Abstract: Systems and methods for advanced pulmonary nodule analysis with malignancy-weighted detection and customized visualization are provided. One or more medical images of a patient are received. One or more candidate nodules are detected in the one or more medical images. Each of the one or more candidate nodules are associated with a nodule detection score. A malignancy score is determined for each of the one or more candidate nodules. The nodule detection score is weighted for each of the one or more candidate nodules based on its malignancy score. The weighted nodule detection scores are output.

Abstract: Systems and methods for calculating a distance between a first point and a second point of an anatomical landmark are provided. An input medical image of an anatomical landmark of a patient is received. One or more probability maps predicting a first point and a second point of the anatomical landmark in the input medical image are generated using a machine learning based model. Locations of the first point and the second point in the input medical image are determined based on the one or more probability maps. A distance between the first point and the second point is calculated based on the locations. The locations of the first point and the second point in the input medical image and/or the calculated distance between the first point and the second point are output.

Abstract: A method for detecting a motion artifact in medical imaging data is provided. The method comprises generating a first image representing at least an overlap region based on a first acquired dataset and generating a second image representing at least the overlap region based on a second acquired dataset; generating a difference map for the overlap region by subtracting the first image and the second image from each other; and detecting the motion artifact by determining a connected region in the difference map, wherein absolute values of the difference map are equal to or greater than a predefined threshold value within the connected region.

Abstract: Systems and methods for quantification of body composition using contrastive learning in computed tomography (CT) data. A segmentation model is provided that is trained using two stages. An encoder of the segmentation model is pretrained using unlabeled data. The encoder is extended by a decoder which is further trained using labeled data.

Abstract: One or more example embodiments relates to a computer-implemented method for dose control with an automatic exposure control. The method comprises receiving a first time-course, receiving a second time-course, and receiving a cut-off dose. The method comprises determining a first weight, determining a second weight, multiplying a latest value of the first time-course with the first weight, wherein a first weighted accumulated dose is determined and multiplying a latest value of the second time-course with the second weight, wherein a second weighted accumulated dose is determined. The method comprises adding the first and the second weighted accumulated dose, wherein a weighted overall applied dose is determined, comparing the weighted overall applied dose with the cut-off dose, and providing a result of the comparison.

Abstract: Systems and methods for performing one or more medical imaging analysis tasks on PCCT (photon-counting computed tomography) images are provided. Image acquisition parameters of a PCCT image acquisition device are determined for acquiring PCCT images. One or more PCCT images of an anatomical object of a patient acquired using the PCCT image acquisition device configured with the image acquisition parameters are received. One or more medical imaging analysis tasks analyzing the anatomical object are performed based on the one or more PCCT images using one or more machine learning based models. Results of the one or more medical imaging analysis tasks are output.

Abstract: One or more example embodiments relates to a computer-implemented method for dose control with an automatic exposure control. The method comprises receiving a first time-course, receiving a second time-course, and receiving a cut-off dose. The method comprises determining a first weight, determining a second weight, multiplying a latest value of the first time-course with the first weight, wherein a first weighted accumulated dose is determined and multiplying a latest value of the second time-course with the second weight, wherein a second weighted accumulated dose is determined. The method comprises adding the first and the second weighted accumulated dose, wherein a weighted overall applied dose is determined, comparing the weighted overall applied dose with the cut-off dose, and providing a result of the comparison.

Abstract: Techniques of determining a quantification of at least one characteristic of a muscle structure comprising at least one muscle and at least one tendon are disclosed. The quantification of the at least one characteristic of the rotator cuff may be determined by using at least one artificial neural network and based on one or more medical images depicting the muscle structure of a patient.

Abstract: Systems and methods for calculating a distance between a first point and a second point of an anatomical landmark are provided. An input medical image of an anatomical landmark of a patient is received. One or more probability maps predicting a first point and a second point of the anatomical landmark in the input medical image are generated using a machine learning based model. Locations of the first point and the second point in the input medical image are determined based on the one or more probability maps. A distance between the first point and the second point is calculated based on the locations. The locations of the first point and the second point in the input medical image and/or the calculated distance between the first point and the second point are output.

Abstract: Systems and methods for an intuitive display of one or more anatomical objects are provided. One or more 3D medical images of one or more anatomical objects of a patient are received. Correspondences between the one or more 3D medical images and points on a 2D map representing the one or more anatomical objects are determined. The 2D map is updated with patient information extracted from the one or more 3D medical images. The updated 2D map with the determined correspondences is output.

Abstract: Systems and methods for quantification of body composition using contrastive learning in computed tomography (CT) data. A segmentation model is provided that is trained using two stages. An encoder of the segmentation model is pretrained using unlabeled data. The encoder is extended by a decoder which is further trained using labeled data.

Abstract: Systems and methods that normalize medical imaging data between different species and breeds of animals in order to allow for cross-species and crossbreed usage of machine trained models. Image data of a non-human subject is acquired, registered using a standardized model, and segmented using a machine trained model.

Abstract: Techniques for determining at least one characteristic of adipose tissue included in an anatomical structure are provided. The at least one characteristic of the adipose tissue is determined based on one or more segmented CT images using a trained neural network. For example, the at least one characteristic of the adipose tissue may be determined by inputting the one or more segmented CT images into the trained neural network. The one or more segmented CT images are obtained by segmenting each one of one or more CT images depicting the anatomical structure including the adipose tissue to determine a contour of the adipose tissue.

Abstract: A computer-implemented method is provided for classifying a malignancy risk of a kidney, in particular a human kidney. Imaging data of an anatomy of a subject patient at least partially includes a representation of a kidney of the subject patient. A first neural network segments at least one region of the kidney representation based on the imaging data. A second neural network detects one or more suspected lesions of the segmented kidney representation. A third neural network classifies the detected suspected lesion with a malignancy risk. The third neural network is a deep profiler.

Abstract: Techniques of determining a quantification of at least one characteristic of a muscle structure comprising at least one muscle and at least one tendon are disclosed. The quantification of the at least one characteristic of the rotator cuff may be determined by using at least one artificial neural network and based on one or more medical images depicting the muscle structure of a patient.

Abstract: Systems and methods for an intuitive display of one or more anatomical objects are provided. One or more 3D medical images of one or more anatomical objects of a patient are received. Correspondences between the one or more 3D medical images and points on a 2D map representing the one or more anatomical objects are determined. The 2D map is updated with patient information extracted from the one or more 3D medical images. The updated 2D map with the determined correspondences is output.

Abstract: A method for controlling an X-ray projection imaging device includes applying preferences for a serial radiography image acquisition to the X-ray projection imaging device. The method further includes executing the serial radiography image acquisition with the X-ray projection imaging device; recording data frames at different times during the serial radiography image acquisition; applying a trigger signal to the X-ray projection imaging device during the serial radiography image acquisition; and generating a snapshot image from a subset of the data frames, wherein the subset is chosen from the data frames recorded based on the trigger signal. A related system, a related control unit and a related X-ray projection imaging system are also disclosed.

Abstract: System and methods are provided for localizing a target object in a medical image. The medical image is discretized into a plurality of images having different resolutions. For each respective image of the plurality of images, starting from a first image and progressing to a last image with the progression increasing in resolution, a sequence of actions is performed for modifying parameters of a target object in the respective image. The parameters of the target object comprise nonlinear parameters of the target object. The sequence of actions is determined by an artificial intelligence agent trained for a resolution of the respective image to optimize a reward function. The target object is localized in the medical image based on the modified parameters of the target object in the last image.

Abstract: Systems and methods are provided for generating a visualization of a lung fissure. Medical imaging data is processed to identify a lung mesh and fissures data. The lung mesh is augmented with the fissures data and projected onto a straight plane for rendering into a concise two-dimensional image in which completeness of the lung fissure may be detected.

Abstract: For patient positioning for scanning, a current pose of a patient is compared to a desired pose. The desired pose may be based on a protocol or a pose of the same patient in a previous examination. Any differences in pose, such as arm position, leg position, head orientation, and/or torso orientation (e.g., laying on side, back, or stomach), are communicated. By changing the current pose of the patient to be more similar to the desired pose, a more consistent and/or registerable dataset may be acquired by scanning the patient.