Multi-Target Deep Learning for Algal Detection and Classification.

Water quality has a direct impact on industry, agriculture, and public health. Algae species are common indicators of water quality. It is because algal communities are sensitive to changes in their habitats, giving valuable knowledge on variations in water quality. However, water quality analysis requires professional inspection of algal detection and classification under microscopes, which is very time-consuming and tedious. In this paper, we propose a novel multi-target deep learning framework for algal detection and classification. Extensive experiments were carried out on a large-scale colored microscopic algal dataset. Experimental results demonstrate that the proposed method leads to the promising performance on algal detection, class identification and genus identification.

MOVES-Matrix for high-performance on-road energy and running emission rate modeling applications.

The MOVES model was developed by the U.S. Environmental Protection Agency (U.S. EPA) to estimate emissions from on-road mobile sources and nonroad sources in the United States. Coupling high-resolution on-road vehicle activity data with appropriate MOVES emission rates further advances research efforts designed to assess the environmental impacts of transportation design and operation strategies. However, the complicated MOVES interface and slow performance makes it difficult to assess large, regional scale transportation networks and to undertake analyses of large-scale systems that are dynamic in nature. The MOVES-Matrix system develops an initial Large Matrix of MOVES outputs by running MOVES 146,853 times on the PACE high performance computing cluster to generate more than 90 billion emission rates to populate the matrix for a single area with one fuel regime and one inspection and maintenance program. A total of 117 such Large Matrices would be needed for the entire United States. The MOVES-Matrix system developed can be used to conduct the emissions modeling 200-times faster than using MOVES. The hypothetical case study shows that MOVES-Matrix is able to generate the exact same emission results as the MOVES model to ensure the validity for regulatory analysis. The resulting matrix allows users to link emission rates to big data projects and to evaluate changes in emissions for dynamic transportation systems in near-real-time. MOVES-Matrix does not currently estimate emissions from starts, hoteling or evaporative emissions, and the research team is working on MOVES-Matrix version 2 that supports incorporating off-network modeling.Implications: MOVES-Matrix should be of interest to a broad readership including those interested in vehicle emission modeling, near-road air quality modeling, transportation conformity analysis. The paper should also interest engineers who are involved in transportation regulatory and conformity analysis, state implementation plan, and who are seeking an efficient way of conducting regulatory emission modeling and air quality analysis in the United States.

Understanding the emission impacts of high-occupancy vehicle (HOV) to high-occupancy toll (HOT) lane conversions: Experience from Atlanta, Georgia.

Converting a congested high-occupancy vehicle (HOV) lane into a high-occupancy toll (HOT) lane is a viable option for improving travel time reliability for carpools and buses that use the managed lane. However, the emission impacts of HOV-to-HOT conversions are not well understood. The lack of emission impact quantification for HOT conversions creates a policy challenge for agencies making transportation funding choices. The goal of this paper is to evaluate the case study of before-and-after changes in vehicle emissions for the Atlanta, Georgia, I-85 HOV/HOT lane conversion project, implemented in October 2011. The analyses employed the Motor Vehicle Emission Simulator (MOVES) for project-level analysis with monitored changes in vehicle activity data collected by Georgia Tech researchers for the Georgia Department of Transportation (GDOT). During the quarterly field data collection from 2010 to 2012, more than 1.5 million license plates were observed and matched to vehicle class and age information using the vehicle registration database. The study also utilized the 20-sec, lane-specific traffic operations data from the Georgia NaviGAtor intelligent transportation system, as well as a direct feed of HOT lane usage data from the State Road and Tollway Authority (SRTA) managed lane system. As such, the analyses in this paper simultaneously assessed the impacts associated with changes in traffic volumes, on-road operating conditions, and fleet composition before and after the conversion. Both greenhouse gases and criteria pollutants were examined. IMPLICATIONS: A straight before-after analysis showed about 5% decrease in air pollutants and carbon dioxide (CO2). However, when the before-after calendar year of analysis was held constant (to account for the effect of 1 yr of fleet turnover), mass emissions at the analysis site during peak hours increased by as much as 17%, with little change in CO2. Further investigation revealed that a large percentage decrease in criteria pollutants in the straight before-after analysis was associated with a single calendar year change in MOVES. Hence, the Atlanta, Georgia, results suggest that an HOV-to-HOT conversion project may have increased mass emissions on the corridor. The results also showcase the importance of obtaining on-road data for emission impact assessment of HOV-to-HOT conversion projects.

MOVES-Matrix and distributed computing for microscale line source dispersion analysis.

MOVES and AERMOD are the U.S. Environmental Protection Agency's recommended models for use in project-level transportation conformity and hot-spot analysis. However, the structure and algorithms involved in running MOVES make analyses cumbersome and time-consuming. Likewise, the modeling setup process, including extensive data requirements and required input formats, in AERMOD lead to a high potential for analysis error in dispersion modeling. This study presents a distributed computing method for line source dispersion modeling that integrates MOVES-Matrix, a high-performance emission modeling tool, with the microscale dispersion models CALINE4 and AERMOD. MOVES-Matrix was prepared by iteratively running MOVES across all possible iterations of vehicle source-type, fuel, operating conditions, and environmental parameters to create a huge multi-dimensional emission rate lookup matrix. AERMOD and CALINE4 are connected with MOVES-Matrix in a distributed computing cluster using a series of Python scripts. This streamlined system built on MOVES-Matrix generates exactly the same emission rates and concentration results as using MOVES with AERMOD and CALINE4, but the approach is more than 200 times faster than using the MOVES graphical user interface. Because AERMOD requires detailed meteorological input, which is difficult to obtain, this study also recommends using CALINE4 as a screening tool for identifying the potential area that may exceed air quality standards before using AERMOD (and identifying areas that are exceedingly unlikely to exceed air quality standards). CALINE4 worst case method yields consistently higher concentration results than AERMOD for all comparisons in this paper, as expected given the nature of the meteorological data employed. IMPLICATIONS: The paper demonstrates a distributed computing method for line source dispersion modeling that integrates MOVES-Matrix with the CALINE4 and AERMOD. This streamlined system generates exactly the same emission rates and concentration results as traditional way to use MOVES with AERMOD and CALINE4, which are regulatory models approved by the U.S. EPA for conformity analysis, but the approach is more than 200 times faster than implementing the MOVES model. We highlighted the potentially significant benefit of using CALINE4 as screening tool for identifying potential area that may exceeds air quality standards before using AERMOD, which requires much more meteorology input than CALINE4.

Manipulating liquid plugs in microchannel with controllable air vents.

An air venting element on microchannel, which can be controlled externally and automatically, was demonstrated for manipulating liquid plugs in microfluidic systems. The element's open and closed statuses correspond to the positioning and movement of a liquid plug in the microchannel. Positioning of multiple liquid plugs at an air venting element enabled the merging and mixing of the plugs. Besides these basic functions, other modes of liquid plug manipulations including plug partitioning, multiple plug mixing, and spacing adjustment between liquid plugs, were realized using combination of multiple elements. The structure, operation, and some functions of the element were demonstrated with a microfluidic chip application. The performances of the element including its failure modes, threshold flow rate, and structural optimization were also discussed.

A high-flux isopore micro-fabricated membrane for effective concentration and recovering of waterborne pathogens.

A high-flux metallic micro/nano-filtration membrane has been fabricated and validated for isolation of waterborne pathogens from drinking water. Obtained membrane with smooth surface and perfectly ordered pores was achieved by a high yield and cost effective multilevel lithography and electroplating technique. The micro-fabricated membrane was also strengthened with an integrated back-support, which can withstand a high pressure during filtration. The results of microfiltration tests with model particles revealed the superior performance of the micro-fabricated filter than current commercial filters in sample throughput, recovery ratio, and reusability. This study highlighted the potential application of micro-fabricated filer in rapid filtration and recovery of C. parvum oocysts for downstream analysis.

Real-time PCR array chip with capillary-driven sample loading and reactor sealing for point-of-care applications.

A major challenge for the lab-on-a-chip (LOC) community is to develop point-of-care diagnostic chips that do not use instruments. Such instruments include pumping or liquid handling devices for distribution of patient's nucleic-acid test sample among an array of reactors and microvalves or mechanical parts to seal these reactors. In this paper, we report the development of a primer pair pre-loaded PCR array chip, in which the loading of the PCR mixture into an array of reactors and subsequent sealing of the reactors were realized by a novel capillary-based microfluidics with a manual two-step pipetting operations. The chip is capable of performing simultaneous (parallel) analyses of multiple gene targets and its performance was tested by amplifying twelve different gene targets against cDNA template from human hepatocellular carcinoma using SYBR Green I fluorescent dye. The versatility and reproducibility of the PCR-array chip are demonstrated by real-time PCR amplification of the BNI-1 fragment of SARS cDNA cloned in a plasmid vector. The reactor-to-reactor diffusion of the pre-loaded primer pairs in the chip is investigated to eliminate the possibility of primer cross-contamination. Key technical issues such as PCR mixture loss in gas-permeable PDMS chip layer and bubble generation due to different PDMS-glass bonding methods are investigated.