Methodological framework of the CRC | Cancer Registy of Crete - Department of Social Medicine - University of Crete

Methodological framework of the CRC

The CRC has recently introduced a new research framework that aims to reform and enhance research capacity in Crete.

The new digital cancer monitoring system (CMS) is suitable for accommodating and managing “big data” according to international standards of disease coding (ICD10o) and data privacy. It is connected with a GIS system that exports instant reports, maps and other results by applying spatio-temporal analysis and dynamic models. A pilot study was performed to test the following: functionality, validity, reliability and accuracy.

Accredited personnel of the CRC will collect and import data using a unique password in an online platform (dbase). Validity of data is tested at real time through three stages:

a) during data entry

b) after data entry by the CMS auto-functions

c) after data entry by the CMS administrator

The CRC follows a circular monitoring process. A sample of this process is presented in the following figure.

Figure 1: Diagram of an integrated circular monitoring process.

The above circular process of the CRC ensures:

  1. Direct access to the information, input/output records, error management, mining data, feedback network.
  2. Instant reporting
  3. Connection with other statistical and spatial software systems such as SPSS, STATA, System for Geografical information [GIS (Arcmap)] or/and remote sensing  (ERDAS) in order to define probable correlation with environmental exposure etc. [figure 1]
  4. Creation of interactive maps and information tables [figure 2]

Figure 2: GIS flowchart

Figure 3: GIS visualization of real world and attributes (hidden information)

Furthermore, there is a three-stage data control through three different nodes  (1. recorders: during data entry, 2. System and analysis administrator: responsible for data mining and analysis 3. MD-oncologist: responsible for data quality).

All this information could help to perform spatio-temporal analysis, which can predict the occurrence of new cases (interpolation prediction models) in regions lacking data based on the registered data (time or location trends). In addition, the organization of the collected data would allow the identification of factors influencing the occurrence of cancer and the monitoring of cases of spatial clusters and hot spots of increased cancer incidence facilitating the implementation of preventive action and measures.