Clasificación de eritrocitos empleando modelos ocultos de Márkov / Red Blood cell classification with hidden Markov models

Se realiza un estudio del desempeño de los modelos ocultos de Márkov (HMM) en la clasificación morfológica supervisada de eritrocitos en muestras de sangre periférica de pacientes con anemia drepanocítica. Los contornos se representan de forma novedosa considerando las diferencias angulares en la curvatura de los puntos del mismo. El entrenamiento de cada modelo se realiza tanto con la descripción normal de los contornos como con la representación de la rotación de los mismos, para garantizar una mayor estabilidad en los parámetros estimados. Se desarrolla un proceso de validación cruzada de 5x1 para estimación del error. Se obtienen las medidas de sensibilidad, precisión y especificidad de la clasificación. Los mejores resultados en cuanto a sensibilidad se obtienen al clasificar eritrocitos pertenecientes a dos clases: normales (96 por ciento) y elongados (99 por ciento). Al considerar además una clase de eritrocitos con otras deformaciones los mejores resultados se obtienen realizando el entrenamiento de los modelos con la rotación de todos los contornos, que alcanzó sensibilidades de normales (94 por ciento), elongados (82 por ciento) y con otras deformaciones (76 por ciento)(AU)

A study of the performance of Hidden Markov Models (HMM) in morphologic supervised classification of erythrocytes in peripheral blood smears of patients with sickle cell disease is realized. Contours are represented in original way considering the angular differences in the curvature of the points of the same. The training of every model comes true with the normal description of the contours and with the representation of the rotation of the same, in order to guarantee a bigger stability in the esteemed parameters. A process of validation crossed of 5x1 for estimate of the error is developed. The measures of sensibility, precision and specificity of classification are obtained. The best results obtain when classifying erythrocytes in two classes, with sensibility values in normal of 96 percent and elongated 99 percent. In the classification of erythrocytes considering the class of other deformations better results obtain accomplishing the training of the models with the rotation of all the contours, that it attained sensibilities of normal (94 percent), elongated (82 percent) and with other deformations (76 percent)(AU)

[Incorporation of transferrin into the erythron in anemic and polycythemic conditions]. / Incorporación de la transferrina al eritrón en estados anémicos y policitémicos.

A ferrokinetic study was performed in 79 patients, 25 with absolute polycythaemia, 19 with relative polycythaemia and 35 presenting anaemia of different aetiology. The incorporation of transferrin into the erythron (ITE) was estimated on each case. The mean ITE values found in absolute primary and secondary polycythaemia and in secondary polycythaemia were, respectively, 281 +/- 85, 181 +/- 99 and 74 +/- 19 mumol/L tb/d (i.e., micromoles per litre of total blood per day). The mean ITE in iron deficiency anaemia was 110 +/- 43 mumol/L tb/d, in haemolytic anaemia this was 274 +/- 151 mumol/L tb/d, and in patients with myelodysplastic syndromes such values was 116 +/- 49 mumol/L tb/d. Significant differences were found for each group of patients with respect to a normal control group. Such findings are in accordance with the pathophysiology of the different pathologies studied. The values found in haemolytic anaemia and myelodysplastic syndromes are similar to those reported by others. The results of this study stress the advantages of evaluating the erythropoietic activity of the bone marrow by means of the ITE.