Monitoring Hyperkalemia Using Point-of-care AI-enabled ECG Device Among Patients With Chronic Kidney Disease Stages 3b-5

A mobile AI-enhanced electrocardiogram device will be used to help patients monitor hyperkalemia. The device, shaped like a stick of chewing gum, collects time-series data of electrocardiogram lead I through two electrodes. Deep convolutional neural network algorithm can read the data and generated categories of plasma potassium (\<5 mmol/L, 5 - \<5.5 mmol/L, 5.5 - \<6.0 mmol/L and ≥ 6.0 mmol/L). Study participants receive the device after randomization. The staff establishes a WeChat group for each participant and reminds them to monitor plasma potassium every 3 days. When the detected plasma potassium category falls in 5 - \<5.5 mmol/L, they will be reminded to watch diet and to verify plasma potassium again. More intensive reminding (once daily) will be sent if plasma potassium category falls in 5.5 - \<6.0 mmol/L or ≥ 6.0 mmol/L. An extra outpatient appointment will also be made as soon as possible for the patients when their detected plasma potassium category is ≥ 6.0 mmol/L.