Biomedical subjects

M Hagerup

Publications and source records attributed to M Hagerup.

4 recordsLinked to original sources

PubMed1992

Classification of crossed immunoelectrophoretic patterns using digital image processing and artificial neural networks.

A method is presented which makes it possible to present crossed immunoelectrophoretic patterns to an artificial neural network. The electrophoretic patterns are presented for the artificial neural network as three-dimensional vectors and it is shown that it is possible with this representation to train the network to learn the patterns and classify them. It was found that the ability to generalize was substantially increased by the addition of noise to the input patterns during training. Furthermore, the addition of noise decreased the number of presentations needed to reach the predetermined error level. The trained neural network was able to classify all distorted patterns correctly within an error range of 1%.

Image Processing, Computer-Assisted↗

PubMed1989

Concanavalin A crossed affinity immunoelectrophoresis and image analysis for semiquantitative evaluation of microheterogeneity profiles of human serum transferrin from alcoholics and normal individuals.

The microheterogeneity profile of human serum transferrin from normal and alcoholic subjects was investigated qualitatively and quantitatively by means of Concanavalin A crossed affinity immunoelectrophoresis and an image analysis program. Differences in amounts of nonreacting transferrin molecules were found, suggesting an increase in triantennary glycosylation of transferrin from alcoholics compared with normal individuals. The increased amount of a highly retarded fraction in crude sera from alcoholics was demonstrated to be artefactual, probably due to entrapment or coprecipitation as the fraction disappeared after repeating the analysis with immunosorbent-purified transferrin. In conclusion, affinity electrophoresis represents a simple approach for demonstration of variations in the neutral monosaccharides of glycans and can discriminate between transferrin from alcoholics and normal individuals.

Adult↗

PubMed1987

Image processing and pattern recognition algorithms for evaluation of crossed immunoelectrophoretic patterns (crossed radioimmunoelectrophoresis analysis manager; CREAM).

A computerized method for automatic evaluation and comparison of crossed immunoelectrophoretic and crossed radioimmunoelectrophoretic patterns that requires limited hardware resources has been developed. For the initial reading of the plates an ordinary video camera is used. Feature extractors that allow the computer to recognize a point on the precipitation curve as being a peak point have been developed. After this automatic procedure the program allows for an interactive menu-driven proofreading phase during which it is possible to force the system to take into consideration any number of extra points along the precipitation curve in the curve-fitting process. The system has been tested on crossed immunoelectrophoretic patterns as well as crossed radioimmunoelectrophoretic patterns and it has been shown that the system can recognize the same precipitation curves on different immunoplates and autoradiographs. In addition, the system reports the mean, the variance, and the area of the precipitation curves, thus allowing not only a qualitative comparison of two or more plates but also a quantitation of individual antigens or antibodies.

Algorithms↗

PubMedDate not supplied

Practical multipeptide synthesis: dedicated software for the definition of multiple, overlapping peptides covering polypeptide sequences.

A personal computer program for the conversion of linear amino acid sequences to multiple, small, overlapping peptide sequences has been developed. Peptide lengths and "jumps" (the distance between two consecutive overlapping peptides) are defined by the user. To facilitate the use of the program for parallel solid-phase chemical peptide syntheses for the synchronous production of multiple peptides, amino acids at each acylation step are laid out by the program in a convenient standard multi-well setup. Also, the total number of equivalents, as well as the derived amount in milligrams (depend-ending on user-defined equivalent weights and molar surplus), of each amino acid are given. The program facilitates the implementation of multipeptide synthesis, e.g., for the elucidation of polypeptide structure-function relationships, and greatly reduces the risk of introducing mistakes at the planning step. It is written in Pascal and runs on any DOS-based personal computer. No special graphic display is needed.

Amino Acid Sequence↗