Developing a personalized remote patient monitoring algorithm: a proof-of-concept in heart failure.

Aims: Non-invasive remote patient monitoring is an increasingly popular technique to aid clinicians in the early detection of worsening heart failure (HF) alongside regular follow-ups. However, previous studies have shown mixed results in the performance of such systems. Therefore, we developed and evaluated a personalized monitoring algorithm aimed at increasing positive-predictive-value (PPV) (i.e. alarm quality) and compared performance with simple rule-of-thumb and moving average convergence-divergence algorithms (MACD). Methods and results: In this proof-of-concept study, the developed algorithm was applied to retrospective data of daily bodyweight, heart rate, and systolic blood pressure of 74 HF-patients with a median observation period of 327 days (IQR: 183 days), during which 31 patients experienced 64 clinical worsening HF episodes. The algorithm combined information on both the monitored patients and a group of stable HF patients, and is increasingly personalized over time, using linear mixed-effect modelling and statistical process control charts. Optimized on alarm quality, heart rate showed the highest PPV (Personalized: 92%, MACD: 2%, Rule-of-thumb: 7%) with an F1 score of (Personalized: 28%, MACD: 6%, Rule-of-thumb: 8%). Bodyweight demonstrated the lowest PPV (Personalized: 16%, MACD: 0%, Rule-of-thumb: 6%) and F1 score (Personalized: 10%, MACD: 3%, Rule-of-thumb: 7%) overall compared methods. Conclusion: The personalized algorithm with flexible patient-tailored thresholds led to higher PPV, and performance was more sensitive compared to common simple monitoring methods (rule-of-thumb and MACD). However, many episodes of worsening HF remained undetected. Heart rate and systolic blood pressure monitoring outperformed bodyweight in predicting worsening HF. The algorithm source code is publicly available for future validation and improvement.

Monitoring left ventricular assist device parameters to detect flow- and power-impacting complications: a proof of concept.

Aims: The number of patients on left ventricular assist device (LVAD) support increases due to the growing number of patients with end-stage heart failure and the limited number of donor hearts. Despite improving survival rates, patients frequently suffer from adverse events such as cardiac arrhythmia and major bleeding. Telemonitoring is a potentially powerful tool to early detect deteriorations and may further improve outcome after LVAD implantation. Hence, we developed a personalized algorithm to remotely monitor HeartMate3 (HM3) pump parameters aiming to early detect unscheduled admissions due to cardiac arrhythmia or major bleeding. Methods and results: The source code of the algorithm is published in an open repository. The algorithm was optimized and tested retrospectively using HeartMate 3 (HM3) power and flow data of 120 patients, including 29 admissions due to cardiac arrhythmia and 14 admissions due to major bleeding. Using a true alarm window of 14 days prior to the admission date, the algorithm detected 59 and 79% of unscheduled admissions due to cardiac arrhythmia and major bleeding, respectively, with a false alarm rate of 2%. Conclusion: The proposed algorithm showed that the personalized algorithm is a viable approach to early identify cardiac arrhythmia and major bleeding by monitoring HM3 pump parameters. External validation is needed and integration with other clinical parameters could potentially improve the predictive value. In addition, the algorithm can be further enhanced using continuous data.

Data-driven monitoring in patients on left ventricular assist device support.

INTRODUCTION: Despite an increasing population of patients supported with a left ventricular assist device (LVAD), it remains a complex therapy, and patients are frequently admitted. Therefore, a strict follow-up including frequent hospital visits, patient self-management and telemonitoring is needed. AREAS COVERED: The current review describes the principles of LVADs, the possibilities of (tele)monitoring using noninvasive and invasive devices. Furthermore, possibilities, challenges, and future perspectives in this emerging field are discussed. EXPERT OPINION: Several studies described initial experiences on telemonitoring in LVAD patients, using mobile phone applications to collect clinical data and pump data. This may replace frequent hospital visits in near future. In addition, algorithms were developed aiming to early detect pump thrombosis or driveline infections. Since not all complications are reflected by pump parameters, data from different sources should be combined to detect a broader spectrum of complications in an early stage. We need to focus on the development of sophisticated but understandable algorithms and infrastructure combining different data sources, while addressing essential aspects such as data safety, privacy, and cost-effectiveness.