Abstract. What is the right supervisory signal to train visual representations? Current approaches in computer vision use category labels from datasets such as ImageNet to train ConvNets. However, in case of biological agents, visual representation learning does not require millions of semantic labels. We argue that biological agents use physical interactions with the world to learn visual representations unlike current vision systems which just use passive observations (images and videos downloaded from web). For example, babies push objects, poke them, put them in their mouth and throw them to learn representations. Towards this goal, we build one of the first systems on a Baxter platform that pushes, pokes, grasps and observes objects in a tabletop environment. It uses four different types of physical interactions to collect more than 130K datapoints, with each datapoint providing supervision to a shared ConvNet architecture allowing us to learn visual representations. We show the quality of learned representations by observing neuron activations and performing nearest neighbor retrieval on this learned representation. Quantitatively, we evaluate our learned ConvNet on image classification tasks and show improvements compared to learning without external data. Finally, on the task of instance retrieval, our network outperforms the ImageNet network on recall@1 by 3%.
We present the design of a soft wearable robotic device composed of elastomeric artificial muscle actuators and soft fabric sleeves, for active assistance of knee motions. A key feature of the device is the two-dimensional design of the elastomer muscles that not only allows the compactness of the device, but also significantly simplifies the manufacturing process. In addition, the fabric sleeves make the device lightweight and easily wearable. The elastomer muscles were characterized and demonstrated an initial contraction force of 38N and maximum contraction of 18mm with 104kPa input pressure, approximately. Four elastomer muscles were employed for assisted knee extension and flexion. The robotic device was tested on a 3D printed leg model with an articulated knee joint. Experiments were conducted to examine the relation between systematic change in air pressure and knee extensionflexion. The results showed maximum extension and flexion angles of 95 • and 37 • , respectively. However, these angles are highly dependent on underlying leg mechanics and positions. The device was also able to generate maximum extension and flexion forces of 3.5N and 7N, respectively.
Recently, there has been active research in finding robotized solutions for the treatment of atrial fibrillation (AF) by augmenting catheter systems through the integration of force sensors at the tip. However, limited research has been aimed at providing automatic force control by also integrating actuation of the catheter tip, which can significantly enhance safety in such procedures. This article solves the demanding challenge of miniaturizing both actuation and sensing for integration into flexible catheters. Fabrication strategies are presented for a series of novel soft thick-walled cylindrical actuators, with embedded sensing using eutectic gallium-indium. The functional catheter tips have a diameter in the range of 2.6-3.6 mm and can both generate and detect forces in the range of < 0.4 N, with a bandwidth of 1-2 Hz. The deformation modeling of thick-walled cylinders with fiber reinforcement is presented in the article. An experimental setup developed for static and dynamic characterization of these units is presented. The prototyped units were validated with respect to the design specifications. The preliminary force control results indicate that these units can be used in tracking and control of contact force, which has the potential to make AF procedures much safer and more accurate.
PurposeThere are few reports from Asian countries about the long-term results of aromatase inhibitor adjuvant treatment for breast cancer. This observational study aimed to evaluate the long-term effects of letrozole in postmenopausal Korean women with operable breast cancer.MethodsSelf-reported quality of life (QoL) scores were serially assessed for 3 years during adjuvant letrozole treatment using the Korean version of the Functional Assessment of Cancer Therapy-Breast questionnaires (version 3). Changes in bone mineral density (BMD) and serum cholesterol levels were also examined.ResultsAll 897 patients received the documented informed consent form and completed a baseline questionnaire before treatment. Adjuvant chemotherapy was administered to 684 (76.3%) subjects, and 410 (45.7%) and 396 (44.1%) patients had stage I and II breast cancer, respectively. Each patient completed questionnaires at 3, 6, 12, 18, 24, 30, and 36 months after enrollment. Of 897 patients, 749 (83.5%) completed the study. The dropout rate was 16.5%. The serial trial outcome index, the sum of the physical and functional well-being subscales, increased gradually and significantly from baseline during letrozole treatment (p<0.001). The mean serum cholesterol level increased significantly from 199 to 205 after 36 months (p=0.042). The mean BMD significantly decreased from −0.39 at baseline to −0.87 after 36 months (p<0.001).ConclusionQoL gradually improved during letrozole treatment. BMD and serum cholesterol level changes were similar to those in Western countries, indicating that adjuvant letrozole treatment is well tolerated in Korean women, with minimal ethnic variation.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.