Abstract: Systems and methods for performing medical audio summarizing for medical conversations are disclosed. An audio file and meta data for a medical conversation are provided to a medical audio summarization system. A transcription machine learning model is used by the medical audio summarization system to generate a transcript and a natural language processing service of the medical audio summarization system is used to generate a summary of the transcript. The natural language processing service may include at least four machine learning models that identify medical entities in the transcript, identify speaker roles in the transcript, determine sections of the transcript corresponding to the summary, and extract or abstract phrases for the summary. The identified medical entities and speaker roles, determined sections, and extracted or abstracted phrases may then be used to generate the summary.

Abstract: Techniques for determining a drift impact score in a machine learning model are disclosed. The techniques can include: obtaining a reference distribution of a machine learning model; obtaining a current distribution of the machine learning model; determining a statistical distance based on the reference distribution and the current distribution; determining a local feature importance parameter for each feature associated with a prediction made by the machine learning model; determining a cohort feature importance parameter for a cohort of multiple features based on the local feature importance parameter of each feature in the cohort; and determining a drift impact score for the cohort based on the statistical distance and the cohort feature importance parameter.

Abstract: Techniques for optimizing a machine learning model. The techniques can include: obtaining one or more embedding vectors based on a prediction of a machine learning model; mapping the embedding vectors from a higher dimensional space to a 2D/3D space to generate one or more high density points in the 2D/3D space; clustering the high-density points by running a clustering algorithm multiple times, each time with a different set of parameters to generate one or more clusters; applying a purity metric to each cluster to generate a normalized purity score of each cluster; identifying one or more clusters with a normalized purity score lower than a threshold; and optimizing the identifying one or more clusters.

Abstract: Techniques for determining a drift impact score in a machine learning model are disclosed. The techniques can include: obtaining a reference distribution of a machine learning model; obtaining a current distribution of the machine learning model; determining a statistical distance based on the reference distribution and the current distribution; determining a local feature importance parameter for each feature associated with a prediction made by the machine learning model; determining a cohort feature importance parameter for a cohort of multiple features based on the local feature importance parameter of each feature in the cohort; and determining a drift impact score for the cohort based on the statistical distance and the cohort feature importance parameter.

Abstract: Techniques for determining embedding drift score in a machine learning model. The techniques can include: obtaining one or more first embedding vectors based on at least one first prediction of a machine learning model; filtering the first embedding vectors based on a slice of the first prediction; determining a first average vector by averaging each dimension of the filtered first embedding vectors; obtaining one or more second embedding vectors on at least one second prediction of the machine learning model; filtering the second embedding vectors based on a slice of the second prediction; generating a second average vector by averaging each dimension of the filtered second embedding vectors; and determining an embedding drift score based on a distance measure of the first average vector and the second average vector.

Abstract: The present disclosure relates to systems, non-transitory computer-readable media, and methods to generate sketches for clearing-bid values and bid-success rates based on multi-dimensional targeting criteria for a digital-content campaign and dynamically determine predicted values for the digital-content campaign based on the sketches. To illustrate, the disclosed systems can use a running-average-tuple-sketch to generate tuple sketches of historical clearing-bid values and tuple sketches of historical bid-success-rates from historical auction data. Based on the tuple sketches, the disclosed systems can determine one or more of a predicted cost per quantity of impressions, a predicted number of impressions, or a predicted expenditure for the digital-content campaign—according to user-input targeting criteria and expenditure constraints.

Abstract: The present disclosure relates to systems, non-transitory computer-readable media, and methods to generate sketches for clearing-bid values and bid-success rates based on multi-dimensional targeting criteria for a digital-content campaign and dynamically determine predicted values for the digital-content campaign based on the sketches. To illustrate, the disclosed systems can use a running-average-tuple-sketch to generate tuple sketches of historical clearing-bid values and tuple sketches of historical bid-success-rates from historical auction data. Based on the tuple sketches, the disclosed systems can determine one or more of a predicted cost per quantity of impressions, a predicted number of impressions, or a predicted expenditure for the digital-content campaign—according to user-input targeting criteria and expenditure constraints.

Abstract: A development application can be configured to facilitate editing multiple fragments already associated with an interface design simultaneously using a graphic design application, with the fragments arranged in a logical form supported by the graphic design application. The logical form of the artwork fragments as provided to the graphic design application can correspond to the relationship between different elements of the interface design using the respective fragments. Fragments may correspond to different components, parts of components, and part states, for instance. A method can comprise accessing data defining an interface component and comprising a plurality of logically arranged elements of the component. The method can comprise extracting the artwork and arranging the artwork into a form compatible with a graphic design application. The artwork can be edited in a graphic design application and then re-integrated into the data defining the design elements of the interface after editing is complete.

Abstract: Methods and systems for animating visual components representing data items. One embodiment comprises a method for producing an application using declarative language code to specify animation behavior for data item representations. A programming application may be used to create the declarative language code using a display design area for placing and adjusting objects such as data item containers and/or an editor for entering and editing code. One embodiment comprises a method that allows an application, such as a rich Internet application, to create representations of displayed objects and virtually displayed objects to facilitate animation. One embodiment involves facilitating animation using initial and changed layouts, such layouts including representing of a limited number of data items both inside and outside the content display area.

Abstract: Methods and systems for animating visual components representing data items. One embodiment comprises a method for producing an application using declarative language code to specify animation behavior for data item representations. A programming application may be used to create the declarative language code using a display design area for placing and adjusting objects such as data item containers and/or an editor for entering and editing code. One embodiment comprises a method that allows an application, such as a rich Internet application, to create representations of displayed objects and virtually displayed objects to facilitate animation. One embodiment involves facilitating animation using initial and changed layouts, such layouts including representing of a limited number of data items both inside and outside the content display area.

Abstract: The application provides, in part, vectors based on novel tobacco rattle virus replicons, as well as methods for using such vectors and transgenic plants.

Abstract: This disclosure provides a system for generating animal tissue with carbohydrate antigens that are compatible for transplantation into human patients. The tissue is inactivated homozygously for expression of ?(1,3)galactosyltransferase, and comprises a transgene for ?(1,2)fucosyltransferase. As a result, cell-surface N-acetyl lactosamine is not converted to the Gal?(1,3)Gal xenoantigen. Instead, it is converted to Fuc?(1,2)Gal, which is H substance, a self-antigen in humans. The tissue may also contain A or B-transferase, which will cause H substance to be converted into other ABO blood group antigens for compatibility with patients of the same blood type. This invention improves transplant compatibility of the xenograft tissue by lessening the risk of reactions resulting from xenoantigen and unconverted N-acetyl lactosamine acceptor determinants.

Abstract: This invention provides a system for producing differentiated cells from a stem cell population for use wherever a relatively homogenous cell population is desirable. The cells contain an effector gene under control of a transcriptional control element (such as the TERT promoter) that causes the gene to be expressed in relatively undifferentiated cells in the population. Expression of the effector gene results in expression of a cell-surface antigen that can be used to deplete the undifferentiated cells. Model effector sequences encode glycosyl transferases that synthesize carbohydrate xenoantigen or alloantigen, which can be used for immunoseparation or as a target for complement-mediated lysis. The differentiated cell populations produced are suitable for use in tissue regeneration and non-therapeutic applications such as drug screening.

Abstract: This invention provides a system for producing differentiated cells from a stem cell population for use wherever a relatively homogenous cell population is desirable. The cells contain an effector gene under control of a transcriptional control element (such as the TERT promoter) that causes the gene to be expressed in relatively undifferentiated cells in the population. Expression of the effector gene results in expression of a cell-surface antigen that can be used to deplete the undifferentiated cells. Model effector sequences encode glycosyl transferases that synthesize carbohydrate xenoantigen or alloantigen, which can be used for immunoseparation or as a target for complement-mediated lysis. The differentiated cell populations produced are suitable for use in tissue regeneration and non-therapeutic applications such as drug screening.

Abstract: This disclosure provides a system for specifically killing cancer cells which can be used in the course of human therapy. Vectors of the invention comprise an encoding sequence for a glycosyltransferase, under control of a tumor or tissue specific transcriptional control element, such as the promoter for telomerase reverse transcriptase. Exemplary glycosyltransferases are the A or B transferase enzymes, which cause the cancer cells to express ABO histo blood group allotypes against which humans have naturally occurring antibody. This provides for ongoing surveillance for newly emerging cells with a malignant phenotype.

Abstract: Described in this disclosure is a new process whereby cells of one tissue type can be reprogrammed to produce cells of a different tissue type. Cells from a human donor are reprogrammed by culturing adjacent to primate pluripotent stem cells (in an undifferentiated or newly differentiated state) or in an environment supplemented by components taken from pPS cells. Simultaneously or in a subsequent step, the donor cells can be treated in a manner that enhances differentiation towards a different tissue type. In this manner, patients in need of tissue regeneration can be treated with cells differentiated and reprogrammed from their own autologous cell donation.

Abstract: The application provides, in part, vectors based on novel tobacco rattle virus replicons, as well as methods for using such vectors and transgenic plants.

Abstract: It has been discovered that the specificity of multiple transcriptional regulatory elements can be combined to make vector systems that selectively target cancer cells. The promoter for telomerase reverse transcriptase (TERT) can be combined in a remarkably synergistic fashion with another promoter that has expression restricted to cancer cells or a particular tissue type. The two promoters work synergistically for exquisite targeting of the malignant cells—where it causes cell lysis while leaving neighboring healthy cells intact.

Abstract: This disclosure provides a system for generating animal tissue with carbohydrate antigens that are compatible for transplantation into human patients. The tissue is inactivated homozygously for expression of &agr;(1,3)galactosyltransferase, and comprises a transgene for &agr;(1,2)fucosyltransferase. As a result, cell-surface N-acetyl lactosamine is not converted to the Gal&agr;(1,3)Gal xenoantigen. Instead, it is converted to Fuc&agr;(1,2)Gal, which is H substance, a self-antigen in humans. The tissue may also contain A or B-transferase, which will cause H substance to be converted into other ABO blood group antigens for compatibility with patients of the same blood type. This invention improves transplant compatibility of the xenograft tissue by lessening the risk of reactions resulting from xenoantigen and unconverted N-acetyl lactosamine acceptor determinants.

Abstract: This invention provides a system for producing differentiated cells from a stem cell population for use wherever a relatively homogenous cell population is desirable. The cells contain an effector gene under control of a transcriptional control element (such as the TERT promoter) that causes the gene to be expressed in relatively undifferentiated cells in the population. Expression of the effector gene results in expression of a cell-surface antigen that can be used to deplete the undifferentiated cells. Model effector sequences encode glycosyl transferases that synthesize carbohydrate xenoantigen or alloantigen, which can be used for immunoseparation or as a target for complement-mediated lysis. The differentiated cell populations produced are suitable for use in tissue regeneration and non-therapeutic applications such as drug screening.