Corresponding Author: Gerald C. Hsu, eclaireMD Foundation, USA.
The author has extended his 8.5-year T2D research to focus on the relationship between metabolic diseases and the risk probability of having a heart attack or stroke.
He has developed big data based numerical simulation models using ~1.5M data he collected. Initially, he chose age, gender, race, family history, smoking, drinking, drug abuse, personal medical history, and weight/waistline to establish a static baseline. He then applied his learned hemodynamics knowledge to develop a macro-simulated mathematical model for the dynamic situations of blood blockage (using fluid dynamics concept) and artery rupture (using solid mechanics concept). He utilized 81,900 data of chronic disease conditions (output data of obesity, diabetes, hypertension, and hyperlipidemia) within the past 2,555 days (1/1/2012 – 12/31/2018) to compute the risk probability of having a heart attack or stroke in the near future. Finally, he conducted sensitivity analyses to cover the probability variance by using different weighting factors (WF).
Comparing the results from the worst year, 2000, to the health-improving period of 2012-2018, the probability values are:
- 2000 with BMI 31: 74%
- 2001-2006: Three episodes of chest pain
- 2012 with BMI 29: 62%
- 2017: 26.4% (Comparable with Framingham’s 2017 result: 26.7%)
- 2018 with BMI 25: 31%
In summary, over eight years, he has an average of 34% probability with +/- 10% variance of WF sensitivity.
The mathematical simulation results are validated by his past health examination reports. This big data based dynamic simulation approach using GH-Method: math-physical medicine will provide an early warning to patients with chronic disease of having a heart attack or stroke in the future.