial Acoustoelectric Brain Imaging (tABI) combines ultrasound (US) with radiofrequency detection to map electrical currents in the brain at the mm and ms scales. To optimize tABI and ensure safe exposure, a feedback system is required to monitor and track the US beam through the skull. This study proposes combining a real-time neuronavigation system with acoustic modeling and MRI-based skull segmentation to optimize transcranial US delivery. We developed a testbed to estimate targeting accuracy and algorithms to optimally steer and focus US through the human skull for tABI and other applications that require US delivery to the brain. The optimized system is modeled to deliver focused US to brain targets through the skull with a FWHM of 3.88mm at the focus with steering at sub-mm accuracy, enabling safe and accurate guidance of tABI. The integrated system allows for a prediction of the US beam trajectory through the human skull with registration to clinical images for guidance and improving accuracy of real-time tABI.

As a real-time high resolution imaging modality, pulse echo ultrasound (US) is a common screening tool for breast, thyroid, and other types of cancer. However, US has low specificity for detecting cancer due to poor contrast in soft tissue. Hybrid US techniques like photoacoustic (PA) and thermoacoustic (TA) imaging provide additional sources of contrast for cancer detection but are still mostly independent research tools not readily available in the clinic. The goal of this study was to develop and demonstrate advantages of a real-time trimodal imaging system that combines US, PA, and TA integrated into a commercial US system. Our trimodal system provided real-time feedback on near-infrared absorbing contrast agents, water content, and acoustic backscatter for structure, all important contrast features for accurate diagnostics and feedback during thermal therapy. Real-time video of the pumped Methylene Blue (MB, 0.025% in diH2O) was captured at a velocity resolution of 280 mm/s. Spectrally probed MB resulted in unique PA spectral responses, as well as qualitative images in false colors. Thermometry revealed a correlation between the PA and TA signal responses and MB temperature each with $\mathrm{R}^{2}=0.99$ . By combining all three modalities, the system is capable of detecting different sources of contrast, including acoustic backscatter (US), optical (PA) and microwave (TA) absorption, and changes in temperature (PA and TA).