• Design of a smart system for rapid bacterial test

      Patil, Rajshree; Levin, Saurabh; Rajkumar, Samuel; Ajmal, Tahmina; Institute of Chemical Technology (ICT), Mumbai; Foundation for Environmental Monitoring, Bangalore; University of Bedfordshire (MDPI, 2019-12-19)
      In this article, we present our initial findings to support the design of an advanced field test to detect bacterial contamination in water samples. The system combines the use of image processing and neural networks to detect an early presence of bacterial activity. We present here a proof of concept with some tests results. Our initial findings are very promising and indicate detection of viable bacterial cells within a period of 2 h. To the authors' knowledge this is the first attempt to quantify viable bacterial cells in a water sample using cell splitting. We also present a detailed design of the complete system that uses the time lapse images from a microscope to complete the design of a neural network based smart system.
    • Developing a novel water quality prediction model for a South African aquaculture farm

      Eze, Elias Chinedum; Halse, Sarah; Ajmal, Tahmina; University of Bedfordshire; Abagold Limited (MDPI, 2021-06-28)
      Providing an accurate prediction of water quality parameters for improved water quality management is a topical issue in the aquaculture industry. Conventional prediction methods have shown different challenges like a poor generalization, poor prediction accuracy, and high time complexity. Aiming at these challenges, a novel hybrid prediction model with ensemble empirical mode decomposition (EEMD) and deep learning (DL) long-short term memory (LSTM) neural network is proposed in this paper. In this innovative hybrid EEMD-DL-LSTM model, firstly, the integrity of the datasets is enhanced by applying moving average filtering and linear interpolation techniques of water quality parameter datasets pre-treatment. Secondly, the measured real sensor water quality parameters dataset is decomposed with the aid of the EEMD algorithm into disparate IMFs and a corresponding residual item. Thirdly, a multi-feature selection process is applied to make a careful selection of a strongly correlated group of IMFs with the measured real water quality parameter datasets and integrate them as inputs to the DL-LSTM neural network. The presented model is built on water quality sensor data collected from an Abalone farm in South Africa. The performance of the novel hybrid prediction model is validated by comparing the results against the real datasets. To measure the overall accuracy of the novel hybrid prediction model, different statistical indices, namely the Mean Absolute Error (MAE), Mean Square Error (MSE), Root Mean Square Error (RMSE), and Mean Absolute Percentage Error (MAPE), are used.