GluCINDa (Glucose: Coherent Import. Neat Data.) an open-source tool for the consolidation of continuous glucose monitoring data for research

Continuous glucose monitoring (CGM) is a central data source in both clinical practice and diabetes research. Although dedicated analysis tools are available, the workflows converting manufacturer exports into analysis-ready datasets remain heterogeneous and largely manual. Existing tools that import CGM exports directly are limited to specific devices or platforms or require data upload to a server. We therefore developed GluCINDa (Glucose: Cohand several additional Python packageserent Import. Neat Data.) to serve as a systematic ingestion layer to harmonise CGM exports in a reproducible and traceable manner. GluCINDa is an open-source Python application that runs locally ensuring data privacy. It is operated via a graphical user interface and supports parallel processing for scalability. The core algorithm processes input files row-by-row, thus handling structural heterogeneities like multiple embedded tables, mid-file header shifts, and mixed numeric and non-numeric glucose values. Manufacturer-specific detection logic is implemented as a fixed rule set applied to header structure, keeping detection deterministic and auditable independent of downstream processing. Subject identifiers are derived exclusively from file names to avoid identifiable metadata in the harmonised output. GluCINDa was validated with 87 real-world CGM exports from three manufacturers, four download platforms, and in two languages. Outputs were identical across four alternative directory structures, three operating systems and multiple Python versions. Line-count verification confirmed that no input row was silently skipped, and file-by-file summaries confirmed plausible CGM stream identification across all 87 validation files, and heuristic-based selection was independently verified against a manual reference. Clinical plausibility of harmonised data was confirmed by comparing GluCINDa-derived ambulatory glucose profile metrics, computed with the iglu package, against manufacturer portal reports. Small observed deviations reflect known differences between platform-level aggregation and raw export content. Benchmarking against generic parsers and existing CGM tools showed that almost none could parse the validation files directly. GluCINDa formalises and standardises the ingestion and structural harmonisation of CGM exports with a deterministic, fully traceable workflow. By separating data ingestion from statistical analysis and retaining explicit provenance information for each extracted value, it reduces manual effort, enhances reproducibility, and supports multi-centre and real-world research settings in which heterogeneous device data must be consolidated reliably.