Personalized symptoms forecasting for pollen-induced allergic rhinitis sufferers

Area of Science:

• Allergy and Immunology
• Computational Intelligence
• E-health

Background:

• Hay fever (pollen-induced allergic reaction) significantly impacts quality of life.
• Symptom severity varies due to individual factors like genetics and pollen thresholds.
• Existing information services lack personalized forecasting for hay fever sufferers.

Purpose of the Study:

• To develop an information service providing personalized pollen-induced symptom forecasts.
• To enhance the quality of life for individuals suffering from hay fever.
• To create customizable forecasting models for individual users.

Main Methods:

• Collected allergic symptom data from the Personal Hay Fever Diary system.
• Utilized European Aeroallergen Network pollen concentration data.
• Applied computational intelligence methods for data analysis and model development.

Main Results:

• Developed highly customizable forecasting models for personalized warnings.
• Achieved a correlation coefficient of r=0.71±0.17 in a pilot study (Vienna and Lower Austria).
• Demonstrated system performance with r=0.66±0.18 in a larger user group.

Conclusions:

• Combining diverse data sources with advanced data engineering enables innovative e-health services.
• Personalized forecasts offer direct and tailored information to allergic rhinitis patients.
• The developed system improves quality of life for hay fever sufferers through customized alerts.