Long-term electrocardiographic and physiological monitoring over a period lasting up to several years in duration can be provided through a continuously-recording subcutaneous insertable cardiac monitor (ICM). The sensing circuitry and the physical layout of the electrodes are specifically optimized to capture electrical signals from the propagation of low amplitude, relatively low frequency content cardiac action potentials, particularly the P-waves that are generated during atrial activation. In general, the ICM is intended to be implanted centrally and positioned axially and slightly to either the left or right of the sternal midline in the parasternal region of the chest. Additionally, the ICM includes an ECG sensing circuit that measures raw cutaneous electrical signals and performs signal processing prior to outputting the processed signals for sampling and storage.

A system and method for atrial fibrillation detection in non-noise ECG data with the aid of a digital computer are provided. Electrocardiography (ECG) features and annotated patterns of the features are maintained in a database, at least some of the patterns associated with atrial fibrillation. A classifier is trained based on the annotated patterns, the classifier implemented by a convolutional neural network. A representation of an ECG signal recorded by one or more ambulatory monitors is received. Noise is detected in the representations and ECG features in the representation falling within each of the non-noise temporal windows are detected. The trained classifier is used to identify patterns of the ECG features. A value indicative of whether portions of the representation are associated the patient experiencing atrial fibrillation is calculated. That one or more of the portions are associated with the patient experiencing atrial fibrillation is determined.

A system and method for atrial fibrillation detection in non-noise ECG data with the aid of a digital computer are provided. Electrocardiography (ECG) features and annotated patterns of the features are maintained in a database, at least some of the patterns associated with atrial fibrillation. A classifier is trained based on the annotated patterns, the classifier implemented by a convolutional neural network. A representation of an ECG signal recorded by one or more ambulatory monitors is received. Noise is detected in the representations and ECG features in the representation falling within each of the non-noise temporal windows are detected. The trained classifier is used to identify patterns of the ECG features. A value indicative of whether portions of the representation are associated the patient experiencing atrial fibrillation is calculated. That one or more of the portions are associated with the patient experiencing atrial fibrillation is determined.

A subcutaneous and cutaneous electrocardiography monitor configured for self-optimizing ECG data compression is provided. The monitors include a housing, an electrocardiographic front end circuit, a memory, and a micro-controller configured to: obtain a series of electrode voltage values based on the sensed electrocardiographic signals; use a plurality of selection schemes to choose one or more of a plurality of compression algorithms associated with each of the selection scheme for testing; test the selected compression algorithms including applying the compression algorithms chosen using each of the selection schemes to a segment of the electrode voltage series; analyze results of the testing; select one or more compression algorithms chosen using one of the selection schemes for compressing at least a portion of the electrode voltage series based on the analysis; obtain a compression of at least the portion of the electrode voltage series; and store the compression within the memory.

A subcutaneous and cutaneous electrocardiography monitor configured for self-optimizing ECG data compression is provided. The monitors include a housing, an electrocardiographic front end circuit, a memory, and a micro-controller configured to: obtain a series of electrode voltage values based on the sensed electrocardiographic signals; use a plurality of selection schemes to choose one or more of a plurality of compression algorithms associated with each of the selection scheme for testing; test the selected compression algorithms including applying the compression algorithms chosen using each of the selection schemes to a segment of the electrode voltage series; analyze results of the testing; select one or more compression algorithms chosen using one of the selection schemes for compressing at least a portion of the electrode voltage series based on the analysis; obtain a compression of at least the portion of the electrode voltage series; and store the compression within the memory.

An electrocardiography patch is provided. A backing includes two rounded ends connected by a middle section that is narrower than the two rounded ends. An electrode is positioned on a contact surface of the backing on each rounded end. A circuit trace is electrically coupled to each of the electrodes. A battery is positioned on an outer surface of the backing, opposite the contact surface, on one of the rounded ends.

An electrocardiography patch is provided. A backing includes two rounded ends connected by a middle section that is narrower than the two rounded ends. An electrode is positioned on a contact surface of the backing on each rounded end. A circuit trace is electrically coupled to each of the electrodes. A battery is positioned on an outer surface of the backing, opposite the contact surface, on one of the rounded ends.

A three-dimensional adhesive device to be attached to the body surface of a mammal including a microelectronic sensing system. The system includes (a) a three-dimensional adhesive body made of a pressure sensitive adhesive having an upper surface and a bottom surface; (b) a microelectronic system embedded in the body of the pressure sensitive adhesive; (c) one or more cover layer(s) attached to the upper surface; and (d) optionally a release liner releasably attached to the bottom surface. Suitably the microelectronic system is capable of sensing physical input such as pressure, vibration, sound, electrical activity (e,g. from muscle activity), tension, blood-flow, moisture, temperature, enzyme activity, bacteria, pH, blood sugar, conductivity, resistance, capacitance, inductance or other chemical, biochemical, biological, mechanical or electrical purposes.

A three-dimensional adhesive device to be attached to the body surface of a mammal including a microelectronic sensing system. The system includes (a) a three-dimensional adhesive body made of a pressure sensitive adhesive having an upper surface and a bottom surface; (b) a microelectronic system embedded in the body of the pressure sensitive adhesive; (c) one or more cover layer(s) attached to the upper surface; and (d) optionally a release liner releasably attached to the bottom surface. Suitably the microelectronic system is capable of sensing physical input such as pressure, vibration, sound, electrical activity (e,g. from muscle activity), tension, blood-flow, moisture, temperature, enzyme activity, bacteria, pH, blood sugar, conductivity, resistance, capacitance, inductance or other chemical, biochemical, biological, mechanical or electrical purposes.

An electrocardiography monitor is provided. A sealed housing includes one end wider than an opposite end of the sealed housing. Electronic circuitry is provided within the sealed housing. The electronic circuitry includes an electrographic front end circuit to sense electrocardiographic signals and a micro-controller interfaced to the electrocardiographic front end circuit to sample the electrocardiographic signals. A buzzer within the housing outputs feedback to a wearer of the sealed housing.