Grantee Research Project Results
2004 Progress Report: Using Neural Networks to Create New Indices and Classification Schemes
EPA Grant Number: R829784Title: Using Neural Networks to Create New Indices and Classification Schemes
Investigators: Brion, Gail M. , Lingireddy, Srinivasa
Institution: University of Kentucky
EPA Project Officer: Page, Angela
Project Period: July 1, 2002 through June 30, 2005 (Extended to June 30, 2006)
Project Period Covered by this Report: July 1, 2004 through June 30, 2005
Project Amount: $523,938
RFA: Microbial Risk in Drinking Water (2001) RFA Text | Recipients Lists
Research Category: Water , Drinking Water , Human Health
Objective:
The objective of this research project is to use advanced computer modeling techniques and newly developed indicators of fecal age to predict enhanced risk, as represented by presence or absence of enteric viruses, in a drinking water source from other more easily measured surrogate parameters.
Progress Summary:
The database was completed for the Kentucky River Study from 108 sampling visits resulting in 100 complete observations of 18 water quality inputs. Multivariate logistic regression has shown that three types of input variables (fecal load, source, and age) are needed to predict enteric virus presence with greater than 80 percent accuracy. Using both forward selection and backward stepwise selection model building programs, the input parameters selected by statistical significance (p < 0.05) from 18 potential input parameters were temperature, the ratio of atypical to typical coliform colonies (AC/TC), and concentrations of fecal coliforms and Enterococci. The AC/TC ratio, an indicator of fecal age and temperature were the most statistically significant input parameters of the four identified by both approaches. The four input multivariate logistic regression model is based on temperature, fecal coliforms, the AC/TC ratio, and Enterococci was compared against an expanded six input parameter model that included concentrations of coprostanol and epicoprostanol along with the aforementioned inputs. The larger model that contained the additional indicator of human specific sterols was found to be superior to the four input model that contained only indicators for fecal load and age. Sensitivity and specificity of 82.5 percent and 81.4 percent were obtained by the six input model. The relative importance of the input variables in the model were log temperature (p = 0.001), log AC/TC (p = 0.004), log fecal coliforms (p = 0.018), log Enterococci (p = 0.033), log coprostanol (p = 0.249), and log epicoprostanol (p = 0.34). A second logistic regression modeling study with input selection based on expert judgment rather than statistical optimization showed that a simple two input parameter model applied to a 102 observation dataset could obtain predictive results after fitting of 83.1 percent sensitivity and 69.8 percent selectivity using only log-transformed values for the AC/TC ratio and concentrations of coprostanol. It was found that only chemical indicators of fecal load and source were needed for prediction of enteric virus. The new bacterial ratio initially proposed for study (AC/TC) has been shown to be vital in the prediction of enteric virus presence for all models, has been conclusively related to fecal age and is the foundation of a new watershed sampling scheme for the indication of hot-spots in local drinking water supply watersheds.
In-house modeling of the completed 100 observation database, along with a database from a smaller local watershed, have resulted in the publication of one paper in Year 3 with initial acceptance of four other papers that have been refined and will be published in 2006. The work has resulted in four national/international conference presentations in Year 3, with an award being received for one of the international presentations. In addition, two national/international conference presentations were submitted and accepted for presentation in the no-cost extension year. Two new journal articles are in preparation for submission from work that will be completed in the no-cost extension year.
Future Activities:
The principal investigators (PIs) will continue comparison studies of artificial neural network analysis against logistic regression and other methods of data analysis on multiparameter databases. The PIs are promoting the use of the new AC/TC bacterial ratio with other researchers and have expanded the fecal load, source, and age input parameter approach to identifying remediation sites in a local recharge area for a karst spring used for water supply.
Journal Articles on this Report : 5 Displayed | Download in RIS Format
Other project views: | All 21 publications | 9 publications in selected types | All 9 journal articles |
---|
Type | Citation | ||
---|---|---|---|
|
Brion GM. The AC/TC bacterial ratio: a tool for watershed quality management. Journal of Water and Environment Technology 2005;3(2):271-277. |
R829784 (2003) R829784 (2004) R829784 (Final) |
Exit Exit |
|
Brion G, Viswanathan C, Neelakantan TR, Lingireddy S, Girones R, Lees D, Allard A, Vantarakis A. Artificial neural network prediction of viruses in shellfish. Applied and Environmental Microbiology 2005;71(9):5244-5253. |
R829784 (2003) R829784 (2004) R829784 (Final) |
Exit Exit |
|
Chandramouli V, Brion G, Neelakantan TR, Lingireddy S. Backfilling missing microbial concentrations in a riverine database using artificial neural networks. Water Research 2007;41(1):217-227. |
R829784 (2003) R829784 (2004) R829784 (Final) |
Exit Exit |
|
Chandramouli V, Lingireddy S, Brion GM. Robust training termination criterion for back-propagation ANNs applicable to small datasets. Journal of Computing in Civil Engineering 2007;21(1):39-46. |
R829784 (2004) R829784 (Final) |
Exit |
|
Freitas SJ, Brion GM, Black L, Coakley T. Predictive input parameters for enteric virus presence at the inlet of a potable water supply. Water Science and Technology 2006;54(3):17-21. |
R829784 (2004) R829784 (Final) |
|
Supplemental Keywords:
water quality, pathogens, indicators, modeling, artificial neural networks, risk management, enteric viruses, environmental microbiology, engineering, remediation,, RFA, Scientific Discipline, Water, Geographic Area, Environmental Chemistry, Ecological Risk Assessment, Ecology and Ecosystems, Drinking Water, Engineering, Chemistry, & Physics, Environmental Engineering, EPA Region, microbial risk assessment, alternative disinfection methods, microbial contamination, environmental monitoring, water contamination detection, region 4, bacteria, microbiological organisms, early warning, microbial pathogens, cryptosporidium , neural networks, emerging pathogens, water quality, ecological risk, drinking water contaminantsProgress and Final Reports:
Original AbstractThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.