Abstract: A method for rejecting biased data using a machine learning model includes receiving a cluster training data set including a known unbiased population of data and training a clustering model to segment the received cluster training data set into clusters based on data characteristics of the known unbiased population of data. Each cluster of the cluster training data set includes a cluster weight. The method also includes receiving a training data set for a machine learning model and generating training data set weights corresponding to the training data set for the machine learning model based on the clustering model. The method also includes adjusting each training data set weight of the training data set weights to match a respective cluster weight and providing the adjusted training data set to the machine learning model as an unbiased training data set.

Abstract: Implementations relate to providing, in response to a query, machine learning model output that is based on output from a trained machine learning model. The machine learning model output can include a predicted answer to the query, that is predicted based on the trained machine learning model. The machine learning model output can additionally or alternatively include an interactive interface for the trained machine learning model. Some implementations relate to generating a trained machine learning model “on the fly” based on a search query. Some implementations additionally or alternatively relate to storing, in a search index, an association of a machine learning model with a plurality of content items from resource(s) on which the machine learning model was trained.

Abstract: Implementations relate to providing, in response to a query, machine learning model output that is based on output from a trained machine learning model. The machine learning model output can include a predicted answer to the query, that is predicted based on the trained machine learning model. The machine learning model output can additionally or alternatively include an interactive interface for the trained machine learning model. Some implementations relate to generating a trained machine learning model “on the fly” based on a search query. Some implementations additionally or alternatively relate to storing, in a search index, an association of a machine learning model with a plurality of content items from resource(s) on which the machine learning model was trained.

Abstract: A method for rejecting biased data includes receiving a bias training data set based on a probability distribution of bias-sensitive variables of a target population and segmenting the bias training data set into clusters based on at least one respective bias-sensitive variable of the target population, each cluster including a bias cluster weight. The method also includes receiving a training data set for a machine learning model and segmenting the training data set into training clusters. Each training cluster is associated with at least one corresponding bias-sensitive variable of the target population and includes a corresponding training data set weight. The method also includes adjusting each training data set weight to match a respective bias cluster weight to form an adjusted training data set and providing the adjusted training data set to the machine learning model as an unbiased training data set.

Abstract: A method for rejecting biased data using a machine learning model includes receiving a cluster training data set including a known unbiased population of data and training a clustering model to segment the received cluster training data set into clusters based on data characteristics of the known unbiased population of data. Each cluster of the cluster training data set includes a cluster weight. The method also includes receiving a training data set for a machine learning model and generating training data set weights corresponding to the training data set for the machine learning model based on the clustering model. The method also includes adjusting each training data set weight of the training data set weights to match a respective cluster weight and providing the adjusted training data set to the machine learning model as an unbiased training data set.

Abstract: A method for rejecting biased data using a machine learning model includes receiving a cluster training data set including a known unbiased population of data and training a clustering model to segment the received cluster training data set into clusters based on data characteristics of the known unbiased population of data. Each cluster of the cluster training data set includes a cluster weight. The method also includes receiving a training data set for a machine learning model and generating training data set weights corresponding to the training data set for the machine learning model based on the clustering model. The method also includes adjusting each training data set weight of the training data set weights to match a respective cluster weight and providing the adjusted training data set to the machine learning model as an unbiased training data set.

Abstract: A method for rejecting biased data includes receiving a bias training data set based on a probability distribution of bias-sensitive variables of a target population and segmenting the bias training data set into clusters based on at least one respective bias-sensitive variable of the target population, each cluster including a bias cluster weight. The method also includes receiving a training data set for a machine learning model and segmenting the training data set into training clusters. Each training cluster is associated with at least one corresponding bias-sensitive variable of the target population and includes a corresponding training data set weight. The method also includes adjusting each training data set weight to match a respective bias cluster weight to form an adjusted training data set and providing the adjusted training data set to the machine learning model as an unbiased training data set.

Abstract: A method for rejecting biased data using a machine learning model includes receiving a cluster training data set including a known unbiased population of data and training a clustering model to segment the received cluster training data set into clusters based on data characteristics of the known unbiased population of data. Each cluster of the cluster training data set includes a cluster weight. The method also includes receiving a training data set for a machine learning model and generating training data set weights corresponding to the training data set for the machine learning model based on the clustering model. The method also includes adjusting each training data set weight of the training data set weights to match a respective cluster weight and providing the adjusted training data set to the machine learning model as an unbiased training data set.

Abstract: An imaging system is provided that includes at least one detector configured to acquire imaging information, a processing unit, and a display unit. The processing unit is operably coupled to the at least one detector, and is configured to reconstruct an image using the imaging information. The image is organized into voxels having non-uniform dimensions. The processing unit is configured to perform a penalized likelihood (PL) image reconstruction using the imaging information. The PL image reconstruction includes a penalty function. Performing the penalty function includes interpolating a voxel size in at least one dimension from an original size to an interpolated size before determining a penalty function, determining the penalty function using the interpolated size to provide an initial penalty, interpolating the initial penalty to the original size to provide a modified penalty, and applying the modified penalty in the PL image reconstruction.

Abstract: Implementations relate to providing, in response to a query, machine learning model output that is based on output from a trained machine learning model. The machine learning model output can include a predicted answer to the query, that is predicted based on the trained machine learning model. The machine learning model output can additionally or alternatively include an interactive interface for the trained machine learning model. Some implementations relate to generating a trained machine learning model “on the fly” based on a search query. Some implementations additionally or alternatively relate to storing, in a search index, an association of a machine learning model with a plurality of content items from resource(s) on which the machine learning model was trained.

Abstract: An imaging system is provided that includes at least one detector configured to acquire imaging information, a processing unit, and a display unit. The processing unit is operably coupled to the at least one detector, and is configured to reconstruct an image using the imaging information. The image is organized into voxels having non-uniform dimensions. The processing unit is configured to perform a penalized likelihood (PL) image reconstruction using the imaging information. The PL image reconstruction includes a penalty function. Performing the penalty function includes interpolating a voxel size in at least one dimension from an original size to an interpolated size before determining a penalty function, determining the penalty function using the interpolated size to provide an initial penalty, interpolating the initial penalty to the original size to provide a modified penalty, and applying the modified penalty in the PL image reconstruction.

Abstract: Methods and systems are provided for scatter correction in Positron Emission Tomography (PET) imaging. In one embodiment, a method comprises performing an emission scan to acquire emission data, identifying outliers in a tail region of the emission data, discarding a portion of the outliers from the emission data, calculating a linear fit to remaining emission data in the tail region, and correcting the emission data based on the linear fit. In this way, scatter coincidence events can be eliminated even if the emission data is spatially misaligned with transmission data.

Abstract: A multimedia presentation system for presenting multimedia data comprising a demultiplexing unit and at least one processing unit. The demultiplexing unit demultiplexes a plurality of streams of data. At least one presentation processing unit provides the plurality of stream of data for presentation according to predetermined timing and detecting an end of any one of the plurality of streams of data.

Abstract: A computer-implemented method for penalized-likelihood reconstruction of a Positron Emission Tomography (PET) image includes generating a regularization function in which a smoothing parameter is modulated by one or more data-independent spatially variable modulation factors to compensate for sensitivity variations in a PET voxel dataset, and reconstructing the PET image from the PET emission dataset using the regularization function.

Abstract: A publication-and-subscription mechanism for team rooms, referred to as team room “channels”, through which teams are able to selectively share resources from their team rooms with non-members, and that allows selective contributions, modifications, and discussions from non-members. The team room channels operate as bidirectional information pipelines to other team rooms. The team room channels may either be broadcast to all known teams, or be published selectively to one or more designated “target” teams. The receiving team rooms may then subscribe to specific published channels as appropriate. Information items from a team's team room, such as documents, tasks, representations of team members, and/or other resources, may be added to one or more of a team's published channels. Teams can associate specific permissions with each channel, including Read, Contribute, Modify, and/or Discuss. These permissions apply to all items associated with the channel.

Abstract: A computer-implemented method for partial volume correction in Positron Emission Tomography (PET) image reconstruction includes receiving emission data related to an activity distribution, reconstructing the activity distribution from the emission data by maximizing a penalized-likelihood objective function to produce a reconstructed PET image, quantifying an activity concentration in a region of interest of the reconstructed PET image to produce an uncorrected quantitation, and correcting the uncorrected quantitation based on a pre-calculated contrast recovery coefficient value to account for a partial volume error in the uncorrected quantitation.

Abstract: This invention relates to an assay label comprising an amorphous carbon particle, a functionalised dextran polymer attached to the surface of the carbon particle and a first member of a complementary binding pair covalently bonded to the functionalised dextran polymer. The invention also provides a device incorporating the assay label, which further comprises a radiation source adapted to generate a series of pulses of electromagnetic radiation at a wavelength such that the absorption of the radiation by the label generates energy by non-radiative decay; a sample chamber containing a transducer having a pyroelectric or piezoelectric element and electrodes which is capable of transducing energy generated by non-radiative decay into an electrical signal; and a detector which is capable of detecting the electrical signal generated by the transducer.

Abstract: An electronic device may be used to support user authentication based on biometric readings. In this regard, a unique identification parameter may be generated for each user associated with the electronic device. The unique identification parameter may comprise a user identification input parameter (e.g., alphanumerical password) combined with a set of values (e.g., alphanumerical) generated based on biometrics data generated for the user. In this regard, the biometric based values may be generated based on configuring, for each possible biometric identifier, a range of valid values, such as based on a type of biometric identifier and a specified degree of accuracy. User access may be permitted based on obtaining of a subsequent biometric reading, and generating based thereon a second identification parameter that is compared with the unique identification parameters recognized by the electronic device.

Abstract: A computer-implemented method for penalized-likelihood reconstruction of a Positron Emission Tomography (PET) image includes generating a regularization function in which a smoothing parameter is modulated by one or more data-independent spatially variable modulation factors to compensate for sensitivity variations in a PET voxel dataset, and reconstructing the PET image from the PET emission dataset using the regularization function.

Abstract: A computer-implemented method for partial volume correction in Positron Emission Tomography (PET) image reconstruction includes receiving emission data related to an activity distribution, reconstructing the activity distribution from the emission data by maximizing a penalized-likelihood objective function to produce a reconstructed PET image, quantifying an activity concentration in a region of interest of the reconstructed PET image to produce an uncorrected quantitation, and correcting the uncorrected quantitation based on a pre-calculated contrast recovery coefficient value to account for a partial volume error in the uncorrected quantitation.