The portable device improves the detection of atrial fibrillation after ablation

Songqun Huang, Teng Zhao, Chao Liu, Aihong Qin, Shaohua Dong, Binhang Yuan, Wenhui Xing, Zhifu Guo, xin miao huang, Yong‐Mei Cha, Jiang CaoAuthorea 2020 · DOI

Abstract

Asymptomatic recurrences of atrial fibrillation (AF) are common following catheter ablation, often going undetected by traditional intermittent monitoring. This study evaluates the performance of a handheld single-lead ECG device (BigThumb®) paired with an artificial intelligence (AI) algorithm for post-ablation AF screening. In a randomized controlled trial of 218 patients, the BigThumb group (BT Group) demonstrated a significantly higher detection rate of AF recurrence (35.8% vs. 21.1% in the traditional group) and nearly double the adherence rate to oral anticoagulation therapy. The integrated AI algorithm achieved a sensitivity of 94.4% and specificity of 98.5%, outperforming standard automated detection. These findings suggest that AI-powered portable monitoring provides a more effective strategy for long-term post-ablation management.

The Challenge of Asymptomatic Recurrence

Catheter ablation is a primary treatment for atrial fibrillation, yet the procedure does not guarantee a permanent cure. A significant clinical hurdle is asymptomatic recurrence—episodes where the heart returns to an irregular rhythm without the patient experiencing noticeable symptoms.

Traditional follow-up typically relies on periodic Holter monitoring (24–48 hour recordings every few months). However, because AF is often paroxysmal (intermittent), these brief "snapshots" frequently miss recurrences. Undetected AF leaves patients at high risk for stroke and heart failure, particularly if they discontinue necessary anticoagulation therapy under the false impression that they are AF-free.

The BigThumb System and AI Integration

To address these monitoring gaps, researchers utilized the BigThumb® device, a handheld single-lead ECG monitor that transmits data to a smartphone application.

Handheld single-lead ECG device (BigThumb) and smartphone interface.

Unlike standard devices that rely on simple heuristic algorithms, this system employs a deep learning AI algorithm to analyze recordings. The AI was trained to distinguish between normal sinus rhythm and AF while filtering out the "noise" and signal variability common in patient-operated handheld devices. In head-to-head comparisons, the AI's sensitivity (94.4%) and specificity (98.5%) were significantly higher than those of traditional automated software (90.7% and 96.2%, respectively).

Clinical Trial and Results

The study randomized 218 post-ablation patients into two groups: the BT Group (using the portable device at least three times daily) and the TF Group (traditional follow-up with scheduled Holter monitors).

Improved Detection Rates

Over a follow-up period of approximately one year, the BT Group generated a massive dataset of 26,133 ECG recordings. The "AF-free survival rate" was found to be 64.2% in the BT Group compared to 78.9% in the TF Group. In this clinical context, a lower survival rate indicates superior diagnostic performance; the portable device caught many recurrences that traditional methods simply missed.

Intelligent diagnosis report showing AF risk and heart rate.

Behavioral Impact and Adherence

One of the most significant findings was the impact on patient behavior. Patients in the BT Group were significantly more likely to adhere to their oral anticoagulation (OAC) regimens (51.0% vs. 25.4%). The ability to monitor their own heart rhythm and receive immediate AI-driven feedback appeared to empower patients, making them more diligent in managing their condition and potentially reducing their long-term stroke risk.

Limitations and Future Directions

While the results are promising, the study noted a decline in patient engagement over time, with recording frequency dropping from 1.19 per day in the first quarter to 0.36 per day after three months. This "user fatigue" remains a challenge for long-term digital health compliance. Additionally, as a single-center study, further multi-center trials are needed to confirm these results across diverse populations.

Conclusion

The integration of AI-powered portable ECG devices into post-ablation care significantly enhances the detection of silent AF recurrences. By providing continuous, high-fidelity monitoring and improving medication adherence, these tools represent a major step forward in the personalized management of atrial fibrillation.

Summary updated

Abstract

Background: Asymptomatic recurrences of atrial fibrillation (AF) are common after ablation of AF. Objective: We aimed to analyze the performance of the mobile ECG device using artificial intelligence (AI) algorithm in detection of AF after ablation. Method: A randomized controlled trial of AF screening using a handheld single-lead ECG monitor (BigThumb®) or a traditional follow-up strategy was conducted in patients with non-valvular AF after catheter ablation. Consecutive patients were randomized to either BigThumb Group (BT Group) or Traditional Follow-up Group (TF Group). Monitoring data was collected and analyzed. The ECGs collected by BigThumb were compared using the automated AF detection algorithm, AI algorithm and cardiologists’ manual review. Subsequent changes in adherence on oral anticoagulation of patients were also recorded. Result: We studied 218 patients (109 in BT Group, 109 in TF Group). After a follow-up of 345.4±60.2 days, AF-free survival rate was 64.2% in BT Group and 78.9% in TF Group (P=0.0163), with more adherence on oral anticoagulation in BT Group (P=0.0052). The participants in the BT Group recorded 26133 ECGs during the follow-up, among which 3299 (12.6%) were diagnosed as AF by cardiologists’ manual review. The sensitivity and specificity of the AI algorithm were 94.4% and 98.5% respectively, which are significantly higher than the automated AF detection algorithm (90.7% and 96.2%). Conclusion: We found that follow-up after AF ablation using BigThumb leads to a more frequent detection of AF recurrence and more adherence on oral anticoagulation. Artificial intelligence algorithm improves the accuracy of ECG diagnosis.
BibTeX
@misc{huang2020portable,
  author = {Songqun Huang and Teng Zhao and Chao Liu and Aihong Qin and Shaohua Dong and Binhang Yuan and Wenhui Xing and Zhifu Guo and xin miao huang and Yong‐Mei Cha and Jiang Cao},
  title = {{The portable device improves the detection of atrial fibrillation after ablation}},
  year = {2020},
  doi = {10.22541/au.159657690.03276294},
  howpublished = {Authorea},
}
Figure