The PASCAL Visual Object Classes Challenge

The volume and complexity of diagnostic imaging is increasing at a pace faster than the availability of human expertise to interpret it. Artificial intelligence has shown great promise in classifying two-dimensional photographs of some common diseases and typically relies on databases of millions of annotated images. Until now, the challenge of reaching the performance of expert clinicians in a real-world clinical pathway with three-dimensional diagnostic scans has remained unsolved. Here, we apply a novel deep learning architecture to a clinically heterogeneous set of three-dimensional optical coherence tomography scans from patients referred to a major eye hospital. We demonstrate performance in making a referral recommendation that reaches or exceeds that of experts on a range of sight-threatening retinal diseases after training on only 14,884 scans. Moreover, we demonstrate that the tissue segmentations produced by our architecture act as a device-independent representation; referral accuracy is maintained when using tissue segmentations from a different type of device. Our work removes previous barriers to wider clinical use without prohibitive training data requirements across multiple pathologies in a real-world setting.

/pdf/clinically-applicable-deep-learning-for-diagnosis-and-56r182cw0z.pdf

Clinically applicable deep learning for diagnosis and referral in retinal disease

Objective: To demonstrate optical coherence tomography for high-resolution, noninvasive imaging of the human retina. Optical coherence tomography is a new imaging technique analogous to ultrasound B scan that can provide cross-sectional images of the retina with micrometer-scale resolution. Design: Survey optical coherence tomographic examination of the retina, including the macula and optic nerve head in normal human subjects. Settings Research laboratory. Participants: Convenience sample of normal human subjects. Main Outcome Measures: Correlation of optical coherence retinal tomographs with known normal retinal anatomy. Results: Optical coherence tomographs can discriminate the cross-sectional morphologic features of the fovea and optic disc, the layered structure of the retina, and normal anatomic variations in retinal and retinal nerve fiber layer thicknesses with 10- μm depth resolution. Conclusion: Optical coherence tomography is a potentially useful technique for high depth resolution, cross-sectional examination of the fundus.

Optical Coherence Tomography of the Human Retina

Medical Image Computing and Computer-Assisted Intervention

The diffraction tomography theorem is adapted to one-dimensional length measurement. The resulting spectral interferometry technique is described and the first length measurements using this technique on a model eye and on a human eye in vivo are presented.

Measurement of intraocular distances by backscattering spectral interferometry

In this paper, we suggest a novel data-driven approach to active learning (AL). The key idea is to train a regressor that predicts the expected error reduction for a candidate sample in a particular learning state. By formulating the query selection procedure as a regression problem we are not restricted to working with existing AL heuristics; instead, we learn strategies based on experience from previous AL outcomes. We show that a strategy can be learnt either from simple synthetic 2D datasets or from a subset of domain-specific data. Our method yields strategies that work well on real data from a wide range of domains.

Learning Active Learning from Data

/pdf/learning-active-learning-from-data-47vsizvacp.pdf

Instrument detection, pose estimation, and tracking in surgical videos are an important vision component for computer-assisted interventions. While significant advances have been made in recent years, articulation detection is still a major challenge. In this paper, we propose a deep neural network for articulated multi-instrument 2-D pose estimation, which is trained on detailed annotations of endoscopic and microscopic data sets. Our model is formed by a fully convolutional detection-regression network. Joints and associations between joint pairs in our instrument model are located by the detection subnetwork and are subsequently refined through a regression subnetwork. Based on the output from the model, the poses of the instruments are inferred using maximum bipartite graph matching. Our estimation framework is powered by deep learning techniques without any direct kinematic information from a robot. Our framework is tested on single-instrument RMIT data, and also on multi-instrument EndoVis and in vivo data with promising results. In addition, the data set annotations are publicly released along with our code and model.

Articulated Multi-Instrument 2-D Pose Estimation Using Fully Convolutional Networks

We consider the problem of 20 questions with noisy answers, in which we seek to nd a target by repeatedly choosing a set, asking an oracle whether the target lies in this set, and obtaining an answer corrupted by noise. Starting with a prior distribution on the target’s location, we seek to minimize the expected entropy of the posterior distribution. We formulate this problem as a dynamic program and show that any policy optimizing the one-step expected reduction in entropy is also optimal over the full horizon. Two such Bayes-optimal policies are presented: one generalizes the probabilistic bisection policy due to Horstein and the other asks a deterministic set of questions. We study the structural properties of the latter, and illustrate its use in a computer vision application.

/pdf/twenty-questions-with-noise-bayes-optimal-policies-for-1yx63m90d4.pdf

Twenty questions with noise: bayes optimal policies for entropy loss

The swimming behavior of the nematode Caenorhabditis elegans is investigated in aqueous solutions of increasing viscosity. Detailed flow dynamics associated with the nematode’s swimming motion as well as propulsive force and power are obtained using particle tracking and velocimetry methods. We find that C. elegans delivers propulsive thrusts on the order of a few nanonewtons. Such findings are supported by values obtained using resistive force theory; the ratio of normal to tangential drag coefficients is estimated to be approximately 1.4. Over the range of solutions investigated here, the flow properties remain largely independent of viscosity. Velocity magnitudes of the flow away from the nematode body decay rapidly within less than a body length and collapse onto a single master curve. Overall, our findings support that C. elegans is an attractive living model to study the coupling between small-scale propulsion and low Reynolds number hydrodynamics.

https://repository.upenn.edu/cgi/viewcontent.cgi?article=1274&context=meam_papers

Raphael Sznitman

Papers

Learning Active Learning from Data

Learning Active Learning from Data

Articulated Multi-Instrument 2-D Pose Estimation Using Fully Convolutional Networks

Twenty questions with noise: bayes optimal policies for entropy loss

Propulsive force measurements and flow behavior of undulatory swimmers at low Reynolds number