# Hypertension Risk ## Description Hypertension Risk is the percentage likelihood that the user has hypertension (as diagnosed by a physician). It corresponds to the percentage of people with the user’s risk profile who report that they have been diagnosed with hypertension (irrespective of their current blood pressure reading). Individuals with elevated hypertension risk might consider having their blood pressure monitored by a medical professional to determine if they meet the diagnostic criteria for hypertension. ## Participants Adults (18+ years of age) recruited from several hospital health clinics. ## Additional Data Collection Subjects were also asked whether they have been diagnosed with hypertension. ## Modeling Approach Blood flow signal was extracted and processed from facial video, and then blood flow features were extracted from blood flow signal. Feature selection was carried out on blood flow and demographic features to identify features predictive of hypertension status. A machine-learning based classifier was then created to predict an individual’s hypertension status based on those features. The model was created/trained using 80% of the subjects. The distribution of sex and hypertension status in this group is summarized in Table 1. | Subjects in videos | Composition | | ------------------------- | -------------------------- | | Sex distribution | 49.6 % Male; 50.4 % Female | | Hypertension distribution | 20 % Hypertensive | *Table 1: Hypertension distribution of training set* ## Model Performance The Hypertension Risk model was then validated for accuracy on an independent portion of the dataset (n=2,012) that was not used in training - validation set. Then the final accuracy was obtained on an independent portion of the dataset (n=2,012) that was not used in training - test set. The distribution of sex and hypertension status in these groups is summarized in Table 2. | Subjects in videos | Composition in Validation | Composition in Test | | ------------------------- | ------------------------- | ------------------------ | | Sex distribution | 49.8% Male; 50.2 % Female | 48.5 Male; 51.5 % Female | | Hypertension distribution | 20 % Hypertensive | 20 % Hypertensive | *Table 2: Hypertension distribution of validation and test sets* The accuracy on the test set calculated as area under the curve was 88.9% as shown in Figure 2.
*Figure 2: AUC of hypertension risk prediction* Predictions are displayed to the user as a percentage likelihood of having hypertension. A percentage of 45% or more suggests a risk of hypertension, while 55% or more suggests high risk of hypertension.