Abstract: Apparatuses, systems, and techniques for continuous training of an object detection/classification model are provided. A first image depicting an environment is identified. Data for an object detected in the first image is obtained based on outputs of a machine learning (ML) model. A determination is made of whether a level of confidence that the object corresponds to an object class satisfies a confidence criterion. If so, first conditions corresponding to the environment depicted in the first image are identified. Noise characteristics associated with a second image depicting the environment are determined based on a difference between the first conditions and second conditions of the second image. The first image is augmented based on the noise characteristics to generate a third image depicting the environment based on the second conditions and the detected object. Training data associated with the third image is provided to train the ML model.

Abstract: In various examples, systems and methods are disclosed relating to encoding and decoding high-resolution images on systems supporting limited resolutions. A system can identify an image to be encoded using an image encoding process. The system can extract a plurality of regions from the image and can generate a plurality of encoded regions by encoding each of the plurality of regions using the image encoding process. The system can generate a media package using the plurality of encoded regions and metadata corresponding to the plurality of regions.

Abstract: Various examples, systems, and methods are disclosed relating to frame selection via activity-based ranking and optimization. A first computing system can receive a plurality of frames and metadata from a capture device capturing a video stream. The first computing system can generate, using a ranking model, a plurality of rankings for the plurality of frames based on a plurality of video parameters of the plurality of frames and the metadata, wherein the plurality of rankings correspond to a summarization of the video stream. The first computing system can determine at least one of the plurality of frames to provide to at least one buffer based on the plurality of rankings, wherein the at least one buffer stores a subset of frames of the plurality of frames. The first computing system can provide, from the at least one buffer, the subset of frames as input to a machine-learning model.

Abstract: In various examples, a microservices-based architecture may be configured to detect, analyze, and/or reduce instances of PCB defects using AI-based models and other algorithms. For instance, image data representing an image of a PCB may be applied to a model running within a first service associated with an application. The model may process the image data to generate one or more predictions relating to defects associated with the PCB. In some instances, one or more second services associated with the application may use the predictions to generate visualizations and/or compute metrics corresponding to the defects associated with the PCB. Additionally, in some examples, the predictions, the visualizations, and/or the metrics may be used to update one or more parameters associated with one or more systems to, for instance, improve the performance of AOI systems, SMT systems, or any other systems.

Abstract: Disclosed are apparatuses, systems, and techniques for implementing bidirectional tracking in computer vision applications. In one embodiment, the techniques include obtaining digital representations of an object depicted in video frames and for each of a forward direction (FD) of tracking and a reverse direction (RD) of tracking, obtaining, using (i) a current state of the object associated with an upstream video frame and (ii) the digital representation of the object for a downstream video frame, an updated state of the object associated with the downstream video frame. The techniques further include obtaining, using the updated state of the object for the FD and/or the updated state of the object for the RD, a bidirectional state of the object, and determining, using the bidirectional state of the object, a trajectory of the object across the video frames.

Abstract: Apparatuses, systems, and techniques for continuous training of an object detection and/or classification model. A first image including a depiction of an environment based on a first set of conditions is identified. Object data associated with an object detected in the first image is obtained based on one or more outputs of a machine learning model. A determination is made of whether a level of confidence that an object corresponds to an object class satisfies a level of confidence criterion. If so, a first set of conditions corresponding to the environment depicted in the first image. One or more noise characteristics associated with a second image including a depiction of the environment is determined based on a difference between the first set of conditions and a second set of conditions of the second image. The first image is augmented based on the one or more determined noise characteristics to generate a third image.

Abstract: The present invention provides pharmacological compounds including an effector moiety conjugated to a binding moiety that directs the effector moiety to a biological target of interest. Likewise, the present invention provides compositions, kits, and methods (e.g., therapeutic, diagnostic, and imaging) including the compounds. The compounds can be described as a protein interacting binding moiety-drug conjugate (SDC-TRAP) compounds, which include a protein interacting binding moiety and an effector moiety. For example, in certain embodiments directed to treating cancer, the SDC-TRAP can include an Hsp90 inhibitor conjugated to a cytotoxic agent as the effector moiety.

Abstract: The present invention provides pharmacological compounds including an effector moiety conjugated to a binding moiety that directs the effector moiety to a biological target of interest. Likewise, the present invention provides compositions, kits, and methods (e.g., therapeutic, diagnostic, and imaging) including the compounds. The compounds can be described as a protein interacting binding moiety-drug conjugate (SDC-TRAP) compounds, which include a protein interacting binding moiety and an effector moiety. For example, in certain embodiments directed to treating cancer, the SDC-TRAP can include an Hsp90 inhibitor conjugated to a cytotoxic agent as the effector moiety.

Abstract: Apparatuses, systems, and techniques to balance processing load between a plurality of hardware accelerators. In at least one embodiment, operations performed on batches of frames of a video (e.g., as part of a video analytics pipeline) are distributed by a load balancer between a first hardware accelerator and a second hardware accelerator.

Abstract: For personalizing a vessel stent, images associated with a subject generated by various imaging modalities are aggregated. The images are then processed for identifying Regions of Interest (ROIs) and various parameters associated with the ROIs. Further, a model and a composition of the vessel stent to be administered to the subject to alleviate the condition in the vessel are computed. Thereafter, the model is verified for compatibility using information derived from patient stratification parameters. Upon successful verification of the model, a format of the model is generated that can be used directly for fabricating the vessel stent using additive manufacturing processes known in the art.

Abstract: The present invention provides pharmacological compounds including an effector moiety conjugated to a binding moiety that directs the effector moiety to a biological target of interest. Likewise, the present invention provides compositions, kits, and methods (e.g., therapeutic, diagnostic, and imaging) including the compounds. The compounds can be described as a protein interacting binding moiety-drug conjugate (SDC-TRAP) compounds, which include a protein interacting binding moiety and an effector moiety. For example, in certain embodiments directed to treating cancer, the SDC-TRAP can include an Hsp90 inhibitor conjugated to a cytotoxic agent as the effector moiety.

Abstract: The disclosure relates to method and device for determining at least a portion of an anatomy from a set of medical images. Thereafter, the image acquisition protocol associated with the medical images is determined automatically based on the determined anatomy. The image acquisition protocol is automatically determined by extracting one or more features associated with the medical images characterizing the acquisition protocol. Thereafter, the medical images are classified based in the features. Further, the medical images are assigned with an acquisition protocol label based on the classification of the medical images. The medical images are displayed with the respective protocol labels. Further, medical images acquired with an identified image acquisition protocol at different time instances are displayed for comparative analysis.

Abstract: In order to install a ground loop for a geothermal heating and/or cooling system in ground material, a drilling machine having a drill bit connected to an end of first tubing may be used to create an borehole in the ground material. After the borehole is created, grout may be pumped into the borehole through the first tubing as the first tubing is removed from the borehole. This allows the borehole to be grouted from the bottom towards the top. Thereafter, second tubing may be inserted into the grout in order to create the ground loop. This eliminates the need for a “tremie” pipe to insert the ground loop and pump grout into the borehole, while still allowing for the borehole to be grouted from the bottom towards the top to reduce the likelihood of voids.

Abstract: Perceptual visual quality of different image sets is automatically assessed. A computing device includes a processor and a memory coupled to the processor. The memory includes an image assessment module configured to obtain features from each of the image sets, and generate feature vectors corresponding to the obtained features. The image quality assessment module is configured to compute a set of quality scores corresponding to the feature vectors associated with each of the image sets. The image quality assessment module is also configured to automatically determine an image set having an optimal perceptual visual quality among the image sets using the corresponding set of quality scores.

Abstract: For personalizing a vessel stent, images associated with a subject generated by various imaging modalities are aggregated. The images are then processed for identifying Regions of Interest (ROIs) and various parameters associated with the ROIs. Further, a model and a composition of the vessel stent to be administered to the subject to alleviate the condition in the vessel are computed. Thereafter, the model is verified for compatibility using information derived from patient stratification parameters. Upon successful verification of the model, a format of the model is generated that can be used directly for fabricating the vessel stent using additive manufacturing processes known in the art.

Abstract: The disclosure relates to method and device for determining at least a portion of an anatomy from a set of medical images. Thereafter, the image acquisition protocol associated with the medical images is determined automatically based on the determined anatomy. The image acquisition protocol is automatically determined by extracting one or more features associated with the medical images characterizing the acquisition protocol. Thereafter, the medical images are classified based in the features. Further, the medical images are assigned with an acquisition protocol label based on the classification of the medical images. The medical images are displayed with the respective protocol labels. Further, medical images acquired with an identified image acquisition protocol at different time instances are displayed for comparative analysis.

Abstract: The embodiments relate to a method and device for determining plaque deposits in coronary arteries. According to the method, a coronary angiography radiography image of a subject is obtained using an imaging modality. The obtained radiography image is calibrated for further analysis of the radiography image. Image processing techniques such as thresholding and histogram equalization are applied on the radiography image to extract vessel tree structure for analysis. Further, an inner lumen width and an outer vessel width are computed based on the processed radiography image. Further, a level of plaque deposition is determined based on the inner lumen and outer vessel dimensions. Further, the result based on the analysis is displayed to the user. The result includes the level of plaque deposits within the lumen on the vascular structure and possible risks posed by the plaque deposits to a subject.

Abstract: The embodiments relate to a method and device for determining plaque deposits in coronary arteries. According to the method, a coronary angiography radiography image of a subject is obtained using an imaging modality. The obtained radiography image is calibrated for further analysis of the radiography image. Image processing techniques such as thresholding and histogram equalization are applied on the radiography image to extract vessel tree structure for analysis. Further, an inner lumen width and an outer vessel width are computed based on the processed radiography image. Further, a level of plaque deposition is determined based on the inner lumen and outer vessel dimensions. Further, the result based on the analysis is displayed to the user. The result includes the level of plaque deposits within the lumen on the vascular structure and possible risks posed by the plaque deposits to a subject.

Abstract: A system for processing electrocardiograph signals includes an interface and a processor. The interface provides a first electrocardiograph signal and a second electrocardiograph signal from an electrocardiograph to the processor. The processor receives the electrocardiograph signals, processes the first electrocardiograph signal to identify a feature of the signal, and processes the second electrocardiograph signal to identify energy over time (e.g., at various time instances), and identifies a first time instance of interest based on the feature.

Abstract: Perceptual visual quality of different image sets is automatically assessed. A computing device includes a processor and a memory coupled to the processor. The memory includes an image assessment module configured to obtain features from each of the image sets, and generate feature vectors corresponding to the obtained features. The image quality assessment module is configured to compute a set of quality scores corresponding to the feature vectors associated with each of the image sets. The image quality assessment module is also configured to automatically determine an image set having an optimal perceptual visual quality among the image sets using the corresponding set of quality scores.