Semantic learning from keyframe demonstration using object attribute constraints.

Learning from demonstration is an approach that allows users to personalize a robot's tasks. While demonstrations often focus on conveying the robot's motion or task plans, they can also communicate user intentions through object attributes in manipulation tasks. For instance, users might want to teach a robot to sort fruits and vegetables into separate boxes or to place cups next to plates of matching colors. This paper introduces a novel method that enables robots to learn the semantics of user demonstrations, with a particular emphasis on the relationships between object attributes. In our approach, users demonstrate essential task steps by manually guiding the robot through the necessary sequence of poses. We reduce the amount of data by utilizing only robot poses instead of trajectories, allowing us to focus on the task's goals, specifically the objects related to these goals. At each step, known as a keyframe, we record the end-effector pose, object poses, and object attributes. However, the number of keyframes saved in each demonstration can vary due to the user's decisions. This variability in each demonstration can lead to inconsistencies in the significance of keyframes, complicating keyframe alignment to generalize the robot's motion and the user's intention. Our method addresses this issue by focusing on teaching the higher-level goals of the task using only the required keyframes and relevant objects. It aims to teach the rationale behind object selection for a task and generalize this reasoning to environments with previously unseen objects. We validate our proposed method by conducting three manipulation tasks aiming at different object attribute constraints. In the reproduction phase, we demonstrate that even when the robot encounters previously unseen objects, it can generalize the user's intention and execute the task.

Smartphone-based particle tracking velocimetry for the in vitro assessment of coronary flows.

The present study adopts a smartphone-based approach for the experimental characterization of coronary flows. Technically, Particle Tracking Velocimetry (PTV) measurements were performed using a smartphone camera and a low-power continuous wave laser in realistic healthy and stenosed phantoms of left anterior descending artery with inflow Reynolds numbers approximately ranging from 20 to 200. A Lagrangian-Eulerian mapping was performed to convert Lagrangian PTV velocity data to a Eulerian grid. Eulerian velocity and vorticity data obtained from smartphone-based PTV measurements were compared with Particle Image Velocimetry (PIV) measurements performed with a smartphone-based setup and with a conventional setup based on a high-power double-pulsed laser and a CMOS camera. Smartphone-based PTV and PIV velocity flow fields substantially agreed with conventional PIV measurements, with the former characterized by lower average percentage differences than the latter. Discrepancies emerged at high flow regimes, especially at the stenosis throat, due to particle image blur generated by smartphone camera shutter speed and image acquisition frequency. In conclusion, the present findings demonstrate the feasibility of PTV measurements using a smartphone camera and a low-power light source for the in vitro characterization of cardiovascular flows for research, industrial and educational purposes, with advantages in terms of costs, safety and usability.

Predicting Lipid-Rich Plaque Progression in Coronary Arteries Using Multimodal Imaging and Wall Shear Stress Signatures.

BACKGROUND: Plaque composition and wall shear stress (WSS) magnitude act as well-established players in coronary plaque progression. However, WSS magnitude per se does not completely capture the mechanical stimulus to which the endothelium is subjected, since endothelial cells experience changes in the WSS spatiotemporal configuration on the luminal surface. This study explores WSS profile and lipid content signatures of plaque progression to identify novel biomarkers of coronary atherosclerosis. METHODS: Thirty-seven patients with acute coronary syndrome underwent coronary computed tomography angiography, near-infrared spectroscopy intravascular ultrasound, and optical coherence tomography of at least 1 nonculprit vessel at baseline and 1-year follow-up. Baseline coronary artery geometries were reconstructed from intravascular ultrasound and coronary computed tomography angiography and combined with flow information to perform computational fluid dynamics simulations to assess the time-averaged WSS magnitude (TAWSS) and the variability in the contraction/expansion action exerted by WSS on the endothelium, quantifiable in terms of topological shear variation index (TSVI). Plaque progression was measured as intravascular ultrasound-derived percentage plaque atheroma volume change at 1-year follow-up. Plaque composition information was extracted from near-infrared spectroscopy and optical coherence tomography. RESULTS: Exposure to high TSVI and low TAWSS was associated with higher plaque progression (4.00±0.69% and 3.60±0.62%, respectively). Plaque composition acted synergistically with TSVI or TAWSS, resulting in the highest plaque progression (≥5.90%) at locations where lipid-rich plaque is exposed to high TSVI or low TAWSS. CONCLUSIONS: Luminal exposure to high TSVI, solely or combined with a lipid-rich plaque phenotype, is associated with enhanced plaque progression at 1-year follow-up. Where plaque progression occurred, low TAWSS was also observed. These findings suggest TSVI, in addition to low TAWSS, as a potential biomechanical predictor for plaque progression, showing promise for clinical translation to improve patient prognosis.