Reasons and features of big data preprocessing in neurosurgery. Practical guideline
- 2 February 2026
- journal article
- Published by Federal State Budgetary Institution "National Medical Research Center named after. V.A. Almazov" Ministry of Health of the Russian Federation" in Russian Neurosurgical Journal named after Professor A. L. Polenov
- Vol. 17 (4) , 22-27
Abstract
Current paper presents the features of big data preprocessing, to which artificial intelligence technologies can be applied in clinical neurosurgery. At the initial stage of creating patient registries, the data are characterized by missing values and variability, which requires the use of specific preprocessing methods. Data cleaning, handling missing values, and addressing variability issues at the initial stage are critically important for subsequent analysis and the application of artificial intelligence technologies. Data preprocessing includes detecting statistical outliers, normalization, feature extraction, and feature selection. These steps are performed using methods such as complete case analysis, imputation techniques, the last observation carried forward method, and dimensionality reduction methods. The application of these methods improves the accuracy and efficiency of artificial intelligence algorithms and enhances the interpretation of results. This paper provides practical recommendations for neurosurgeons on the selection and application of data preprocessing methodsKeywords
This publication has 11 references indexed in Scilit:
- Establishing of big data clinical dataset in brain vessel aneurysm researchSiberian Scientific Medical Journal, 2023
- Digital medicine and the curse of dimensionalitynpj Digital Medicine, 2021
- Managing Missing Data in Patient Registries2018
- Dimension reduction methods for microarray data: a reviewAIMS Bioengineering, 2017
- Statistical description for survival dataAnnals of Translational Medicine, 2016
- Data Cleaning2016
- Collinearity: a review of methods to deal with it and a simulation study evaluating their performanceEcography, 2012
- Pearson Correlation Coefficient2009
- Model-Dependent Variance Inflation Factor Cutoff ValuesQuality Engineering, 2002
- Regression Shrinkage and Selection Via the LassoJournal of the Royal Statistical Society Series B: Statistical Methodology, 1996