Table 1.
Questionnaire input table and the numeric coding of the answers.
Table 2.
Summary of the questionnaire data (row 1–11), average data (12–13) and the imputed digital biomarkers (14–15).
The imputation increases the number of days (N) with available data with a factor of ~2.
Table 3.
Generic algorithm for constructing a questionnaire based digital biomarker.
Fig 1.
Distribution of averaged wellbeing data, WeBe, and motivation and self-confidence data, MotSC, (top) and the corresponding digital biomarkers, WeBe-i and MotSC-i (bottom).
Table 4.
Principal components (PC1-3), varimax rotated factor loadings and the belonging to the 2 digital biomarkers.
Fig 2.
Depicting the clinical course of patients using wellbeing and motivation/self-confidence data.
The average/digital biomarker view of wellbeing (WeBe/WeBe-i), motivation/self-confidence (MotSc/MotSc-i), and Addiction Monitoring Index (AMI) data for 3 patients (A-C) as time series during 4 months (x-axis = Treatment day). Symbols: Green circle = no alcohol detected; Red square = alcohol detected; Black diamond = all breathalyzer tests omitted.
Table 5.
The performance of the LSTM neural network model’s capability to predict exacerbation events (EE) of unseen patients from the Test dataset.
True Positives (TP), True Negatives (TN), False Positives (FP), False Negatives (FN), Sensitivity, Specificity and Matthews Correlation Coefficient (MCC) were calculated using the threshold that yielded the maximum MCC. Section A refers to predicting EEs occurring 1–3 days in the future, and section B refers to predicting EEs occurring 5–7 days in the future.