Artificial Neural Networks (ANNs) and Response Surface Methodology (RSM) Approach for Modelling the Optimization of Chromium (VI) Reduction by Newly Isolated Acinetobacter radioresistens Strain NS-MIE from Agricultural Soil.

Numerous technologies and approaches have been used in the past few decades to remove hexavalent chromium (Cr[VI]) in wastewater and the environment. However, these conventional technologies are not economical and efficient in removing Cr(VI) at a very low concentration (1-100 ppm). As an alternative, the utilization of bioremediation techniques which uses the potential of microorganisms could represent an effective technique for the detoxification of Cr(VI). In this study, we reported a newly isolated bacterium identified as Acinetobacter radioresistens sp. NS-MIE from Malaysian agricultural soil. The chromate reduction potential of strain NS-MIE was optimized using RSM and ANN techniques. The optimum condition predicted by RSM for the bacterium to reduce hexavalent chromium occurred at pH 6, 10 g/L ppm of nutrient broth (NB) concentration and 100 ppm of chromate concentration while the optimum condition predicted by ANN is at pH 6 and 10 g/L of NB concentration and of 60 ppm of chromate concentration with chromate reduction (%) of 75.13 % and 96.27 %, respectively. The analysis by the ANN model shows better prediction data with a higher R2 value of 0.9991 and smaller average absolute deviation (AAD) and root mean square error (RMSE) of 0.33 % and 0.302 %, respectively. Validation analysis showed the predicted values by RSM and ANN were close to the validation values, whereas the ANN showed the lowest deviation, 2.57%, compared to the RSM. This finding suggests that the ANN showed a better prediction and fitting ability compared to the RSM for the nonlinear regression analysis. Based on this study, A. radioresistens sp. NS-MIE exhibits strong potential characteristics as a candidate for the bioremediation of hexavalent chromium in the environment.