Apparatus and a method for generating a diagnostic label
Inventors
BABU, Melwin • Lalam, Sravan Kumar • Barve, Rakesh • Nandan, Kirnesh • Kunderu, Hari Krishna
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-12288621-B2
Publication Date
2025-04-29
Expiration Date
Abstract
An apparatus for generating a diagnostic label is disclosed. The apparatus includes at least a processor and memory communicatively connected to the at least a processor. The memory instructs the processor to receive a plurality of electrocardiogram signals and a plurality of electronic health records from a user. The memory instructs the processor to generate a plurality of structured electronic health records using the plurality of electronic health records. The memory instructs the processor to generate a plurality of representations as a function of the plurality of electrocardiogram signals and the plurality of structured electronic health records using a representation machine learning model. The memory instructs the processor to generate a diagnostic label as a function of the plurality of representations. The memory instructs the processor to display the diagnostic label using a display device.
Core Innovation
The invention relates to an apparatus for generating a diagnostic label from a plurality of electrocardiogram signals and a plurality of electronic health records. The apparatus receives the electrocardiogram signals from a user and receives the electronic health records, including a plurality of metadata, by utilizing optical character recognition (OCR) to convert the electronic health records into machine-encoded text and by extracting features from the electronic health records to reduce a dimensionality of a representation of the plurality of health records.
The apparatus generates a plurality of structured electronic health records using the plurality of electronic health records. The plurality of structured electronic health records includes a plurality of diagnostic codes, and generating the plurality of structured electronic health records includes classifying health data in the plurality of diagnostic codes with at least a corresponding time code representing a medical condition of the user within a given time period.
The apparatus then generates a plurality of representations as a function of the plurality of electrocardiogram signals and the plurality of structured electronic health records using a representation machine learning model. Generating the representations includes generating representation training data by inputting electrocardiogram signal representations with dummy pixels into the representation machine learning model, and outputting representation training data comprising the electrocardiogram signal representations with the dummy pixels replaced with filled-in values. The apparatus generates the diagnostic label as a function of the plurality of representations and displays the diagnostic label using a display device.
Claims Coverage
The consolidated content includes two independent claims, an apparatus claim and a method claim. Both center on OCR conversion of electronic health records, structured electronic health records with diagnostic codes and corresponding time codes, representation learning from electrocardiogram signals and structured electronic health records, and generation and display of a diagnostic label. The claims share the same core inventive features.
Diagnostic label generation using multi-modal representations
Generating a diagnostic label by receiving a plurality of electrocardiogram signals and a plurality of electronic health records, generating a plurality of representations as a function of the electrocardiogram signals and structured electronic health records using a representation machine learning model, generating the diagnostic label as a function of the representations, and displaying the diagnostic label using a display device.
OCR-based machine-encoded text for EHR intake
Receiving the plurality of electronic health records further comprises utilizing optical character recognition (OCR) to convert the plurality of electronic health records into machine-encoded text.
Structured EHR with diagnostic codes and time codes
Generating a plurality of structured electronic health records includes classifying health data in the plurality of diagnostic codes comprising at least a corresponding time code representing a medical condition of the user within a given time period.
Representation training with dummy pixels replaced with filled-in values
Generating representation training data includes inputting electrocardiogram signal representations with dummy pixels into the representation machine learning model and outputting representation training data comprising the electrocardiogram signal representations with the dummy pixels replaced with filled-in values, then training the representation machine learning model using the representation training data.
Across the independent apparatus and method claims, the core inventive coverage consists of OCR conversion of electronic health records into machine-encoded text; structured electronic health records with diagnostic codes and corresponding time codes; representation training data in which dummy pixels are replaced with filled-in values; generation of multiple representations from electrocardiogram signals and structured electronic health records; and generation and display of a diagnostic label as a function of those representations.
Stated Advantages
Not explicitly described in patent.
Documented Applications
Not explicitly described in patent.
Interested in licensing this patent?