Related Experiment Videos
Validation of a Web-based prognostic system for breast cancer
Johan Lundin1, Mikael Lundin, Jorma Isola
1Department of Oncology, HUCH Clinical Research Institute, University of Helsinki, Haartmaninkatu 4, PO Box 105, FIN-00290 Helsinki, Finland. johan.lundin@helsinki.fi
Studies in Health Technology and Informatics
|September 14, 2004
Summary
A new website provides breast cancer survival curves using Finnish patient data. Comparing this with US data shows good agreement, aiding clinical decision-making.
Area of Science:
- Oncology
- Biostatistics
- Health Informatics
Background:
- Breast cancer prognosis relies on accurate survival estimates.
- Predictive models are crucial for personalized treatment strategies.
- Utilizing large population-based databases enhances prognostic accuracy.
Purpose of the Study:
- To develop and validate a web-based tool for breast cancer survival prediction.
- To compare survival estimates derived from a Finnish nationwide database (FinProg) with US data (SEER).
- To assess the agreement between prognostic profiles across different large, unselected patient cohorts.
Main Methods:
- Development of a website tool for inputting patient prognostic factors.
- Generation of survival curves using the Finnish FinProg database (n=2,842).
- Comparison of FinProg-derived estimates with US SEER database estimates for breast cancer patients (1991-2).
Main Results:
- A reasonable agreement was observed between Finnish and US survival estimates.
- Significant differences in estimates were found for only 2 out of 19 analyzed prognostic profiles.
- The developed system demonstrates the utility of large, unselected databases for survival estimation.
Conclusions:
- Web-based tools utilizing nationwide data can provide reliable breast cancer survival predictions.
- Cross-database comparisons confirm the generalizability of prognostic models.
- This system facilitates knowledge sharing and supports clinical decision-making in breast cancer treatment.