Stevens Institute of Technology

About: Stevens Institute of Technology is a education organization based out in Hoboken, New Jersey, United States. It is known for research contribution in the topics: Cognitive radio & Wireless network. The organization has 5440 authors who have published 12684 publications receiving 296875 citations. The organization is also known as: Stevens & Stevens Tech.

One-Pass Multi-Task Networks With Cross-Task Guided Attention for Brain Tumor Segmentation

Abstract: Class imbalance has emerged as one of the major challenges for medical image segmentation. The model cascade (MC) strategy, a popular scheme, significantly alleviates the class imbalance issue via running a set of individual deep models for coarse-to-fine segmentation. Despite its outstanding performance, however, this method leads to undesired system complexity and also ignores the correlation among the models. To handle these flaws in the MC approach, we propose in this paper a light-weight deep model, i.e., the One-pass Multi-task Network (OM-Net) to solve class imbalance better than MC does, while requiring only one-pass computation for brain tumor segmentation. First, OM-Net integrates the separate segmentation tasks into one deep model, which consists of shared parameters to learn joint features, as well as task-specific parameters to learn discriminative features. Second, to more effectively optimize OM-Net, we take advantage of the correlation among tasks to design both an online training data transfer strategy and a curriculum learning-based training strategy. Third, we further propose sharing prediction results between tasks, which enables us to design a cross-task guided attention (CGA) module. By following the guidance of the prediction results provided by the previous task, CGA can adaptively recalibrate channel-wise feature responses based on the category-specific statistics. Finally, a simple yet effective post-processing method is introduced to refine the segmentation results of the proposed attention network. Extensive experiments are conducted to demonstrate the effectiveness of the proposed techniques. Most impressively, we achieve state-of-the-art performance on the BraTS 2015 testing set and BraTS 2017 online validation set. Using these proposed approaches, we also won joint third place in the BraTS 2018 challenge among 64 participating teams. The code is publicly available at https://github.com/chenhong-zhou/OM-Net .

Acoustic emission detection using fiber Bragg gratings

Abstract: A systematic study of acoustic emission detection using fiber Bragg grating sensors has been carried out over the last year. In this, we attempt to use the fiber Bragg grating to sense the dynamic strain created by a passing ultrasonic wave signal. Our goal here is to see if such a sensor is possible, and if so, what the detection sensitivity and limitations will be. To answer these questions, we carried out several experiments involving the detection of simulated acoustic emission events. In the first experiment, we attach fiber Bragg grating to the surface of a piezoceramic resonator which is driven by a signal generator. We were able to detect the resulting surface vibration of the resonator up to 2.1 MHz. In the second experiment, we attach a fiber Bragg grating to the surface of an aluminum plate. We excite an acoustic wave using an ultrasonic transducer located at various positions of the aluminum plate. In this way, we demonstrated that the fiber Bragg Grating sensor is capable of picking up the signal coming from a distance (up to 30 cm) for up to 2.5 MHz. In a third experiment, we use the same fiber Bragg grating on aluminum plate set up, but set up an acoustic signal by either a gentle knock on the plate by a pin, or by breaking a pencil lead on the plate. We were able to detection acoustic emission set up by pencil lead breaking up to a frequency of 30 kHz. Higher frequency components were not detected mainly due to the limitation of available electronic equipment at this time (higher frequency band-pass filters and amplifiers. In all the above-mentioned experiments we use a match Bragg grating to demodulate the detected optical signal and use a dual channel scheme for electronic data acquisition and processing (a signal channel and a reference channel).

LVI-SAM: Tightly-coupled Lidar-Visual-Inertial Odometry via Smoothing and Mapping

Abstract: We propose a framework for tightly-coupled lidar-visual-inertial odometry via smoothing and mapping, LVI-SAM, that achieves real-time state estimation and map-building with high accuracy and robustness. LVI-SAM is built atop a factor graph and is composed of two sub-systems: a visual-inertial system (VIS) and a lidar-inertial system (LIS). The two sub-systems are designed in a tightly-coupled manner, in which the VIS leverages LIS estimation to facilitate initialization. The accuracy of the VIS is improved by extracting depth information for visual features using lidar measurements. In turn, the LIS utilizes VIS estimation for initial guesses to support scan-matching. Loop closures are first identified by the VIS and further refined by the LIS. LVI-SAM can also function when one of the two sub-systems fails, which increases its robustness in both texture-less and feature-less environments. LVI-SAM is extensively evaluated on datasets gathered from several platforms over a variety of scales and environments. Our implementation is available at https://git.io/lvi-sam.

Biphasic osteochondral scaffold for reconstruction of articular cartilage

Abstract: An osteochondral scaffold has a chondrogenic spiral scaffold in one end of an outer shell made of sintered microspheres, and an osteogenic spiral scaffold in the other end of the outer shell. Each spiral scaffold has nanofibers of a composition selected to promote attachment and proliferation of the desired types of cells. The nanofibers for the chondrogenic spiral scaffold have a different composition than the nanofibers for the osteogenic spiral scaffold. The nanofibers of each spiral scaffold are aligned to orient the attached cells so as to recreate the structure of the native tissue.

Cooperative Spectrum Sensing in Cognitive Radio Networks in the Presence of the Primary User Emulation Attack

Abstract: In recent years, the security issues of the cognitive radio (CR) networks have drawn a lot of research attentions. Primary user emulation attack (PUEA), as one of common attacks, compromises the spectrum sensing, where a malicious user forestalls vacant channels by impersonating the primary user to prevent other secondary users from accessing the idle frequency bands. In this paper, we propose a new cooperative spectrum sensing scheme, considering the existence of PUEA in CR networks. In the proposed scheme, the sensing information of different secondary users is combined at a fusion center and the combining weights are optimized with the objective of maximizing the detection probability of available channels under the constraint of a required false alarm probability. We also investigate the impact of the channel estimation errors on the detection probability. Simulation and numerical results illustrate the effectiveness of the proposed scheme in cooperative spectrum sensing in the presence of PUEA.