Pattern recognition in flow cytometry.

BACKGROUND: Analytical flow cytometry (AFC), by quantifying sometimes more than 10 optical parameters on cells at rates of approximately 10(3) cells/s, rapidly generates vast quantities of multidimensional data, which provides a considerable challenge for data analysis. We review the application of multivariate data analysis and pattern recognition techniques to flow cytometry. METHODS: Approaches were divided into two broad types depending on whether the aim was identification or clustering. Multivariate statistical approaches, supervised artificial neural networks (ANNs), problems of overlapping character distributions, unbounded data sets, missing parameters, scaling up, and estimating proportions of different types of cells comprised the first category. Classic clustering methods, fuzzy clustering, and unsupervised ANNs comprised the second category. We demonstrate the state of the art by using AFC data on marine phytoplankton populations. RESULTS AND CONCLUSIONS: Information held within the large quantities of data generated by AFC was tractable using ANNs, but for field studies the problem of obtaining suitable training data needs to be resolved, and coping with an almost infinite number of cell categories needs further research.

Comparison of five clustering algorithms to classify phytoplankton from flow cytometry data.

BACKGROUND: Artificial neural networks (ANNs) have been shown to be valuable in the analysis of analytical flow cytometric (AFC) data in aquatic ecology. Automated extraction of clusters is an important first stage in deriving ANN training data from field samples, but AFC data pose a number of challenges for many types of clustering algorithm. The fuzzy k-means algorithm recently has been extended to address nonspherical clusters with the use of scatter matrices. Four variants were proposed, each optimizing a different measure of clustering "goodness." METHODS: With AFC data obtained from marine phytoplankton species in culture, the four fuzzy k-means algorithm variants were compared with each other and with another multivariate clustering algorithm based on critical distances currently used in flow cytometry. RESULTS: One of the algorithm variants (adaptive distances, also known as the Gustafson--Kessel algorithm) was found to be robust and reliable, whereas the others showed various problems. CONCLUSIONS: The adaptive distances algorithm was superior in use to the clustering algorithms against which it was tested, but the problem of automatic determination of the number of clusters remains to be addressed.

A proposed taxonomy for nailfold capillaries based on their morphology.

Certain diseases cause permanent changes to the shapes and densities of nailfold capillaries and, therefore, nailfold capillaroscopy is important as a tool for diagnosing and monitoring these diseases. The first aim of the project is to resolve differences in terminology that have developed over the years in previous work. We propose a taxonomy for nailfold capillaries that cover six descriptive classes: cuticulis, open, tortuous, crossed, bushy, and bizarre. The first three are parametric in that they may be distinguished by the ratio of capillary length to width and by the curvature of the capillary limbs. The last three are characterized by their topology; a crossed capillary has a closed area that is not connected to the image background. Bushy and bizarre capillaries have atypical shapes that are characterized by the convex hull of their skeleton. These descriptive classes may be modified according to anomalies in width and length. The second aim is to automate the classification of capillaries by encapsulating the taxonomy in an algorithm; our computer program rivals the most experienced clinicians in classifying capillaries consistently with an overall agreement of 85% with the clinicians' majority view. This was particularly valuable in classifying borderline shapes objectively and consistently.

Training radial basis function neural networks: effects of training set size and imbalanced training sets.

Obtaining training data for constructing artificial neural networks (ANNs) to identify microbiological taxa is not always easy. Often, only small data sets with different numbers of observations per taxon are available. Here, the effect of both size of the training data set and of an imbalanced number of training patterns for different taxa is investigated using radial basis function ANNs to identify up to 60 species of marine microalgae. The best networks trained to discriminate 20, 40 and 60 species respectively gave overall percentage correct identification of 92, 84 and 77%. From 100 to 200 patterns per species was sufficient in networks trained to discriminate 20, 40 or 60 species. For 40 and 60 species data sets an imbalance in the number of training patterns per species always affected training success, the greater the imbalance the greater the effect. However, this could be largely compensated for by adjusting the networks using a posteriori probabilities, estimated as network output values.

Proportion estimation with confidence limits.

A common task in microbiology involves determining the composition of a mixed population of individuals by drawing a sample from the population and using some procedure to identify the individuals in the sample. There may be a significant probability that the identification procedure misidentifies some members of the sample (for example, because the available data are insufficient unambiguously to identify an individual) which makes finding the proportions in the underlying population non-trivial. A further complication arises where individuals are present in the population that do not belong to any of the subpopulations recognised by use of the identification procedure. A simple algorithm is presented to address these problems and construct a maximum likelihood estimate of the proportions, together with confidence limits. The technique is illustrated using an example drawn from flow cytometry in which phytoplankton cells are identified from flow cytometry data by an RBF neural network, and the limitations of the approach are discussed.

Identification of phytoplankton from flow cytometry data by using radial basis function neural networks.

We describe here the application of a type of artificial neural network, the Gaussian radial basis function (RBF) network, in the identification of a large number of phytoplankton strains from their 11-dimensional flow cytometric characteristics measured by the European Optical Plankton Analyser instrument. The effect of network parameters on optimization is examined. Optimized RBF networks recognized 34 species of marine and freshwater phytoplankton with 91. 5% success overall. The relative importance of each measured parameter in discriminating these data and the behavior of RBF networks in response to data from "novel" species (species not present in the training data) were analyzed.

A comparison of some neural and non-neural methods for identification of phytoplankton from flow cytometry data.

Four artificial neural network paradigms (multilayer perceptron networks, learning vector quantization networks, and radial and asymmetric basis function networks) and two statistical methods (parametric statistical classification by modelling each class with Gaussian distributions, and non-parametric density estimation via the K-nearest neighbour method) were compared for their ability to identify seven freshwater and five marine phytoplankton species from flow cytometric data. Kohonen self-organizing maps were also used to examine similarities between species. Optimized networks and statistical methods performed similarly, correctly identifying between 86.8% and 90.1% of data from freshwater species, and between 81.3% and 84.1% of data from marine species. Choice of identification technique must therefore be made on the basis of other criteria. We highlight the way each method partitions the data space and thereby separates the data clusters, and discuss the relative merits of each with reference to complexity of data boundaries, training time, analysis time and behaviour when presented with 'novel' data.

A comparison of Radial Basis Function and backpropagation neural networks for identification of marine phytoplankton from multivariate flow cytometry data.

Two artificial neural network classifiers, the well-known Multi-layer Perception (MLP) (also known as the 'backpropagation network'), and the more recently developed Radial Basis Function (RBF) network, were evaluated and compared for their ability to identify multivariate flow cytometric data from five North Sea plankton groups (Dinoflagellidae, Bacillariophyceae, Prymnesiomonadida, Cryptomonadida, and other flagellates). RBF networks generally performed similarly to MLPs, and slightly better in cases where the data were markedly multimodal; RBF networks also have much shorter training times. The performance of MLPs was improved greatly by the use of a symmetrical bipolar 'transfer function' as opposed to the commonly-used asymmetric form. The issues of network optimisation and computational efficiency in use are discussed.

Neural network analysis of flow cytometric data for 40 marine phytoplankton species.

Flow cytometry data (time of flight, horizontal and vertical forward light scatter, 90 degrees light scatter, and "red" and "orange" integral fluorescence) were collected for laboratory cultures of 40 species of marine phytoplankton, from the following taxonomic classes, the Dinophyceae, Bacillariophyceae, Prymnesiophyceae, Cryptophyceae, and other flagellates. Single-hidden-layer "back-propagation" neural networks were trained to discriminate between species by recognising patterns in their flow cytometric signatures, and network performance was assessed using an independent test data set. Two approaches were adopted employing: (1) a hierarchy of small networks, the first identifying to which major taxonomic group a cell belonged, and then a network for that taxonomic group identified to species, and (2) a single large network. Discriminating some of the major taxonomic groups was successful but others less so. With networks for specific groups, cryptophyte species were all identified reliably (probability of correct classification always being > 0.75); in the other groups half of the species were identified reliably. With the large network, dinoflagellates, cryptomonads, and flagellates were identified almost as well as by networks specific for these groups. The application of neural computing techniques to identification of such a large number of species represents a significant advance from earlier studies, although further development is required.

Computed tomography pelvimetry: accuracy and radiation dose compared with conventional pelvimetry.

Several publications in the past few years have indicated that computed tomography (CT) pelvimetry is preferable to conventional pelvimetry when considering accuracy and radiation dose. Many of the previous publications have, however, compared state of the art CT pelvimetry with outmoded conventional pelvimetry techniques. This study compared both the accuracy and the radiation dose of CT pelvimetry with conventional pelvimetry as practised in two large teaching obstetric hospitals in Melbourne. The study did not demonstrate any significant difference in the accuracy between the two methods. The radiation dose to the fetal gonads was also similar. The radiation dose using CT pelvimetry could be significantly lowered if the axial CT section through the ischial spines were excluded from the CT technique.

Retroperitoneal air dissection associated with mechanical ventilation.

The radiologic patterns, etiology, and possible consequences of gas dissection into the retroperitoneal space from pulmonary sources are reviewed in three patients requiring mechanical ventilation. Airway disruption appears to be related to peak airway pressures, underlying pulmonary disease, and the patient's hemodynamic condition and may cause different forms of interstitial emphysema and air embolization, which may lead to severe respiratory or circulatory dysfunction. Gas migration to the retroperitoneum and, secondarily, into the peritoneal cavity should be considered in the differential diagnosis of free intra-abdominal gas.