Atrial fibrillation (AF) is a cardiac arrhythmia characterized by a highly irregular heart rate. It is the most prevalent arrhythmia in the general population in the United States and most developed countries, and is strongly associated with increased morbidity and mortality from adverse cardiovascular and cerebrovascular events. Moreover, patients with AF are correlated with increased healthcare expenditures, making the burden of AF on society extremely large. In this proof-of-concept study, we present an innovative device to be used as a screening tool for AF. The device consists of a modified electronic scale that is able to obtain an ECG and analyse it for the presence of AF. The classification algorithm is based on the RdR map method that plots RR intervals versus change in RR intervals. After optimizing the algorithm on a learning set of 77 ECGs from 45 patients, the performance of the device during a blind validation of 76 ECGs from 44 patients was: accuracy = 83%, sensitivity = 83%, specificity = 83% (N = 76 ECGs). Applying a constraint that each ECG recording contains a minimum of 7 beats in order to be eligible for classification, accuracy improved to 89% (sensitivity = 83%, specificity = 90%, N = 70). In conclusion, we present an innovative device to detect AF in a manner that can be implemented into current physician workflow without increasing the time or cost of each clinical encounter.