MRC Laboratory of Molecular Biology
CUED - Cambridge University Engineering DepartmentNot a member yet
45551 research outputs found
Sort by
An automated machine-learning approach for road pothole detection using smartphone sensor data
No full textRoad surface monitoring and maintenance are essential for driving comfort, transport safety and preserving infrastructure integrity. Traditional road condition monitoring is regularly conducted by specially designed instrumented vehicles, which requires time and money and is only able to cover a limited proportion of the road network. In light of the ubiquitous use of smartphones, this paper proposes an automatic pothole detection system utilizing the built-in vibration sensors and global positioning system receivers in smartphones. We collected road condition data in a city using dedicated vehicles and smartphones with a purpose-built mobile application designed for this study. A series of processing methods were applied to the collected data, and features from different frequency domains were extracted, along with various machine-learning classifiers. The results indicated that features from the time and frequency domains outperformed other features for identifying potholes. Among the classifiers tested, the Random Forest method exhibited the best classification performance for potholes, with a precision of 88.5% and recall of 75%. Finally, we validated the proposed method using datasets generated from different road types and examined its universality and robustness
Decentralized Stability Conditions for Inverter-Based Microgrids
No full textWe consider the problem of stability analysis of an inverter-based microgrid where higher order models are used for the inverter and line dynamics. Decentralized conditions are derived through which stability of the network can be deduced, with these formulated as input/output conditions on locally defined subsystems. The conditions derived allow to exploit the natural passivity properties of lines when these are represented in a common reference frame, but reduce the conservatism by additionally taking into account the coupling with neighbouring buses. Examples are given to demonstrate the results presented
Bend- and Twist-Insensitive Flexible Multimode Polymer Optical Interconnects
No full textPolymer multimode optical waveguides can enable high-speed short-reach optical interconnection at low cost within high performance electronic systems. The formation of such waveguides on flexible substrates can offer important additional advantages such as light weight, ability to be tightly bent, and reconfigurability which are particularly important in environments where space, weight, and shape conformity are critical, for instance in vehicles and aircraft. The ability of such flexible optical interconnects to be tightly bent and twisted with low excess loss is crucial in enabling their use in systems with limited space and with movable parts. As a result, in this work, we present a new design of such flexible polymer multimode waveguides that achieves improved bending loss performance over the conventional waveguide design. It is experimentally shown that the proposed design achieves a very low excess loss of 0.5 dB for a 3 mm radius bend under a 50 µm MMF launch. In comparison, flexible waveguides with the conventional design exhibit a 2 dB excess loss under the same launch and bend conditions. Additionally, useful rules that associate the twisting loss performance of flexible polymer waveguide samples with their geometric characteristics are derived. It is shown that negligible twisting losses (<0.1 dB for a 50 µm MMF input) can be achieved when the dimensions of the waveguide samples are appropriately selected. The results demonstrate the strong potential of such bend- and twist-insensitive flexible polymer waveguides for use in next-generation vehicles and aircraft
Double-Framed Thin Elastomer Devices
No full textElastomers and, in particular, polydimethylsiloxane (PDMS) are widely adopted as biocompatible mechanically compliant substrates for soft and flexible micro-nanosystems in medicine, biology, and engineering. However, several applications require such low thicknesses (e.g., <100 μm) that make peeling-off critical because very thin elastomers become delicate and tend to exhibit strong adhesion with carriers. Moreover, microfabrication techniques such as photolithography use solvents which swell PDMS, introducing complexity and possible contamination, thus limiting industrial scalability and preventing many biomedical applications. Here, we combine low-adhesion and rectangular carrier substrates, adhesive Kapton frames, micromilling-defined shadow masks, and adhesive-neutralizing paper frames for enabling fast, easy, green, contaminant-free, and scalable manufacturing of thin elastomer devices, with both simplified peeling and handling. The accurate alignment between the frame and shadow masks can be further facilitated by micromilled marking lines on the back side of the low-adhesion carrier. As a proof of concept, we show epidermal sensors on a 50 μm-thick PDMS substrate for measuring strain, the skin bioimpedance and the heart rate. The proposed approach paves the way to a straightforward, green, and scalable fabrication of contaminant-free thin devices on elastomers for a wide variety of applications
Indium silicon oxide TFT fully photolithographically processed for circuit integration
No full textA new class of amorphous oxide semiconductors based on InOx doped with Ti, W or Si seems to show great promise for large area, flexible, electronics. Of particular interest is the In2O3:SiO2 system as it has a relatively large bond dissociation energy, hence highly suited for long-term environmental stability. In this paper, we present a sub-200°C fully photolithographically-processed indium oxide thin film transistor that is fully compatible to circuit integration on plastic substrates. The TFTs typically showed a mobility of 5 cm2/Vs, a threshold voltage of -0.16 V and a subthreshold swing of 312 mV/dec. We report on its stability behavior when subject to electrical bias stress and negative bias illumination stress, along with a pulse-based compensation solution for persistent photoconductivity arising from the latter. Following static characterization and subsequent parameter extraction of the indium silicon oxide TFT, design considerations are presented along with measurement results of a fully integrated TFT voltage amplifier with high impedance subthreshold loading
Accuracy of Commodity Finger Tracking Systems for Virtual Reality Head-Mounted Displays
No full textRepresenting users’ hands and fingers in virtual reality is crucial for many tasks. Recently, virtual reality head-mounted displays, capable of camera-based inside-out tracking and finger and hand tracking, are becoming popular and complement add-on solutions, such as Leap Motion.However, interacting with physical objects requires an accurate grounded positioning of the virtual reality coordinate system relative to relevant objects, and a good spatial positioning of the user’s fingers and hands.In order to get a better understanding of the capabilities of Virtual Reality headset finger tracking solutions for interacting with physical objects, we ran a controlled experiment (n =24) comparing two commodity hand and finger tracking systems (HTC Vive and Leap Motion) and report on the accuracy of commodity hand tracking systems
An AR-Based Inspection System for Monitoring Temperature Abnormalities in Daily O and M Management
No full textAlthough facility management (FM) managers can control the daily operations and maintenance (O&M) events via visual senses, it is still challenging for them to inspect as-is conditions efficiently (especially unobserved pumping and pipes) and identify all possible abnormalities based on their experiences. Previous research has been conducted to facilitate O&M inspection via the FM management systems (e.g., computerized maintenance management systems, CMMS) and distributed sensor systems. However, there is still a lack of a visualized intelligent system that can help to inspect, record, communicate, and verify O&M issues in tandem for continuous improvement. In order to provide an intelligent and visualized inspection environment, this study developed an augment reality (AR)-based inspection system based on a digital twin (DT) and focused on the inspection of temperature abnormalities in daily O&M management. Firstly, intelligent abnormalities algorithms are implemented to detect temperature abnormalities. Next, a comprehensive classification and its corresponding sub-categories of temperature abnormalities relating to building assets are constructed based on fault tree analysis (FTA), which encompasses diverse events to distinguish different kinds of maintenance issues commonly appearing in daily O&M management. Expert interviews are conducted to verify and modify the FTA. Next, based on the developed FTA, a rule-based matching module is developed and refined to assist in the matching with the corresponding assets in the existed building DT. Thus, an AR-based system is developed and used to highlight the target assets on site, especially for unobserved assets and a demonstrator of this developed system is developed based on the Centre for Digital Built Britain (CDBB) West Cambridge digital twin pilot. Finally, the challenges involved in developing inspection system in practice, and future opportunities using dynamic DTs for O&M purposes are discussed. The results fill in the research gaps for asset management practitioners, policy makers, and researchers to improve asset performance in O&M phases
Development and Validation of a Prognostic Risk Score System for COVID-19 Inpatients: A Multi-Center Retrospective Study in China.
No full textCoronavirus disease 2019 (COVID-19) has become a worldwide pandemic. Hospitalized patients of COVID-19 suffer from a high mortality rate, motivating the development of convenient and practical methods that allow clinicians to promptly identify high-risk patients. Here, we have developed a risk score using clinical data from 1479 inpatients admitted to Tongji Hospital, Wuhan, China (development cohort) and externally validated with data from two other centers: 141 inpatients from Jinyintan Hospital, Wuhan, China (validation cohort 1) and 432 inpatients from The Third People's Hospital of Shenzhen, Shenzhen, China (validation cohort 2). The risk score is based on three biomarkers that are readily available in routine blood samples and can easily be translated into a probability of death. The risk score can predict the mortality of individual patients more than 12 d in advance with more than 90% accuracy across all cohorts. Moreover, the Kaplan-Meier score shows that patients can be clearly differentiated upon admission as low, intermediate, or high risk, with an area under the curve (AUC) score of 0.9551. In summary, a simple risk score has been validated to predict death in patients infected with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2); it has also been validated in independent cohorts