Читать книгу The Internet of Medical Things (IoMT) - Группа авторов - Страница 34

Health providers are familiar with the necessity of cleanliness in both the clinic and the operating room, but are not aware of the importance of cleaning their data.

Dirty data can swiftly ruin a large data analytics project, especially if multiple data sources are used to capture clinical or operational elements in slightly different formats. Data cleaning—also known as cleaning or scrubbing—guarantees accuracy, correctness, consistency, relevance, and in no way corruption of datasets.

While most data cleaning activities are still done manually, certain IT vendors provide automated scrubbing instruments that compare, contrast, and rectify big data sets using logic rules. These technologies may grow more sophisticated and accurate as machine learning techniques continue to progress rapidly, lowering time and cost necessary to guarantee high levels of accuracy and integrity in health data stores.