Valores anômalos e dados faltantes em estudos clínicos e experimentais / Anomalous values and missing data in clinical and experimental studies

Durante a análise dos dados de uma pesquisa científica, é habitual deparar-se com valores anômalos ou dados faltantes. Valores anômalos podem ser resultado de erros de registro, de digitação, de aferição instrumental, ou configurarem verdadeiros outliers. Nesta revisão, são discutidos conceitos, exemplos e formas de identificar e de lidar com tais contingências. No caso de dados faltantes, discutem-se técnicas de imputação dos valores para evitar a exclusão do sujeito da pesquisa, caso não seja possível recuperar a informação das fichas de registro ou reabordar o participante

During analysis of scientific research data, it is customary to encounter anomalous values or missing data. Anomalous values can be the result of errors of recording, typing, measurement by instruments, or may be true outliers. This review discusses concepts, examples and methods for identifying and dealing with such contingencies. In the case of missing data, techniques for imputation of the values are discussed in, order to avoid exclusion of the research subject, if it is not possible to retrieve information from registration forms or to re-address the participant

IClustering analysis:assistance to verification of radiotherapy treatment plans / 中华放射肿瘤学杂志

Objective To use clustering analysis to help physicians detect abnormal parameters in radiotherapy treatment plans and improve the efficiency of plan verification. Methods From 2010 to 2015, 835 breast cancer treatment plans for using 4?field hybrid intensity?modulated radiotherapy from MOSAIQ were collectted. Fractional dose, beam angle, and monitor unit were used as featured parameters of a treatment plan to generate a dataset. The K?means clustering algorithm based on principal component analysis was used to perform a clustering analysis of the dataset and divide the dataset into different clusters. The outliers of clusters were automatically detected based on the distance threshold. The outlier?contained treatment plans were manually verified by physicians to determine the accuracy of clustering analysis in detection of abnormal plans. Results In the clustering analysis, the sample space composed by parameters of treatment plans for breast cancer was divided into 4 clusters, 3 of which had outliers detected. In the targeted treatment plans, 3 plans became outliers because of special target volume and the other 4 plans needed improvement. Conclusions Clustering analysis is effective to help physicians to independently verify treatment plans.

Comparison evaluation of statistical analysis methods for inter-laboratory comparison statistical analysis methods / 中国药学杂志

OBJECTIVE: To identify the characteristics and application conditions of currently statistical methods for inter-laboratory comparison and proficiency testing results analysis in drug control, combining with the international standards and practical application. METHODS: Based on the ISO/IEC 17043-GB/T28043, start beging with the definition of outliers, the methods currently used in inter-laboratory comparison and proficiency testing results analysis both at home and abroad methods were summarized and compared according to the ISO/IEC 17043-GB/T28043, with focused on the testing the stability and repeatability of laboratory testing proficiency. The most-used stastical methods of mter-laboratory comparison and proficiency testing during present domestic and international drug testing was sumarrized and contratively analyzed. RESULTS: The application conditions of different methods good contrast were providedidentified application condition of methods. CONCLUSION: Robust Z-score has more advantages in current stage of inter-laboratory comparison and proficiency testing results analysis.

Discrimination and explanation of statistical errors in the analysis of the linear correlation and regression / 基础医学与临床

In this paper,a great many examples of misusing the analysis of the linear correlation and regression are unveiled.Obviously,it is extremely important for people to apply the linear correlation and regression correctly through understanding and mastering the basic concept and approach in this field.It is necessary for a great of practical workers to learn the knowledge on how to perform the analysis of the linear correlation and regression rationally,since people need this kind of konowledge and the textbooks of statistics are not able to meet the needs.

Diagnostic Usefulness of Outliers Count in Quantitative EMG

The reference values of motor unit action potentials(MUAP) of first dorsal interosseous and tibialis anterior muscles were measured in 50 healthy subjects. The MUAPs were recorded with a concentric needle electrode and extracted with a decomposition method. Sixty six patients with neuropathy were also studied in the same way with a count of outliers and measurement of mean values. The mean values of amplitude, area, duration and thickness were 667.74+/-204.34 V, 992.26+/-253.18 Vms, 9.75+/-1.95 ms and 1.49+/-0.26 ms, respectively in the first dorsal interosseous muscles, and 612.88+/-140.13 V, 1172.84+/-199.21 Vms, 11.41+/-2.48 ms and 1.93+/-0.34 ms respectively in the tibialis anterior muscles. There was no significant difference in age and gender of normal subjects. The amplitude was the most sensitive parameter to detect abnormality in a count of outliers and measurement of mean values. The outliers count in duration showed a higher sensitivity than the measurement of mean values(p<0.05), but not in amplitude, area or thickness. Based on the results of this study, the count of outliers was more sensitive than the measurement of mean values in the diagnosis of patients with neuropathy. Further more, less numbers of MUAPs were needed for the evaluation of the outliers count. Through this method we could save the evaluation time and the patients felt more comfortable.