Apparatus and method for dynamically determining a label for a cardiac potential signal via simulation of neural network trained data
Inventors
Ptaszek, Leon • Jain, Rohit • Ramani, Anand • H J, Yogisha • Nadapurohit, Sanjeev Shrinivas • Bharadwaj, Karthik K. • Verma, Shiva
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-12465263-B2
Publication Date
2025-11-11
Expiration Date
Abstract
An apparatus and method for determining a label dynamically using a potential signal. The apparatus includes a memory and at least a processor communicatively connected to the memory, wherein the memory contains instructions configuring the at least a processor to receive a user input comprising a labeled datum associated with the at least a potential signal, receive a plurality of stored data associated with the at least a potential signal, generate, using the at least a processor, a plurality of canonicalized data by processing the plurality of stored data, transmit, using a real time data simulator, the plurality of canonicalized data to a simulation module, generate, using the simulation module, a labeled prediction corresponding to a segmented datum.
Core Innovation
The invention provides an apparatus and a method for dynamically determining labels for cardiac potential signals, including electrocardiogram and electrogram signals. A catheter transducer detects a cardiac phenomenon and outputs a potential signal. Stored data are processed into canonicalized data and transmitted through a real-time data simulator to a trained simulation module, which segments the data and generates labeled predictions for display through a graphical user interface.
The system uses a convolutional neural network in a segmentation model and a recurrent neural network for prediction. Claim elements further recite data normalization into a consistent data format, validation against a predefined protocol, and feedback comprising corrections to a labeled prediction. The content also depicts temporal segmentation of electrogram signals, learning processes, and computing system implementations, while the problem being solved is not explicitly described.
Claims Coverage
The claim coverage includes two inventive features across two independent claims directed to an apparatus and a method for dynamically determining and displaying labeled predictions for cardiac potential signals.
Catheter-based apparatus for dynamically determining labels
An apparatus comprises a catheter with a transducer configured to detect a cardiac phenomenon and output a potential signal, a memory, and a processor. The processor receives labeled datum and stored signal data, generates canonicalized data, and transmits the data through a real-time data simulator to a trained simulation module. The simulation module uses a convolutional-neural-network segmentation model to segment the data and generates a labeled prediction for display via a graphical user interface.
Method for dynamically determining labels
A method comprises detecting a cardiac phenomenon and outputting a potential signal via a catheter transducer, receiving stored signal data and labeled datum, and generating canonicalized data. The method further comprises transmitting the canonicalized data through a real-time data simulator to a trained simulation module, generating segmented data and labeled predictions using a convolutional-neural-network segmentation model, and displaying the labeled predictions via a graphical user interface.
The independent claims cover apparatus and method implementations that process stored cardiac potential signal data into canonicalized data, segment the data using a convolutional-neural-network simulation module, and generate and display labeled predictions. Dependent claims further specify electrocardiogram data, normalization, protocol validation, recurrent-neural-network prediction, accuracy-threshold conditions, and user feedback corrections.
Stated Advantages
Not explicitly described in patent.
Documented Applications
Dynamically determining and displaying labels for cardiac potential signals, including electrocardiogram (ECG) and electrogram (EGM) signals.
Interested in licensing this patent?