Computational characterization of behavioral response of medaka (Oryzias latipes) treated with diazinon.

The behavior of indicator specimens in response to sub-lethal doses of toxic substances has been used to detect contamination in aquatic ecosystems. Changes in the movement behaviors of medaka (Oryzias latipes) were analyzed after being treated with diazinon at a concentration of 0.1 mg/l. The movement tracks of medaka were continuously recorded in two-dimension by a digital image processing system both before and after the treatments. Subsequently, two computational methods--two-dimensional fast Fourier transform (2D FFT) and self-organizing map (SOM), were implemented to extract information from the movement data. The differences in the shapes of the movement tracks before and after the treatments were clearly manifested through 2D FFT. The short-distance, irregular turnings in the movement tracks observed after the treatments in the time domain were characteristically transformed to circular or ellipsoidal patterns in the frequency domain. The amplitudes of 2D FFT were efficiently classified by SOM, demonstrating the effects of the different treatments. To evaluate the feasibility of information extraction by 2D FFT, SOM was similarly carried out on the parameters (speed, meander, stop duration, etc.) conventionally used for characterizing the movement tracks. 2D FFT was more efficient in information extraction from the movement data than the parameters. The 2D FFT and SOM were useful as computational methods for automatically detecting response behaviors of indicator specimens exposed to toxic chemicals in aquatic ecosystems.

Pattern recognition of the movement tracks of medaka (Oryzias latipes) in response to sub-lethal treatments of an insecticide by using artificial neural networks.

Specimens of medaka (Oryzias latipes) were observed continuously through an automatic image recognition system before and after treatments of an anti-cholinesterase insecticide, diazinon (0.1 mg/l), for 4 days in semi-natural conditions (2 days before treatment and 2 days after treatment). The "smooth" pattern was typically shown as a normal movement behavior, while the "shaking" pattern was frequently observed after treatments of diazinon. These smooth and shaking patterns were selected for training with an artificial neural network. Parameters characterizing the movement tracks, such as speed, degree of backward movements, stop duration, turning rate, meander, and maximum distance movements in the y-axis of 1-min duration, were given as input (six nodes) to a multi-layer perceptron with the back propagation algorithm. Binary information for the smooth and shaking patterns was separately given as the matching output (one node), while eight nodes were assigned to a single hidden layer. As new input data were given to the trained network, it was possible to recognize the smooth and shaking patterns of the new input data. Average recognition rates of the smooth pattern decreased significantly while those for the shaking pattern increased to a higher degree after treatments of diazinon. The trained network was able to reveal the difference in the shaking pattern in different light phases before treatments of diazinon. This study demonstrated that artificial neural networks could be useful for detecting the presence of toxic chemicals in the environment by serving as in-situ behavioral monitoring tools.