Artificial intelligence to assist specialists in the detection of haematological diseases.

Artificial intelligence, particularly the growth of neural network research and development, has become an invaluable tool for data analysis, offering unrivalled solutions for image generation, natural language processing, and personalised suggestions. In the meantime, biomedicine has been presented as one of the pressing challenges of the 21st century. The inversion of the age pyramid, the increase in longevity, and the negative environment due to pollution and bad habits of the population have led to a necessity of research in the methodologies that can help to mitigate and fight against these changes. The combination of both fields has already achieved remarkable results in drug discovery, cancer prediction or gene activation. However, challenges such as data labelling, architecture improvements, interpretability of the models and translational implementation of the proposals still remain. In haematology, conventional protocols follow a stepwise approach that includes several tests and doctor-patient interactions to make a diagnosis. This procedure results in significant costs and workload for hospitals. In this paper, we present an artificial intelligence model based on neural networks to support practitioners in the identification of different haematological diseases using only rutinary and inexpensive blood count tests. In particular, we present both binary and multiclass classification of haematological diseases using a specialised neural network architecture where data is studied and combined along it, taking into account the clinical knowledge of the problem, obtaining results up to 96% accuracy for the binary classification experiment. Furthermore, we compare this method against traditional machine learning algorithms such as gradient boosting decision trees and transformers for tabular data. The use of these machine learning techniques could reduce the cost and decision time and improve the quality of life for both specialists and patients while producing more precise diagnoses.

Haemoglobinopathies with high oxygen affinity. Experience of Erythropathology Cooperative Spanish Group.

Haemoglobinopathies are the world's most frequently found monogenic disorders. In the cases with high oxygen affinity, the decrease in the liberation of the oxygen determines a secondary erythrocytosis. In this work, we present 17 unrelated families of Caucasian race and of Spanish origin, with ten variants of haemoglobin or haemoglobinopathies with high oxygen affinity which were diagnosed in our laboratory. Of the ten haemoglobinopathies, in four (the Hb San Diego, the Hb Johnstown, the Hb Malmö and the Hb Columbia-Missouri), the change of amino acid affects zones of the contact alpha(1)beta(2); in two variants (the Hb Strasbourg and the Hb Syracuse), it affects the unions with 2,3-DPG in the central cavity; in the other two (the Hb Badalona and the Hb La Coruña), the cavity of contact with the group haem is affected; in one (Hb Bethesda), it affects the zone of contact alpha(1)beta(1;) and in one (Hb Olympia), the position 20 of the chain in the helix B in the surface of the protein is affected. In all cases, the change of amino acid, though of different form, facilitates that the quaternary structure of the haemoglobin becomes stable in its relaxed configuration so the transfer of oxygen and the P(50) value are decreased. All cases were sent to our laboratory because of shown erythrocytosis. In the majority of them, the diagnosis was done during an analysis of routine or for being relatives of the first ones.

[The H63D mutation in the HFE gene is related to the risk of hepatocellular carcinoma]. / La mutacion H63D del gen HFE se asocia con un riesgo aumentado de carcinoma hepatocelular .

AIM: To disclose whether mutations in the HFE gene inducing liver iron overload are related to the risk of hepatocellular carcinoma (HCC) in otherwise predisposed patients. PATIENTS AND METHODS: One hundred and ninety-six patients (161 males) diagnosed with HCC and 181 healthy controls were included in the study. All subjects were white Spaniards.C282Y and H63D mutations in the HFE gene were identified in leucocyte genomic DNA using a polymerase chain reaction (PCR) and specific restriction enzymes. RESULTS (CASES/CONTROLS): 1. Genotype distribution: a) C282Y mutation: homozygotes 1/0, heterozygotes 12/23, wild type 183/158 (p = 0.07, non significant); b) H63D mutation: homozygotes 9/5, heterozygotes 85/52, wild type 102/124 (0dds ratio 2.00, 95% C.I. 1.29-3.12, p = 0.002. Four cases and 6 controls were carriers of heterozygous mixed genotypes. 2. Allele frequencies: a) C282Y mutation: wild type allele 378/339, mutated allele 14/23 (p = 0.11, non significant); b) H63D mutation: wild type allele 289/300, mutated allele 103/62 (0dds ratio 1.72, 95% C.I. 1.19-2.50, p = 0.004). Age at diagnosis, gender and etiology of the underlying liver disease do not influence these findings. CONCLUSION: The C282Y mutation in the HFE gene is not related to the risk of HCC in non-hemochromatosis patients. The H63D mutation is associated with a higher risk of HCC in cirrhotic patients irrespective of their underlying liver disease.

C282Y and H63D mutations of HFE gene in patients with advanced alcoholic liver disease.

OBJECTIVE: To test the hypothesis that the heterozygous state for HFE gene mutations involved in the pathogenesis of hemochromatosis, that may induce an increase of hepatic iron content, may aggravate the liver damage induced by prolonged and excessive use of ethanol. PATIENTS AND METHODS: C282Y and H63D mutations of HFE gene were identified through polymerase chain reaction (PCR) on leukocyte DNA, in 125 consecutive patients diagnosed of advanced alcoholic liver disease (109 men, mean age 54 years, SD 11) and 181 healthy controls. All subjects were white Spaniards. RESULTS (CASES/CONTROLS): 1. Genotype distribution: a) mutation C282Y: no homozygotes, 10/23 heterozygotes, 115/158 normal (p = 0.60); b) mutation H63D: 9/5 homozygotes, 46/52 heterozygotes, 70/124 normal (Chi square 6.51, p = 0.039). 2. Allele frequencies: a) mutation C282Y: 240/339 normal, 10/23 mutated (p = 0.21); b) mutation H63D: 186/300 normal, 64/62 mutated (odds ratio 1.66, 95% CI 1.10-2.52, p = 0.01). CONCLUSIONS: Our results suggest that H63D mutation of the HFE gene, but not the C282Y mutation, is associated to the risk of developing advanced liver alcoholic disease.

[Serum transferrin receptor in healthy children. Diagnostic yield in ferropenic and infectious anemia]. / Receptor sérico de transferrina en niños sanos. Valor diagnóstico en la anemia infecciosa y en la ferropénica.

BACKGROUND: The serum transferrin receptor (TfR) presents certain advantages over other parameters of cellular iron status because it does not vary in acute or chronic diseases. OBJECTIVE: To establish reference ranges of TfR in healthy children for our laboratory, to define the distribution of this variable in children with acute illness and in those with iron deficiency, and to evaluate the diagnostic yield of TfR, the transferrin-receptor/ferritin ratio (TfR/F) and the transferrin-receptor-ferritin index (TfR-F) in distinguishing ferropenic from infectious anemia. PATIENTS AND METHODS: A descriptive, cross-sectional analysis was conducted in 132 children aged from 6 months to 16 years for a period of 18 months. The subjects were classified in three groups: healthy children, children with acute illness, and children with iron deficiency. The distribution of TfR and its diagnostic yield were evaluated. RESULTS: Of the 132 subjects, 30 were excluded because they lacked one or more of the parameters under analysis and 19 were excluded because they showed a thalassemic trait. In the 30 healthy children, the mean TfR concentration was 1.2 mg/l (SD 0.36) and the median was 1.02 (IQR 0.7-1.7). In the 32 children with acute illness, with or without anemia, TfR values were similar to those found in healthy children (p > 0.05). TfR values were higher in children with iron deficiency (21 patients; mean TfR value: 1.67 mg/l SD 0:98) than in healthy children but this difference was not statistically significant (p 0.08). The highest TfR values were found in the group with ferropenic anemia (mean TfR value: 2.13 mg/l SD 1.14) with a statistically significant difference between healthy children (p 0.04) and those with iron deficiency without anemia (p 0.01). The TfR/F ratio showed an optimal diagnostic yield in distinguishing ferropenic from acute disease anemia. If this ratio is higher than 80.7 ferropenia can be suspected as the cause of the anemia with a global value of the test of 100 % (95 % CI: 75.91-99.42). CONCLUSIONS: TfR could be useful in evaluating intracellular iron status in children. Acute disease does not alter TfR values and, in combination with ferritin, TfR offers an optimal diagnostical yield in distinguishing ferropenic from acute illness anemia.

[Acquired hemoglobin H disease associated with a myelodysplastic syndrome]. / Enfermedad de la hemoglobina H adquirida asociada a síndrome mielodisplásico.

Some patients found to have clonal panmyelopathies develop an acquired defect of haemoglobin synthesis clinically similar to haemoglobin H disease. A 58 year-old male diagnosed of simple refractory anaemia developed microcytosis and hypochromia. At the same time, his myelodysplastic syndrome became a refractory anaemia with excess of blasts. 33% of the red blood cells had "golf ball" inclusions after incubation with brilliant cresyl blue. Cellulose acetate electrophoresis revealed an haemoglobin H band. The globin chain synthesis alpha/beta ratio was 0.69. The molecular analysis demonstrated the integrity of both alpha genes in each chromosome. There were no familiar antecedent of haemoglobinopathy.