Abstract: Embodiments of the disclosure provide systems and methods for analyzing medical images using a learning model. The system receives a medical image acquired by an image acquisition device. The system may additionally include at least one processor configured to apply the learning model to perform an image analysis task on the medical image. The learning model is trained jointly with an error estimator using training images comprising a first set of labeled images and a second set of unlabeled images. The error estimator is configured to estimate an error of the learning model associated with performing the image analysis task.

Abstract: Embodiments of the disclosure provide systems and methods for medical image analysis. A method may include receiving a medical image acquired of a subject by an image acquisition device. The method may also include applying a calcium detection model to detect at least one calcium region relevant in determining a calcium score from the medical image. The method may further include applying a score regression learning model to the at least one calcium region to determine a calcium score for the medical image. The method may additionally include providing the determined calcium score of the medical image for a diagnosis of the subject.

Abstract: The present disclosure provides a method, a device, and a non-transitory computer-readable storage medium for detecting a medical condition of an organ. The method includes obtaining 2D image sequences of the organ in a plurality of different directions and applying a plurality of classification branches to the 2D image sequences. Each classification branch receives a 2D image sequence of one direction and provides a classification result with respect to that direction. Each classification branch includes a convolutional neural network configured to extract first image features from the corresponding 2D image sequence and a recurrent neural network configured to extract second image features from the first image features. The method further includes fusing the classification results provided by the plurality of classification branches for detecting the medical condition.

Abstract: Embodiments of the disclosure provide methods and systems for joint abnormality detection and physiological condition estimation from a medical image. The exemplary method may include receiving, by at least one processor, the medical image acquired by an image acquisition device. The medical image includes an anatomical structure. The method may further include applying, by the at least one processor, a joint learning model to determine an abnormality condition and a physiological parameter of the anatomical structure jointly based on the medical image. The joint learning model satisfies a predetermined constraint relationship between the abnormality condition and the physiological parameter.

Abstract: The disclosure relates to a system and method for predicting physiological-related parameters based on a medical image. The method includes receiving a medical image acquired by an image acquisition device and predicting a sequence of physiological-related parameters at a sequence of positions and simultaneously estimating an uncertainty level of the predicted sequence of physiological parameters from the medical image by using a sequential learning model. The sequential learning model is trained to minimize a loss function associated with the uncertainty level. The method not only provides predictions but also the corresponding uncertainty estimations by using sequential learning model(s), thus improving the transparency and explainability of the sequential learning model.

Abstract: Embodiments of the disclosure provide systems and methods for analyzing a medical image containing a vessel structure using a sequential model. An exemplary system includes a communication interface configured to receive the medical image and the sequential model. The sequential model includes a vessel extraction sub-model and a lesion analysis sub-model. The vessel extraction sub-model and the lesion analysis sub-model are independently or jointly trained. The exemplary system also includes at least one processor configured to apply the vessel extraction sub-model on the received medical image to extract location information of the vessel structure. The at least one processor also applies the lesion analysis sub-model on the received medical image and the location information extracted by the vessel extraction sub-model to obtain a lesion analysis result of the vessel structure. The at least one processor further outputs the lesion analysis result of the vessel structure.

Abstract: The present disclosure relates to training methods for a machine learning model for physiological analysis. The training method may include receiving training data including a first dataset of labeled data of a physiological-related parameter and a second dataset of weakly-labeled data of the physiological-related parameter. The training method further includes training, by at least one processor, an initial machine learning model using the first dataset, and applying, by the at least one processor, the initial machine learning model to the second dataset to generate a third dataset of pseudo-labeled data of the physiological-related parameter. The training method also includes training, by the at least one processor, the machine learning model based on the first dataset and the third dataset, and providing the trained machine learning model for predicting the physiological-related parameter.

Abstract: Methods and Systems for generating a centerline for an object in an image and computer readable medium are provided. The method includes receiving an image containing the object. The method also includes generating the centerline of the object, by a processor, using a reinforcement learning network configured to predict movement of a virtual agent that traces the centerline in the image. The reinforcement learning network is further configured to perform at least one auxiliary task that detects a bifurcation in a trajectory of the object. The reinforcement learning network is trained by maximizing a cumulative reward and minimizing an auxiliary loss of the at least one auxiliary task. Additionally, the method includes displaying the centerline of the object.

Abstract: Embodiments of the disclosure provide methods and systems for disease condition prediction from images of a patient. The system may include a communication interface configured to receive a sequence of images acquired of the patient by an image acquisition device. The sequence of images are acquired at a sequence of prior time points during progression of a disease. The system may include a processor, configured to determine regions of interest based on the sequence of images. The processor applies a progressive condition prediction network to the regions of interest to predict a level of disease progression at a future time point during the progression of the disease. The progressive condition prediction network predicts the level of disease progression based on the regions of interest and disease conditions at the sequence of prior time points. The processor further provides a diagnostic output based on the predicted level of disease progression.

Abstract: Embodiments of the disclosure provide methods and systems for determining a disease condition from a 3D image of a patient. The exemplary system may include a communication interface configured to receive the 3D image acquired of the patient by an image acquisition device. The system may further include a processor, configured to determine a 3D region of interest from the 3D image and apply a detection network to the 3D region of interest to determine the disease condition and a severity of the disease condition. The detection network is a multi-task learning network that determines the disease condition based on one or more lesion masks determined from the 3D region of interest and determines the severity of the disease condition from the 3D region of interest. The processor is further configured to provide a diagnostic output based on the disease condition and the severity of the disease condition.

Abstract: The present disclosure is directed to a method and system for automatically detecting a physiological condition from a medical image of a patient. The method may include receiving the medical image acquired by an imaging device. The method may further include detecting, by a processor, target objects and obtaining the corresponding target object patches from the received medical image. And the method may further include determining, by the processor, a first parameter using a first learning network for each target object patch. The first parameter represents the physiological condition level of the corresponding target object, and the first learning network is trained by adding one or more auxiliary classification layers. This method can quickly, accurately, and automatically predict target object level and/or image (patient) level physiological condition from a medical image of a patient by means of a learning network, such as 3D learning network.

Abstract: Methods and Systems for generating a centerline for an object in an image and computer readable medium are provided. The method includes receiving an image containing the object. The method also includes generating the centerline of the object by tracing a sequence of patches with a virtual agent. For each patch other than the initial patch, the method determines a current patch based on the position and action of the virtual agent at a previous patch. The method further determines a policy function and a value function based on the current patch using a trained learning network, which includes an encoder followed by a first learning network and a second learning network. The learning network is trained by maximizing a cumulative reward. The method also determines the action of the virtual agent at the current patch. Additionally, the method displays the centerline of the object.

Abstract: The disclosure relates to a method and device for predicting a physical parameter based on input physical information, and medium. The method may include predicting, by a processor, an intermediate variable based on the input physical information with an intermediate sub-model, which incorporates a constraint on the intermediate variable according to prior information of the physical parameter. The method may also include transforming, by the processor, the intermediate variable predicted by the intermediate sub-model to the physical parameter with a transformation sub-model.

Abstract: Embodiments of the disclosure provide methods and systems for determining a disease condition from a 3D image of a patient. The exemplary system may include a communication interface configured to receive the 3D image acquired of the patient by an image acquisition device. The system may further include at least one processor, configured to determine a 3D region of interest from the 3D image and apply a detection network to the 3D region of interest to determine the disease condition and a severity of the disease condition. The detection network is a multi-task learning network that includes a first task branch and a second task branch, and the first task branch determines the disease condition and the second task branch determines the severity of the disease condition in a single forward pass. The at least one processor is further configured to provide a diagnostic output based on the determined the disease condition and severity of the disease condition.

Abstract: Embodiments of the disclosure provide methods and systems for predicting a disease condition from images of a patient. The exemplary system may include a communication interface configured to receive a sequence of images acquired of the patient by an image acquisition device. The sequence of images are acquired at a sequence of prior time points during progression of a disease. The system may further include at least one processor, configured to determine regions of interest corresponding to the sequence of prior time points based on the sequence of images. The at least one processor also applies a progressive condition prediction network to the regions of interest to predict a disease condition at a future time point during the progression of the disease. The progressive condition prediction network includes a forward path for predicting the disease condition based on the regions of interest and disease conditions at the sequence of prior time points.

Abstract: The disclosure provides a method, device and a computer-readable medium for performing three-dimensional blood vessel reconstruction. The computer-implemented method includes receiving a first two-dimensional image of a blood vessel of a patient, where the first two-dimensional image is a projection image acquired in a first projection direction. The method further includes reconstructing, by a processor, a three-dimensional model of the blood vessel based on at least the first two-dimensional image. The method additional includes adjusting the three-dimensional model of the blood vessel, based on a comparison of a first optical path length determined from a second two-dimensional image of the blood vessel of the patient and a second optical path length determined from the three-dimensional model.

Abstract: The disclosure relates to systems and methods for vessel image analysis. The method includes receiving a set of images along a vessel acquired by a medical imaging device, and determining a sequence of centerline points along the vessel and a sequence of image patches at the respective centerline points based on the set of images. The method further includes detecting plaques based on the sequence of image patches using a first learning network. The first learning network includes an encoder configured to extract feature maps based on the sequence of image patches and a plaque range generator configured to generate a start position and an end position of each plaque based on the extracted feature maps. The method also includes classifying each detected plaque and determining a stenosis degree for the detected plaque, using a second learning network reusing at least part of the parameters of the first learning network and the extracted feature maps.

Abstract: The disclosure provides a method and device for performing three-dimensional blood vessel reconstruction using projection images of a patient. The computer-implemented method includes receiving a first two-dimensional image of a blood vessel in a first projection direction and a three-dimensional model of the blood vessel. The method further includes determining, by a processor, a first optical path length at a selected position of the blood vessel based on the first two-dimensional image. The method also includes determining, by the processor, a second optical path length at the selected position of the blood vessel in the three-dimensional model. The method additional includes adjusting the three-dimensional model of the blood vessel, based on a comparison of the first optical path length and the second optical path length.

Abstract: Embodiments of the disclosure provide methods and systems for detecting COVID-19 from a lung image of a patient. The exemplary system may include a communication interface configured to receive the lung image acquired of the patient's lung by an image acquisition device. The system may further include at least one processor, configured to determine a region of interest comprising the lung from the lung image and apply a COVID-19 detection network to the region of interest to determine a condition of the lung. The COVID-19 detection network is a multi-class classification learning network that labels the region of interest as one of COVID-19, non-COVID-19 pneumonia, non-pneumonia abnormal, or normal. The at least one processor is further configured to provide a diagnostic output based on the determined condition of the lung.

Abstract: The disclosure relates to a computer-implemented method for analyzing an image sequence of a periodic physiological activity, a system, and a medium. The method includes receiving the image sequence from an imaging device, and the image sequence has a plurality of images. The method further includes identifying at least one feature portion in a selected image, which moves responsive to the periodic physiological activity. The method also includes detecting, by a processor, the corresponding feature portions in other images of the image sequence and determining, by the processor, a phase of a the selected image in the image sequence based on the motion of the feature portion.