The present disclosure relates to systems and methods for processing real-time video and detecting objects in the video. In one implementation, a system is provided that includes an input port for receiving real-time video obtained from a medical image device, a first bus for transferring the received real-time video, and at least one processor configured to receive the real-time video from the first bus, perform object detection by applying a trained neural network on frames of the received real-time video, and overlay a border indicating a location of at least one detected object in the frames. The system also includes a second bus for receiving the video with the overlaid border, an output port for outputting the video with the overlaid border from the second bus to an external display, and a third bus for directly transmitting the received real-time video to the output port.

In one embodiment, a diagnosis support apparatus includes: an input circuit configured to acquire a first medical image; and processing circuitry configured to generate a second medical image from the first medical image in such a manner that information included in the second medical image is reduced from information included in the first medical image, extract auxiliary information from the first medical image, and perform inference of a disease by using the second medical image and the auxiliary information.

A framework for medical data analysis, comprising a tool generation unit configured for automatically generating a first number of data analysis tools based on first medical image data and first analysis data related to the first medical image data.

A method of reading a medical image by a computing device operated by at least one processor is provided. The method includes obtaining an abnormality score of the input image using an abnormality prediction model, filtering the input image so as not to be subsequently analyzed when the abnormality score is less than or equal to a cut-off score based on the cut-off score which makes a specific reading sensitivity; and obtaining an analysis result of the input image using a classification model that distinguishes the input image into classification classes when the abnormality score is greater than the cut-off score.

System and method are disclosed for imaging acquisition from sparse partial scans of distributed wide angle. During real time image reconstruction, artificial intelligence (AI) determines if there is enough information to perform diagnostics based on initial scans. If there is enough information from the fractional scans, then data acquisition stops; if more information is needed, then system performs another round of wide-angle sparse scans in a new location progressively until a result is satisfactory. The system reduces X-ray dose on a patient and performs quicker X-ray scan at multiple pulsed source-in-motion tomosynthesis imaging system. The method and system also significantly reduce the amount of time required to display high quality three-dimensional tomosynthesis images.

Provided are computer-implemented methods and corresponding systems for providing a candidate medical finding based on the analysis of a medical image. The methods and systems are based on obtaining the medical image depicting a body part of a patient, generating a first set of candidate medical findings by subjecting the medical image to a first medical findings detection process, generating a second set of candidate medical findings by subjecting the medical image to a second medical findings detection process different than the first medical finding detection process, obtaining a region of interest in the medical image, identifying, in the region of interest, at least one candidate medical finding comprised in the second set of candidate medical findings and not comprised in the first set of candidate medical findings, and providing the at least one candidate medical finding.

Described herein are means for training a deep model to learn contrastive representations embedded within part-whole semantics via a self-supervised learning framework, in which the trained deep models are then utilized for the processing of medical imaging. For instance, an exemplary system is specifically configured for performing a random cropping operation to crop a 3D cube from each of a plurality of medical images received at the system as input; performing a resize operation of the cropped 3D cubes; performing an image reconstruction operation of the resized and cropped 3D cubes to predict the resized whole image represented by the original medical images received; and generating a reconstructed image which is analyzed for reconstruction loss against the original image representing a known ground truth image to the reconstruction loss function. Other related embodiments are disclosed.

A computer (110)-implemented method for reading an imaging scan (410) includes accessing the imaging scan (410). The imaging scan (410) includes a stack of radiological images. The method also includes generating a plurality of two-dimensional images from cross-sectional data of the imaging scan (410). The plurality of two-dimensional images include projected information from the stack of radiological images. The projected information includes either a full imaged volume or an automatically-selected sub-volume and either a full range of image intensities or an automatically-selected sub-range of image intensities. The method further includes displaying the generated plurality of two-dimensional images or a subset thereof in a user interface (UI) of an advanced interpretation environment (380). The user interface provides access to the stack of radiological images or additional information derived from the stack of radiological images, by enabling interaction with the generated plurality of two-dimensional images.

An improved method of performing object segmentation and classification that reduces the memory required to perform these tasks, while increasing predictive accuracy. The improved method utilizes a capsule network with dynamic routing. Capsule networks allow for the preservation of information about the input by replacing max-pooling layers with convolutional strides and dynamic routing, allowing for the reconstruction of an input image from output capsule vectors. The present invention expands the use of capsule networks to the task of object segmentation and medical image-based cancer diagnosis for the first time in the literature; extends the idea of convolutional capsules with locally-connected routing and propose the concept of deconvolutional capsules; extends the masked reconstruction to reconstruct the positive input class; and proposes a capsule-based pooling operation for diagnosis. The convolutional-deconvolutional capsule network shows strong results for the tasks of object segmentation and classification with substantial decrease in parameter space.

Embodiments classify a region of tissue demonstrating non-small cell lung cancer using quantified vessel tortuosity (QVT). One example apparatus includes annotation circuitry configured to segment a lung region from surrounding anatomy in the region of tissue represented in a radiological image and segment a nodule from the lung region by defining a nodule boundary; vascular segmentation circuitry configured to generate a three dimensional (3D) segmented vasculature by segmenting a vessel associated with the nodule, and to identify a center line of the 3D segmented vasculature; QVT feature extraction circuitry configured to extract a set of QVT features from the radiological image; and classification circuitry configured to compute a probability that the region of tissue will respond to immunotherapy and generate a classification that the region of tissue is a responder or a non-responder based, at least in part, on the probability.