Hemoglobin (Hb) disorders named hemoglobinopathies represent together the most common monogenic disease worldwide. Their biological diagnosis is a complex process relying on the combination of several analytical methods. Currently, the patient phenotype characterization may be difficult due to the lack of selectivity of the techniques used. This thesis project was designed to develop first, a mass spectrometry platform allowing to perform a selective identification of the most common and clinically significant Hb variants and a precise quantification of Hb chains for Hb disorders diagnosis and therapeutic follow-up. These two methods were developed in a low-resolution mass spectrometer already present in clinical laboratories. Second, a strategy relying on high-resolution mass spectrometry was developed for the structural characterization of uncommon Hb variants. Selected diagnostic product ions were used to allow fast and reliable data interpretation of complex high-resolution MS data by unskilled mass spectrometry users.