Evaluated the Potential of Near Infrared Spectroscopy to Detect ABO Blood Grouping in Hair.

Introduction: The use of head hair for the identification of ABO blood groups using the absorption elution (AE) method requires time and expertise. Therefore, developing a new method which is time-saving and more straightforward using Near Infrared spectroscopy is interesting but to date the prior knowledge is insufficient.

Aim: To evaluate the potential of near infrared spectroscopy to detect ABO blood group in hair.

Methodology: The spectral classification of all samples was carried out using principal component analysis (PCA) and soft independent modelling of class analogy (SIMCA).

Results and Conclusions: The accuracy of PCA data by cross validation was more than 90%. However, the PCA models were tested by the test sample sets, which giving a percentage accuracy of only between 14.76% and 42.08%. The reasons for this may be either due to the similarity of the structure of the A, B and H antigens, the in homogeneity of the samples or the dispersal of small amounts of the antigens throughout the cortex of hair.It was not possible to differentiate between the different blood groups from the hair samples collected using NIRS. In further studies larger amounts of hair should be included in the samples and the reference method should be more specific to the antigen.