Agnieszka Wosiak

Intra-uterine growth restriction as a risk factor for hypertension in children six to 10 years old

Summary Introduction Intra-uterine growth restriction (IUGR) is present in about 3–10% of live-born newborns and it is as high as 20–30% in developing countries. Since the 1990s, it has been known that abnormalities during foetal growth may result in cardiovascular disease, including hypertension in adulthood. Methods This study evaluated blood pressure parameters (using ambulatory blood pressure monitoring) in children aged six to 10 years old, born as small for gestational age (SGA), and compared them to their healthy peers born as appropriate for gestational age (AGA). Results In the SGA group, an abnormal blood pressure level (prehypertension or hypertension) was present significantly more often than in the AGA group (50 vs 16%, p < 0.01). This relationship also occurred in association with the type of IUGR (asymmetric p < 0.01, symmetric p < 0.05). Conclusion In SGA children, abnormal blood pressure values occurred more frequently than in AGA children.

Improving Children Diagnostics by Efficient Multi-label Classification Method

Using intelligent computational methods may support children diagnostics process. As in many cases patients are affected by multiple illnesses, multi-perspective view on patient data is necessary to improve medical decision making. In the paper, multi-label classification method—Labels Chain is considered. It performs well when the number of attributes significantly exceeds the number of instances. The effectiveness of the method is checked by experiments conducted on real data. The obtained results are evaluated by using two metrics: Classification Accuracy and Hamming Loss, and compared to the effects of the most popular techniques: Binary Relevance and Label Power-set.

On integrating clustering and statistical analysis for supporting cardiovascular disease diagnosis

Statistical analysis of medical data plays significant role in medical diagnostics development. However in many cases the statistics is not effective enough. In the paper we consider combining statistical inference with clustering in the preprocessing phase of data analysis. The proposed methodology is checked on cardiovascular data and used for developing methods of early diagnosis of hypertension in children. Experiments, conducted on the real data, have demonstrated that the proposed hybrid approach allowed to discover relationships which have not been identified by using only the statistical methods. We have observed approximately 30% growth in the number of correlations between diagnosed attributes. Moreover all the obtained statistically significant dependencies were stronger in clusters rather than in the whole datasets.

Feature selection for classification incorporating less meaningful attributes in medical diagnostics

In medical diagnostics there is a constant need of searching for new methods of attribute acquiring, but it is difficult to asses if these new features can support the existing ones and can be useful in medical inference. In the paper the methodology of discovering features which are less informative while considering independently, however meaningful for diagnosis making, is investigated. The proposed methodology can contribute to better use of attributes, which have not been considered in the diagnostics process so far. The experimental study, which concerns arterial hypertension as one of the civilization diseases demanding early detection and improved treatment is presented. The experiments confirmed that additional attributes enable obtaining the diagnostic results comparable to the ones received by using the most obvious features.

Preprocessing compensation techniques for improved classification of imbalanced medical datasets

The paper describes the study on the problem of applying classification techniques in medical datasets with a class imbalance. The aim of the research is to identify factors that negatively affect classification results and propose actions that may be taken to improve the performance. To alleviate the impact of uneven and complex class distribution, methods of balancing the datasets are proposed and compared. The experiments were conducted on five datasets — three binary and two multiclass. They comprise several data preprocessing methods applied on data and the classification with different techniques. The study shows that for some datasets there exists a combination of a certain preprocessing method and a classification technique which outperforms other approaches. For datasets with complex distribution or too many features the ratio of correctly predicted labels may be low regardless what resampling method and classification technique has been applied.

Unsupervised feature selection using reversed correlation for improved medical diagnosis

Statistical inference has been usually used for medical data analysis, however in many cases it appears not to be efficient enough. Cluster analysis enables finding out groups of similar instances, for which statistical models can be built more effectively. In the paper a feature selection method for finding clustering attributes, which are supposed to improve performance of statistical analysis, is proposed. The method consists in selecting reversed correlated features as attributes of cluster analysis. The proposed technique has been evaluated by experiments done on real data sets of cardiovascular cases. Experiment results showed that the presented approach stimulates efficacy of statistical inference applied to medical diagnosis.

Integrating Correlation-Based Feature Selection and Clustering for Improved Cardiovascular Disease Diagnosis

Based on the growing problem of heart diseases, their efficient diagnosis is of great importance to the modern world. Statistical inference is the tool that most physicians use for diagnosis, though in many cases it does not appear powerful enough. Clustering of patient instances allows finding out groups for which statistical models can be built more efficiently. However, the performance of such an approach depends on the features used as clustering attributes. In this paper, the methodology that consists of combining unsupervised feature selection and grouping to improve the performance of statistical analysis is considered. We assume that the set of attributes used in clustering and statistical analysis phases should be different and not correlated. Thus, the method consisting of selecting reversed correlated features as attributes of cluster analysis is considered. The proposed methodology has been verified by experiments done on three real datasets of cardiovascular cases. The obtained effects have been evaluated regarding the number of detected dependencies between parameters. Experiment results showed the advantage of the presented approach compared to other feature selection methods and without using clustering to support statistical inference.

Glomerular filtration decrease after diagnostic cardiac catheterisation in children with congenital cardiac malformation – the role of serum creatinine, cystatin C, neutrophil gelatinase and urine output monitoring

Introduction Diagnosis of contrast induced-nephropathy (CIN) by a classic renal biomarker such as creatinine concentration can be delayed because of various factors that can influence this marker. Changes in new biomarkers such as neutrophil-gelatinase associated lipocalin (NGAL) and cystatin C are postulated to be more sensitive for recognizing patients prone to CIN-acute kidney injury (AKI). Aim To investigate the role of NGAL and cystatin C as early biomarkers in the diagnosis of kidney injury after cardiac catheterisation. Material and methods The study group consisted of 50 patients with congenital heart malformation admitted for scheduled cardiac catheterisation. The biomarkers serum creatinine, serum NGAL and serum cystatin C were tested at 5 time-points sequentially from start to 48 h after the procedure. Results Significant changes were noted during the research in the serum creatinine concentration (p < 0.001) and serum NGAL concentration (p < 0.001). CIN-AKI, diagnosed by the modified Schwartz formula, occurred in 16 (32%) patients after 24 h and in 8 (16%) after 48 h. Subsequent analysis showed that serum creatinine significantly rose in the first 2 h of the study with simultaneous reduction in the eGFR. Maximum growth in serum NGAL occurred at 6 h after contrast administration and then returned to the baseline values at 24 h. Serum cystatin C level did not significantly change during the study. Conclusions We observed a transient decrease in eGFR and a rise of serum NGAL after 2 h but NGAL was most pronounced at 6 h after the procedure. The potential role of cystatin C as a biomarker of CIN-AKI was not proved.

12-month assessment of increase in total kidney volume in children and young adults with autosomal dominant polycystic kidney disease – a pilot study

IN T R O D U C T IO N: Autosomal dominant polycystic kidney disease (ADPKD) is the most common type of monogenic kidney disease. It is the cause of ESRD in 5–10% of adult patients who undergo renal replacement therapy. Owing to the increasing use of ultrasonography, occurrence of the disease has been observed among children. M A T E R IA L A N D M ET H O D S: The research group consisted of 19 patients with normal kidney function (12 girls and 7 boys aged 1.8–18.8 at the moment of examination) who suffered from ADPKD. 15 patients met the ultrasonographic criteria of the diagnosis. Although the remaining 4 patients did not meet the criteria, they had a strong family history of ADPKD and underwent USG which revealed at least 2 cysts. For each patient, the Total Kidney Volume (TKV), defined as the sum volume of both kidneys, was juxtaposed with BSA. The children underwent the examination approximately every 12 months. R E S U L T S: The members of the research group manifested a statistically significant annual increase in TKV (in 2015, 296.71 ± 178.67 cm 3 versus 350.38 ± 195.86 cm 3 in 2016, p = 0.019), as well as TKV in relation to body surface (in 2015, 191.23 ± 86.29 cm 3 /m 2 versus 221.15 ± 96.99 cm 3 /m 2 in 2016, p = 0.037). There were no apparent differences in the rate of total TKV increase which would depend either on the patients gender (girls 32.45 ± 51.88 cm 3 /m 2 /year versus boys 25.56 ± 71.00 cm 3 /m 2 /year, p = 0.81), or on the number of renal cysts revealed by USG (< 5 cysts 30.77 ± 61.6 cm 3 /m 2 /year versus ≥ 5 cysts 29.41 ± 58.22 cm 3 /m 2 /year, p = 0.96). C O N C L U S IO N: In children and young adults with ADPKD, the increase in total kidney volume (in relation to BSA) can be observed after a 12-month observation.

Supervised and unsupervised machine learning for improved identification of intrauterine growth restriction types

This paper concerns automated identification of intrauterine growth restriction (IUGR) types by use of machine learning methods. The research presents a comparison of supervised and unsupervised learning covering single and hybrid classification, as well as clustering. Supervised learning techniques included bagging with Naïve Bayes, k-nearest neighbours (kNN), C4.5 and SMO as base classifiers, random forest as a variant of bagging with a decision tree as a base classifier, boosting with Naïve Bayes, SMO, kNN and C4.5 as base classifiers, and voting by all single classifiers using majority as a combination rule, as well as five single classification strategies: kNN, C4.5, Naïve Bayes, random tree and sequential minimal optimization algorithm for training support vector machines. Unsupervised learning encompassed k-means and expectation-maximization algorithms. The major conclusion drawn from the study was that hybrid classifiers have demonstrated their potential ability to identify more accurately symmetrical and asymmetrical types of IUGR, whereas the unsupervised learning techniques produced the worst results.