Transfusions of packed red blood cells (PRBCs), given due to an oncological disease and its acute complications, are an indispensable part of anticancer therapy. However, they can lead to post-transfusion iron overload. The study aim was to evaluate the role of ferritin as a nonspecific marker of leukemic growth and marker of transfusion-related iron overload. We performed a longitudinal study of PRBC transfusions and changes in ferritin concentrations during the oncological treatment of 135 patients with childhood acute lymphoblastic and acute myeloblastic leukemia (ALL and AML, median age 5.62 years). At the diagnosis, 41% of patients had a ferritin level over 500 ng/mL, and 14% of patients had a ferritin level over 1000 ng/mL. At the cessation of the treatment, 80% of the children had serum ferritin (SF) over 500 ng/mL, and 31% had SF over 1000 ng/mL. There was no significant difference between SF at the beginning of the treatment between ALL and AML patients, but children with AML finished treatment with statistically higher SF. AML patients had also statistically higher number of transfusions. We found statistically significant positive correlations between ferritin and age, and weight and units of transfused blood. Serum ferritin at the moment of diagnosis can be a useful marker of leukemic growth, but high levels of SF are connected with iron overload in both AML and ALL.

Packed red blood cell (PRBC) transfusions are an important part of supportive treatment in oncology; however, when used frequently, they can be a result of transfusion-related iron overload. The aim of the study was to evaluate the role of ferritin as a non-specific marker of neoplastic growth and transfusion-related iron overload in children with lymphomas and solid tumors. We performed a longitudinal analysis of PRBC transfusions and changes in ferritin concentrations during oncological treatment of 88 children with lymphomas and solid tumors. A ferritin concentration above 500 ng/mL was diagnosed in 14.77% of patients at the moment of admission and 18.18% at the end of treatment. No differences were shown in serum ferritin in the context of tumor type-, sex-, and transfusion-related parameters. Those above the age of 10 demonstrated higher ferritin concentrations compared to subjects younger than 5 years of age. In addition, those over than 10 years old or above 30 kg in weight showed a tendency for better survival. All tested patients demonstrated highly significant correlations between ferritin at the 15th month of treatment or after therapy discontinuation and transfusion-related parameters. Interestingly, ferritin levels were found to lower back to the values before therapy shortly after its discontinuation. Transfusion parameters and ferritin levels had no influence on the survival of the studied cancer patients.

Ranked layers of binary classifiers are used for the linearization of learning sets composed of multivariate feature vectors. After transformation by ranked layer, each learning set can be separated by a hyperplane from the sum of other learning sets. Ranked layers can be designed, among others, from radial binary classifiers. This work elaborates on designing ranked layers from radial binary classifiers with movable centers.

Salvage autologous hematopoietic cell transplantation (auto-HCT) may be used to treat relapse of plasma cell myeloma occurring after previous auto-HCT. When an insufficient number of hematopoietic stem cells have been stored from the initial harvest, remobilization is necessary. Here, we aimed to analyze the efficacy and safety of different doses of cytarabine (total 800 vs. 1600 vs. 2400 mg/m²) for remobilization. Sixty-five patients, 55% male, with a median age at remobilization 63 years, were included. Remobilization was performed with cytarabine_800 in 7, cytarabine_1600 in 36, and cytarabine_2400 in 22 patients. Plerixafor rescue was used in 25% of patients receiving cytarabine_1600 and 27% of those receiving cytarabine_2400. Patients administered cytarabine_800 were not rescued with plerixafor. Remobilization was successful in 80% of patients (57% cytarabine_800; 86% cytarabine_1600; 77% cytarabine_2400; p = 0.199). The yield of collected CD34+ cells did not differ between the different cytarabine doses (p = 0.495). Patients receiving cytarabine_2400 were at the highest risk of developing severe cytopenias, requiring blood product support, or having blood-stream infections. One patient died of septic shock after cytarabine_2400. In summary, remobilization with cytarabine is feasible in most patients. All doses of cytarabine allow for successful remobilization. Cytarabine_2400 is associated with higher toxicity; therefore, lower doses (800 or 1600 mg/m²) seem to be preferable.

Both supervised and unsupervised machine learning algorithms are often based on regression to the mean. However, the mean can easily be biased by unevenly distributed data, i.e., outlier records. Batch normalization methods address this problem to some extent, but they also influence the data. In text-based data, the problem is even more pronounced, as distance distinctions between outlier records diminish with increasing dimensionality. The ultimate solution to achieving unbiased data is identifying the outliers. To address this issue, multidimensional scaling (MDS) and agglomerative-based techniques are proposed for detecting outlier records in text-based data. For both methods, two of the most common distance metrics are applied: Euclidean distance and cosine distance. Furthermore, in the MDS approach, both metric and non-metric versions of the algorithm are used, whereas in the agglomerative approach, the last-p and level cutoff techniques are applied. The methods are also compared with a raw-data-based method, which selects the most distant element from the others based on a given distance metric. Experiments were conducted on overlapping subsets of a dataset containing roughly 2000 records of descriptive image captions. The algorithms were also compared in terms of efficiency with a proposed algorithm and evaluated through human judgment based on the described images. Unsurprisingly, the cosine distance turned out to be the most effective distance metric. The metric-MDS-based algorithm appeared to outperform the others based on human evaluation. The presented algorithms successfully identified outlier records.