Temporally averaged speckle noise in wavefront sensors for beam projection in weak turbulence.

Adaptive optics (AO) compensation for imaging or coherent illumination of a remote object relies on accurate sensing of atmospheric aberrations. When a coherent beacon is projected onto the object to enable wavefront sensing, the reflected reference wave will exhibit random variation in phase and amplitude characteristics of laser speckle. In a Shack-Hartmann wavefront sensor (SHWFS) measurement, speckle effects cause fluctuations in the intensity of focal spots and errors in the position of their centroids relative to those expected from purely atmospheric phase aberrations. The resulting error in wavefront measurements negatively impacts the quality of atmospheric phase conjugation. This paper characterizes the effect of reflected laser speckle on the accuracy of SHWFS measurements for ground-to-space beam projection systems in weak turbulence conditions. We show via simulation that the speckle-induced error in centroiding depends on the ratio between beacon diameter and the diffraction-limited resolution of the lenslet and confirm these results with experimental data. We provide experimental validation that averaging of SHWFS lenslet spot intensities over speckle realizations converges to the incoherent intensity as expected. We further show that the effects of shot noise and speckle noise add in quadrature, simplifying noise analysis. Finally, we characterize the effect of temporal averaging under typical conditions of target motion and integration time. This work provides a straightforward set of relations that can help investigators more accurately estimate the required integration time for wavefront sensing in the presence of laser speckle.

Deep residual learning for low-order wavefront sensing in high-contrast imaging systems.

Sensing and correction of low-order wavefront aberrations is critical for high-contrast astronomical imaging. State of the art coronagraph systems typically use image-based sensing methods that exploit the rejected on-axis light, such as Lyot-based low order wavefront sensors (LLOWFS); these methods rely on linear least-squares fitting to recover Zernike basis coefficients from intensity data. However, the dynamic range of linear recovery is limited. We propose the use of deep neural networks with residual learning techniques for non-linear wavefront sensing. The deep residual learning approach extends the usable range of the LLOWFS sensor by more than an order of magnitude compared to the conventional methods, and can improve closed-loop control of systems with large initial wavefront error. We demonstrate that the deep learning approach performs well even in low-photon regimes common to coronagraphic imaging of exoplanets.

MEMS Deformable Mirrors for Space-Based High-Contrast Imaging.

Micro-Electro-Mechanical Systems (MEMS) Deformable Mirrors (DMs) enable precise wavefront control for optical systems. This technology can be used to meet the extreme wavefront control requirements for high contrast imaging of exoplanets with coronagraph instruments. MEMS DM technology is being demonstrated and developed in preparation for future exoplanet high contrast imaging space telescopes, including the Wide Field Infrared Survey Telescope (WFIRST) mission which supported the development of a 2040 actuator MEMS DM. In this paper, we discuss ground testing results and several projects which demonstrate the operation of MEMS DMs in the space environment. The missions include the Planet Imaging Concept Testbed Using a Recoverable Experiment (PICTURE) sounding rocket (launched 2011), the Planet Imaging Coronagraphic Technology Using a Reconfigurable Experimental Base (PICTURE-B) sounding rocket (launched 2015), the Planetary Imaging Concept Testbed Using a Recoverable Experiment - Coronagraph (PICTURE-C) high altitude balloon (expected launch 2019), the High Contrast Imaging Balloon System (HiCIBaS) high altitude balloon (launched 2018), and the Deformable Mirror Demonstration Mission (DeMi) CubeSat mission (expected launch late 2019). We summarize results from the previously flown missions and objectives for the missions that are next on the pad. PICTURE had technical difficulties with the sounding rocket telemetry system. PICTURE-B demonstrated functionality at >100 km altitude after the payload experienced 12-g RMS (Vehicle Level 2) test and sounding rocket launch loads. The PICTURE-C balloon aims to demonstrate 10 - 7 contrast using a vector vortex coronagraph, image plane wavefront sensor, and a 952 actuator MEMS DM. The HiClBaS flight experienced a DM cabling issue, but the 37-segment hexagonal piston-tip-tilt DM is operational post-flight. The DeMi mission aims to demonstrate wavefront control to a precision of less than 100 nm RMS in space with a 140 actuator MEMS DM.

Automatic Quality Assessment of Echocardiograms Using Convolutional Neural Networks: Feasibility on the Apical Four-Chamber View.

Echocardiography (echo) is a skilled technical procedure that depends on the experience of the operator. The aim of this paper is to reduce user variability in data acquisition by automatically computing a score of echo quality for operator feedback. To do this, a deep convolutional neural network model, trained on a large set of samples, was developed for scoring apical four-chamber (A4C) echo. In this paper, 6,916 end-systolic echo images were manually studied by an expert cardiologist and were assigned a score between 0 (not acceptable) and 5 (excellent). The images were divided into two independent training-validation and test sets. The network architecture and its parameters were based on the stochastic approach of the particle swarm optimization on the training-validation data. The mean absolute error between the scores from the ultimately trained model and the expert's manual scores was 0.71 ± 0.58. The reported error was comparable to the measured intra-rater reliability. The learned features of the network were visually interpretable and could be mapped to the anatomy of the heart in the A4C echo, giving confidence in the training result. The computation time for the proposed network architecture, running on a graphics processing unit, was less than 10 ms per frame, sufficient for real-time deployment. The proposed approach has the potential to facilitate the widespread use of echo at the point-of-care and enable early and timely diagnosis and treatment. Finally, the approach did not use any specific assumptions about the A4C echo, so it could be generalizable to other standard echo views.

Simultaneous Analysis of 2D Echo Views for Left Atrial Segmentation and Disease Detection.

We propose a joint information approach for automatic analysis of 2D echocardiography (echo) data. The approach combines a priori images, their segmentations and patient diagnostic information within a unified framework to determine various clinical parameters, such as cardiac chamber volumes, and cardiac disease labels. The main idea behind the approach is to employ joint Independent Component Analysis of both echo image intensity information and corresponding segmentation labels to generate models that jointly describe the image and label space of echo patients on multiple apical views, instead of independently. These models are then both used for segmentation and volume estimation of cardiac chambers such as the left atrium and for detecting pathological abnormalities such as mitral regurgitation. We validate the approach on a large cohort of echoes obtained from 6,993 studies. We report performance of the proposed approach in estimation of the left-atrium volume and detection of mitral-regurgitation severity. A correlation coefficient of 0.87 was achieved for volume estimation of the left atrium when compared to the clinical report. Moreover, we classified patients that suffer from moderate or severe mitral regurgitation with an average accuracy of 82%.

Rethinking Education: When surgeons and engineering students join forces to solve real problems, success follows.

The Engineers in Scrubs (EiS) training program at the University of British Columbia, affiliated with the Faculty of Applied Science?s Biomedical Engineering Graduate (BMEG) Program, is not a typical graduate school course. Nor does it follow a traditional master?s course rubric that culminates with a tidy end-of-year project. Rather, the course is designed to push students to prototype innovative medical devices, encourage health care collaborations, and create an unprecedented interface between technology and health care to further medicine.

Investigating the performance of a wrist stabilization device for image-guided percutaneous scaphoid fixation.

PURPOSE: Conventional navigated surgery relies on placement of a reference marker on the anatomy of interest. However, placement of such a marker is not readily feasible in small anatomic regions such as the scaphoid bone of the wrist. This study aimed to develop an alternative mechanism for patient tracking that could be used to perform navigated percutaneous scaphoid fixation. METHODS: A prototype wrist stabilization device was developed to immobilize the scaphoid relative to a reference marker attached to the device. A position measurement system and 3D fluoroscopy were used to study the accuracy and limitations of wrist stabilization during simulated clinical usage with a cadaver specimen. Reference markers mounted on the device were used to measure intra-device motion. Radiometallic beads implanted in the scaphoid were used to measure patient-device motion. Navigated planning and guidance of scaphoid fixation were performed in five cadaver and eight "ideally immobilized" plastic specimens. Postoperative 3D fluoroscopy was used to assess the accuracy of navigated drilling. RESULTS: The average intra-device motion was 1.9 mm during load application, which was elastically recovered upon release of the load. Scaphoid motion relative to the reference marker was predominately rotational with an average displacement of 1.25 mm and 2.0°. There was no significant difference in the accuracy of navigated drilling between the cadaver specimens and the ideally immobilized group. CONCLUSIONS: The prototype wrist stabilization device meets the criteria for effective wrist stabilization. This study provides insight concerning proper use of the device to minimize scaphoid displacement and design recommendations to improve immobilization.