Method of diagnosing and treating asphyxia

A method for in vitro early diagnosing and treating asphyxia and disorders related thereto, comprising:

• a) selecting a plurality of asphyxia specific compounds selected from the first 30 ranked compounds of Table 2;
• b) quantitatively detecting a plurality of asphyxia specific compounds from the first 30 ranked compounds of Table 2 in the blood of a neonate;
• wherein said quantitatively detecting comprises:
• analyzing at least one blood sample of a neonate and measuring at least one measured value of concentration, level or amount of each of the selected compounds
• wherein said analyzing and measuring is by chromatography, mass spectrometry, enzyme-linked immunosorbent assay (ELISA), antibody linkage, immunochemical techniques, biochemical or enzymatic reactions or assays, or combinations thereof,
• c) calibrating said at least one measured value, and
• d) diagnosing asphyxia or an asphyxia disorder and administering a treatment to a diagnosed neonate selected from the group consisting of administering oxygen and administering hypothermia
• wherein said calibrating comprises:
• a) mathematically preprocessing the at least one measured value in order to reduce technical errors in said measuring;
• b) selecting at least one suitable classifying algorithm from the group consisting of logistic regression, linear or quadratic discriminant analysis, perceptron, shrunken centroids regularized discriminant analysis, random forests, neural networks, Bayesian networks, hidden Markov models, support vector machines, generalized partial least squares, partitioning around medoids, inductive logic programming, generalized additive models, gaussian processes, regularized least square regression, self organizing maps, recursive partitioning and regression trees, K-nearest neighbor classifiers, fuzzy classifiers, bagging, boosting, and naïve Bayes;
• and applying said selected classifier algorithm to preprocessed data of step a);
• c) training said at least one suitable classifying algorithm of step b) on at least one training data set containing preprocessed data from subjects divided into classes according to their asphyxia-related pathophysiological, physiological, prognostic, or responder conditions, in order to select a classifier function to map said preprocessed data to said conditions;
• and
• d) applying said trained at least one suitable classifying algorithm step c) to a preprocessed data set of a subject with unknown asphyxia-related pathophysiological, physiological, prognostic, or responder condition, and using the trained at least one suitable classifying algorithm to predict a class label of said data set in order to diagnose an asphyxia status of the subject.