What is AimsNet ?
AimsNet is a Windows 95/98/NT program for performing pattern recognition of multivariate data, using
the powerful RBF (radial basis function) artificial neural network algorithm.
It also incorporates several other techniques such as fuzzy clustering, principal component analysis,
and canonical variate analysis. Although AimsNet has been specifically
designed with the analysis of flow cytometry derived data in mind, it is quite capable
of analysing other types of data as well.
AimsNet in use (click on image for full-size version)
The integrated multivariate data viewer (click on image for full-size version)
Using AimsNet, the user constructs a connected structure of modules to describe the desired data processing steps.
Each module performs a single data processing task, such as reading data from disc (in FCS or text formats),
rescaling, cluster analysis, principal component and canonical variate analysis,
RBF neural network analysis etc. The results of processing are typically written back to disc, in listmode or
summary form (such as a misidentification matrix). The results may also be visualised
as 2D and 3D scatterplots using the integrated data viewer.
AimsNet has been designed throughout to be easy to use by those with
limited experience of artificial neural networks. Full online help is included, together
with a tutorial. It was developed under the Aims project
(Automated Identification and Characterisation of Microbial Populations -
CEC grant no. MAS3-CT97-0080).