Cardiovascular assessment of patients suspected of having COVID-19
Inventors
Sharma, Puneet • Rapaka, Saikiran • Schmuecking, Ingo
Assignees
Publication Number
US-12334224-B2
Publication Date
2025-06-17
Expiration Date
2040-05-19
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Abstract
Systems and methods for assessing cardiovascular disease of a patient are provided. Patient data of the patient is received. The patient data may include one or more input medical images of a chest of the patient, results of an assessment of a lung disease performed based on the one or more input medical images, demographic and clinical data of the patient, and cardiovascular imaging exams of the patient. One or more risk scores are computed for the patient based on the patient data using a trained machine learning based network.
Core Innovation
The invention provides systems and methods for assessing cardiovascular disease in patients confirmed as having or suspected of having COVID-19 or other lung diseases. Patient data, including one or more chest medical images, results of lung disease assessment based on these images, demographic and clinical data, and cardiovascular imaging exams, is received. Using a trained machine learning based network, one or more cardiovascular risk scores are computed based on the patient data to assess cardiovascular disease.
The problem addressed is the difficulty in assessing the cardiovascular system of patients suspected of having COVID-19 using existing automatic techniques. Chest CT imaging typically performed for lung disease assessment is not optimized for cardiovascular disease assessment due to acquisition without contrast agents and cardiac gating, leading to lower image quality of cardiovascular structures. Hence, existing methods cannot effectively assess cardiovascular disease from such images.
Embodiments overcome this challenge by using a machine learning system that computes cardiovascular risk scores from patient data captured primarily for lung disease diagnosis, thereby enabling cardiovascular assessment without additional imaging requirements. This allows improved patient management, stratification, and resource use, particularly important for clinicians treating COVID-19 patients. The invention further contemplates integration of sensor data and clinical data for continuous patient monitoring and risk assessment.
Claims Coverage
The patent includes three independent claims covering a computer implemented method, an apparatus, and a non-transitory computer readable medium, all relating to assessing cardiovascular and lung disease using patient data and a trained machine learning based network.
Computing a combination metric representing cardiovascular and lung disease assessment
Receiving patient data comprising input chest medical images, metrics quantifying lung disease abnormality patterns based on the images, and cardiovascular imaging exam results; and computing a combination metric representing the assessment of cardiovascular disease and lung disease using a trained machine learning based network that processes the input images, lung abnormality metrics, and cardiovascular exam results.
Management of patient using continuous physiological sensor data
Continuously receiving sensor data of physiological measurements of the patient and monitoring the patient based on this continuously received sensor data in combination with the computed combination metric.
Application to lung diseases including COVID-19 and viral pneumonia
The patient lung disease can be specifically COVID-19 or any other viral pneumonia.
These inventive features together describe a system and method for integrated cardiovascular and lung disease assessment using multi-modal patient data and machine learning, enabling comprehensive risk scoring and patient management for lung disease patients, particularly COVID-19.
Stated Advantages
Assist clinicians in patient management decisions and patient stratification.
Utilize patient data acquired in clinical workflows for lung disease, minimizing additional clinician time.
Provide improved patient outcomes through accurate cardiovascular risk assessment despite lower quality cardiovascular imaging.
Enable continuous monitoring of cardiac risk via integration of sensor data with cardiovascular risk scores.
Documented Applications
Cardiovascular assessment of patients confirmed as having or suspected of having COVID-19 or other lung diseases.
Patient management decisions including ordering additional tests, changing therapies, and managing clinical resources based on cardiovascular risk scores.
Continuous patient monitoring for cardiac injury or stress by combining sensor data and cardiovascular risk assessment.
Interested in licensing this patent?