ClustAll: A Tool for Patient Stratification in Complex Diseases
Author Information
Author(s): Ortega-Legarreta Asier, Palomino-Echeverria Sara, Huergo Estefania, Lagani Vincenzo, Kiani Narsis A., Rautou Pierre-Emmanuel, Picola Nuria Planell, Tegner Jesper, Gomez-Cabrero David
Primary Institution: Navarrabiomed—Fundación Miguel Servet, Universidad Publica de Navarra
Hypothesis
Can ClustAll effectively stratify patients with complex diseases using clinical data?
Conclusion
ClustAll is a powerful tool for patient stratification that improves personalized medicine strategies.
Supporting Evidence
- ClustAll was validated using two public clinical datasets.
- It effectively handles missing values and mixed data types.
- ClustAll achieved high sensitivity and specificity in patient stratification.
- The tool allows for the identification of multiple stratifications within the same population.
- ClustAll incorporates parallel computing for enhanced efficiency.
- The package includes user-friendly tools for exploring stratifications.
- ClustAll's performance was benchmarked against standard clustering algorithms.
- Results showed ClustAll outperformed traditional methods in accuracy.
Takeaway
ClustAll helps doctors group patients based on their health data, making it easier to give them the right treatment.
Methodology
ClustAll uses a clustering framework to analyze clinical data, handling mixed data types and missing values.
Potential Biases
Potential biases may arise from the imputation of missing data.
Limitations
The study may not account for all complexities in clinical data and relies on the quality of input data.
Participant Demographics
The study included 659 patients with breast cancer and 918 patients with heart disease.
Statistical Information
P-Value
0.001
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website