Invented by Puneet Sharma, Dorin Comaniciu, Siemens Healthcare GmbH

The market for plaque vulnerability assessment in medical imaging is rapidly growing as healthcare professionals recognize the importance of early detection and prevention of cardiovascular diseases. Plaque vulnerability assessment refers to the evaluation of the stability and composition of atherosclerotic plaques, which are fatty deposits that build up on the walls of arteries. Cardiovascular diseases, including heart attacks and strokes, are the leading cause of death worldwide. It is estimated that approximately 17.9 million people die each year due to cardiovascular diseases, accounting for 31% of all global deaths. Atherosclerosis, the underlying cause of most cardiovascular diseases, is a progressive condition characterized by the accumulation of plaques in the arteries. These plaques can rupture, leading to the formation of blood clots that can block blood flow to vital organs, causing severe damage or even death. Traditionally, the assessment of plaque vulnerability has been challenging, as it requires invasive procedures such as angiography or intravascular ultrasound. However, advancements in medical imaging technology have revolutionized the field, enabling non-invasive assessment of plaque vulnerability using techniques such as computed tomography (CT) angiography, magnetic resonance imaging (MRI), and positron emission tomography (PET). The market for plaque vulnerability assessment in medical imaging is driven by several factors. Firstly, the increasing prevalence of cardiovascular diseases globally has created a demand for more accurate and efficient diagnostic tools. Early detection of vulnerable plaques can help identify high-risk patients who may benefit from preventive interventions, such as lifestyle modifications or medication. Secondly, technological advancements in medical imaging have significantly improved the accuracy and resolution of plaque vulnerability assessment. For instance, CT angiography can provide detailed images of the coronary arteries, allowing healthcare professionals to identify and characterize plaques with high precision. Similarly, MRI can assess plaque composition and identify features associated with instability, such as lipid-rich cores and thin fibrous caps. Furthermore, the growing adoption of artificial intelligence (AI) and machine learning algorithms in medical imaging has further enhanced the capabilities of plaque vulnerability assessment. AI algorithms can analyze large volumes of imaging data and identify subtle patterns or features that may not be easily detectable by human observers. This can help improve the accuracy and efficiency of plaque vulnerability assessment, leading to better patient outcomes. The market for plaque vulnerability assessment in medical imaging is expected to witness significant growth in the coming years. According to a report by Grand View Research, the global plaque vulnerability assessment market is projected to reach $1.3 billion by 2027, growing at a compound annual growth rate (CAGR) of 6.5% from 2020 to 2027. The report attributes this growth to factors such as the increasing prevalence of cardiovascular diseases, advancements in medical imaging technology, and the rising adoption of AI algorithms in healthcare. In conclusion, the market for plaque vulnerability assessment in medical imaging is expanding rapidly due to the growing need for early detection and prevention of cardiovascular diseases. Technological advancements in medical imaging, coupled with the adoption of AI algorithms, have significantly improved the accuracy and efficiency of plaque vulnerability assessment. As healthcare professionals continue to prioritize preventive care, the demand for non-invasive and accurate diagnostic tools for plaque vulnerability assessment is expected to drive the market’s growth in the coming years.

The Siemens Healthcare GmbH invention works as follows

A machine-implemented classification system uses medical imaging data and other information to predict the rupture of plaques, rather than relying on variations from one physician to another and limited imaging information. To classify plaque, anatomical, morphological and hemodynamic features, as well as biochemical ones, are combined.

Background for Plaque Vulnerability Assessment in Medical Imaging

The present embodiments are related to the assessment of plaque vulnerability. Plaque analysis has as a main goal to determine and characterize a plaque’s vulnerability to rupture. Plaques in critical vessels such as cerebral arteries and coronaries can rupture, causing a stroke or heart attack.

Multiple medical imaging modalities have been used for the analysis of plaques in blood vessels, including computed tomography(CT), x ray angiography(XA), optical-coherent tomography(OCT), ultra-sound (US), intravascular ultrasound (IVUS), near-infrared spectroscopy and (NIRS). The ability to predict rupture of plaque may vary. The human prediction from medical imaging can be subjective and vary depending on who is predicting.

By way or introduction, the preferred embodiments are described below. They include methods, systems and instructions as well as non-transitory computer-readable media that can be used to assess plaque vulnerability in a patient’s medical imaging. Instead of relying on variations from doctor to physician or limited imaging information to predict rupture, a machine-implemented classification uses medical imaging and other data to predict plaque rupture. To classify plaque, two or more anatomical features, morphological features, hemodynamic and biochemical characteristics are combined.

In a first aspect, the method for assessing plaque vulnerability of a patient is described. A medical imaging scanner extracts anatomical features of a plaque or vessel and morphological features of the plaque from a scan. The scan is used to extract a patient’s hemodynamic characteristic from a sensor that measures the blood flow or based on individualized vessel models. An interface receives a biochemical characteristic from a patient’s blood test. A machine-implemented classification system calculates the risk score of plaque rupture for a patient based on anatomical and morphological features, as well as hemodynamics and biochemicals. The risk score is displayed on a display.

In a second aspect, the system provides a method for assessing the plaque vulnerability of patients in medical imaging. A medical scanner can be configured to scan the vessel and plaques of a patient. A memory can be configured to store up to three first features of a blood test. Image processors are configured to extract from scan data one or more second features about the vessel, plaque or vessel and plaque and determine the risk of rupture based on the first features and the second features. “An output is configured to output for the patient the risk of plaque rupture.

In a third aspect, the method for assessing plaque vulnerabilities of a patient is described. As input features, a morphological feature of a plaque on a patient’s body, a hemodynamic character, an anatomy of a vessel with a plaque, and biochemical characteristics of the plaque are loaded. A machine-trained classification classifier responds to the input feature vector by classifying the plaque of the patient. The user is informed of the result.

The following claims define the invention. Nothing in this section should be construed as limiting those claims. Additional aspects and benefits of the invention will be discussed below, in conjunction with the preferred embodiments. They may be claimed later independently or together.

There is a lot of research being done to determine the physiological mechanisms that lead to plaque rupture. There are several factors that influence the likelihood of a plaque rupture. These include morphological features of the plate, hemodynamic properties (e.g. wall shear stress), the anatomy of the parent vessels, and biochemical qualities of the plaque.

A machine-implemented assessment of plaque vulnerability is available to assist clinicians in understanding the risk for plaque rupture, and thereby identify stable and susceptible plaques for a patient. The machine-based assessment can be used as a second opinion, or to inform the physician about diagnosis and/or treatment planning. A machine-implemented classifier provides more objective risk information than one that relies on variables such as training, skill level, work environment, etc. Since a machine-implementation is used, the plaque assessment in medical imaging may be improved. The machine-implementation may better or more consistently combine different types of information linked to plaque vulnerability.

The assessment is primarily on medical imaging. It may also include information from blood tests or computational modeling. Instead of relying solely on imaging, different types of information will be gathered to classify. The machine-implementation may more efficiently and rapidly gather and provide the assessment as compared to manual performance.

In one embodiment, medical images are taken. Medical images are used to extract anatomical and morphological characteristics of the plaque, and related vessels (e.g. parent vessel and/or branching vessel). To extract hemodynamic characteristics of the plaque, a hemodynamic computation is carried out. To find hemodynamic features, invasive or non-invasive pressure and/or flow measurements are used. A computational fluid dynamics analysis (CFD) based on a personalized model derived from medical imaging can be used as an alternative or in addition. The biochemical features of plaques are extracted from blood-based markers provided by a test. The classifier, which is machine-implemented, determines the risk score for rupture of plaque based on anatomical features, morphological features, hemodynamics, and biochemical characteristics. The risk-score indicates whether a plaque in a patient is stable or vulnerable.

FIG. The first figure shows the overall approach. The analysis of heterogeneous sources is combined. An image processor detects the morphology of plaques, such as lipid or fibrous. Computational fluid dynamics is used to calculate the hemodynamic stress, or any other force. For computation fluid dynamics, the model is customized to the anatomy and patient. This can be done by using medical imaging data. Clinical tests, such a blood test, can be used to obtain biochemical markers such as Troponin, creactive protein, or another feature. The features can be different or more, and they may also be linked to or correlated to the risk of rupture. They may also be used in order determine another feature that is linked or correlative with this risk. A machine calculates the risk-score from these features (e.g. plaque vulnerability or another plaque-related risk). Combining data from multiple sources improves the accuracy of risk calculations, which helps in diagnosis.

FIG. The flow chart diagram in FIG. 2 illustrates one embodiment of the method of assessing plaque vulnerabilities of patients using medical imaging. The information from three or four sources, or three or four different types of features is loaded and then used to classify risk of plaque rupture. Risk is calculated using different types of data, including anatomical and morphological information, biochemical and/or hemodynamic information. This may result in a more accurate risk. The information from any feature that is linked or correlated to the risk can be combined with a machine in order to determine the risks efficiently and/or consistently.

The method is implemented using the system shown in FIG. The method can be implemented by the system shown in Figure 4 or any other system. The method can be implemented, for example, on an image processing system associated with a magnetic resonance (MR), computed-tomography (CT), ultrasonic, emission, x ray (e.g. angiography), and other imaging systems. Another example is the implementation of the method on a Picture Archiving and Communications System (PACS) server or workstation. In some embodiments, the methods are implemented on a computer network with different nodes or server performing different functions. An imaging system or a PACS system is used to acquire the medical data. The output can be seen on a screen or via a network.

The acts are performed according to the order indicated (e.g. from top to bottom). Acts 40, 42 and 44 are performed simultaneously or in any order.

Another, different or fewer acts can be provided. In act 52, for example, the method can be performed without transmitting any risk. Another example is that act 42, 44 and/or 46 were not performed.

In act 40, the medical scanner or imager scans a patient. The medical image is created. The medical image consists of a frame containing data that represents the patient. Data can be in any form. The terms “image” and “imaging” are not synonymous. While the terms?image? The image or imaging data can be displayed in a different format than the actual image. The medical image could be, for example, a collection of scalars representing different locations using a Cartesian coordinate format or polar format that is different from the display format. Another example is a set of red, green and blue (e.g. RGB) values that are output to an display in order to generate the image. The medical image can be a display format or another image that has been displayed in the past or is currently being displayed. The image is a dataset which can be used to create images, for example scan data that represents the patient.

Any type medical image can be used.” In one embodiment, a CT image obtained with a CT scanner is used as the medical image. A CT dataset can be used to detect vessels, for example. Another example is the acquisition of magnetic resonance (MR), or MR data, representing a patient using an MR imaging device. The MR data are acquired by using an imaging sequence to scan a patient. The data represents an interior area of the patient. The magnetic resonance data for MR is k space data. The data is reconstructed using Fourier analysis from the k space into a three dimensional object or image. The raw data for CT is then reconstructed to a three-dimensional image. “Other medical imaging modalities that can be used to acquire scan data are X-ray angiogram (XA), optical-coherence-tomography (OCT), ultrasonic (US), intravascular ultrasound (IVUS) and near-infrared spectroscopy

The medical image data, or scan data, represents the tissue of the patient. The medical image can also represent fluids, flow, or velocity within the patient. In some embodiments, data can represent both structure and flow.

The medical data represent a region in one, two or three dimensions of the patient.” The medical data can represent a slice or an area of the patient. The values are given for multiple locations that are distributed in two- or three-dimensional space. Medical data are acquired in the form of a data frame. The frame of data is the scan area at a certain time or period. The dataset can represent an area or volume in time. For example, it could provide a 4D image of the patient.

The medical image or dataset was acquired through a patient scan. The acquisition is part of the scanning process. The acquisition can also be from memory or storage, for example, acquiring an existing dataset from a PACS.

Click here to view the patent on Google Patents.