Abstract: A computer-implemented system and method for identifying citations within regulatory content is disclosed. The method involves receiving image data representing a format and layout of the regulatory content, receiving a language embedding including a plurality of tokens representing words or characters in the regulatory content, and generating a token mapping associating each of the tokens with a portion of the image data. The method also involves receiving the plurality of tokens and token mapping at an input of a citation classifier, the citation classifier having been trained to generate a classification output for each token based on the language embedding and the token mapping, the classification output identifying a plurality of citation tokens within the plurality of tokens.

Abstract: A computer-implemented method for generating regulatory content requirement descriptions is disclosed and involves receiving requirement data including a plurality of requirements including hierarchical information extracted from regulatory content. The method involves identifying parent requirements based on the existence of child requirements on a lower hierarchical level and generating requirement pairs including the parent requirement and at least one child requirement. The method also involves feeding each of the pairs through a conjunction classifier which has been trained to generate a classification output indicative of the pair being not a conjunction (NC), a single requirement conjunction (CSR), or a multiple requirement conjunction (CMR). The method involves generating a set of requirement descriptions based on the classification output generated for each parent requirement.

Abstract: A computer-implemented method for extracting data from a table comprising a plurality of cells is provided. The method comprises: receiving image data representing a physical layout of the table and words or characters in the table; generating a plurality of tokens, each token representing a semantic unit of the words or characters; and generating at least one language embedding and at least one position embedding associated with each token. The method further comprises: associating at least one token with a cell; receiving the tokens at an input of a cell classifier trained to generate a cell classification for each cell based on the language embeddings and the position embeddings of the at least one token associated with the cell; and determining a relationship exists between at least two cells based at least in part on the cell classifications and the position embeddings of the tokens.

Abstract: This disclosure provides systems and methods for the production of formulations of active pharmaceutical ingredients (APIs). In some embodiments, the disclosure provides an automated medicine formulation system comprising a portable and self-contained API formulating apparatus where the API and excipients are formulated to make a drug product meeting drug quality and safety specifications. The automated formulation system produces liquid formulations including, for example, injectable and intravenous medicines. The systems are capable of producing a plurality of individual sterile injectable doses of drug comprising a specific API and excipient(s), which can be formulated on demand in a GMP and FDA acceptable manner.

Abstract: Aldehyde generation includes providing a first input stream, a second input, and an alkene substrate to a reactor system. The first input stream includes a catalyst, a ligand, and an organic solvent. The second input stream includes a mixture of carbon monoxide (CO) and hydrogen gas (H2). The alkene substrate is in either gaseous form or liquid form, the liquid form of the alkene substrate being provided with the first input stream, the gaseous form of the alkene substrate being provided with the second input stream. The reactor system includes a first reactor and a second reactor, where the second reactor is gas permeable and positioned within the first reactor.

Abstract: Modular reactors comprising a chassis, reactor tubing and optionally a cover are disclosed. The chassis comprises a plurality of channels of different lengths into which a length of reactor tubing is placed to create the reactor portion of the flow reactor.

Abstract: System for synthesis of colloidal nanomaterial includes a multi-stage modular flow reactor that includes four distinct reactor modules for in-flow synthesis of colloidal nanomaterial. The system further includes a computer module for monitor and control of operations of the four reactor modules.

Abstract: Modular reactors comprising a chassis, reactor tubing and optionally a cover are disclosed. The chassis comprises a plurality of channels of different lengths into which a length of reactor tubing is placed to create the reactor portion of the flow reactor.

Abstract: A neural network system for performing a processing task on regulatory content and a method for training the system are disclosed. The method for training involves configuring a neural network language model capable of generating a language embedding output in response to receiving content. The method further involves fine-tuning the language model using regulatory content training data to generate a regulatory content language embedding output for regulatory content processed by the language model. The method also involves configuring at least one task specific output layer to generate task specific results in response to receiving the regulatory content language embedding output from the language model, and training the neural network system using task specific training data to output the task specific results, at least a portion of the task specific training data having been labeled prior to configuring the task specific neural network.

Abstract: A computer-implemented method for generating regulatory content requirement descriptions is disclosed and involves receiving requirement data including a plurality of requirements including hierarchical information extracted from regulatory content. The method involves identifying parent requirements based on the existence of child requirements on a lower hierarchical level and generating requirement pairs including the parent requirement and at least one child requirement. The method also involves feeding each of the pairs through a conjunction classifier which has been trained to generate a classification output indicative of the pair being not a conjunction (NC), a single requirement conjunction (CSR), or a multiple requirement conjunction (CMR). The method involves generating a set of requirement descriptions based on the classification output generated for each parent requirement.

Abstract: A neural network system for performing a processing task on regulatory content and a method for training the system are disclosed. The method for training involves configuring a neural network language model capable of generating a language embedding output in response to receiving content. The method further involves fine-tuning the language model using regulatory content training data to generate a regulatory content language embedding output for regulatory content processed by the language model. The method also involves configuring at least one task specific output layer to generate task specific results in response to receiving the regulatory content language embedding output from the language model, and training the neural network system using task specific training data to output the task specific results, at least a portion of the task specific training data having been labeled prior to configuring the task specific neural network.

Abstract: A computer processor implemented method and system for monitoring a condition associated with operating heavy equipment is disclosed. The method involves receiving a plurality of images at an interface of an embedded processor disposed on the heavy equipment, the images providing a view of at least an operating implement of the heavy equipment. The method also involves processing each of the plurality of images using a first neural network implemented on the embedded processor, the first neural network having been previously trained to identify regions of interest within the image. Each region of interest has an associated designation as at least one of a critical region suitable for extraction of critical operating condition information required for operation of the heavy equipment, and a non-critical region suitable for extraction of non-critical operating condition information associated with the operation of the heavy equipment.

Abstract: A computer-implemented system and method for identifying citations within regulatory content is disclosed. The method involves receiving image data representing a format and layout of the regulatory content, receiving a language embedding including a plurality of tokens representing words or characters in the regulatory content, and generating a token mapping associating each of the tokens with a portion of the image data. The method also involves receiving the plurality of tokens and token mapping at an input of a citation classifier, the citation classifier having been trained to generate a classification output for each token based on the language embedding and the token mapping, the classification output identifying a plurality of citation tokens within the plurality of tokens.

Abstract: A method and apparatus for processing an image of fragmented material to identify fragmented material portions within the image is disclosed. The method involves receiving pixel data associated with an input plurality of pixels representing the image of the fragmented material. The method also involves processing the pixel data using a convolutional neural network, the convolutional neural network having a plurality of layers and producing a pixel classification output indicating whether pixels in the input plurality of pixels are located at one of an edge of a fragmented material portion, inwardly from the edge, and at interstices between fragmented material portions. The convolutional neural network includes at least one convolution layer configured to produce a convolution of the input plurality of pixels, the convolutional neural network having been previously trained using a plurality of training images including previously identified fragmented material portions.

Abstract: A method and apparatus for locating and/or determining the condition of a wear part in an image of an operating implement associated with heavy equipment is disclosed. The method involves capturing at least one image of the operating implement during operation of the heavy equipment, the image including a plurality of pixels each having an intensity value. The method also involves selecting successive pixel subsets within the plurality of pixels, and processing each pixel subset to determine whether pixel intensity values in the pixel subset meet a matching criterion indicating a likelihood that the pixel subset corresponds to the wear part. The matching criterion is based on processing a labeled set of training images during a training exercise prior to capturing the at least one image of the operating implement.

Abstract: A method and apparatus for processing an image of fragmented material to identify fragmented material portions within the image is disclosed. The method involves receiving pixel data associated with an input plurality of pixels representing the image of the fragmented material. The method also involves processing the pixel data using a convolutional neural network, the convolutional neural network having a plurality of layers and producing a pixel classification output indicating whether pixels in the input plurality of pixels are located at one of an edge of a fragmented material portion, inwardly from the edge, and at interstices between fragmented material portions. The convolutional neural network includes at least one convolution layer configured to produce a convolution of the input plurality of pixels, the convolutional neural network having been previously trained using a plurality of training images including previously identified fragmented material portions.

Abstract: A method and apparatus for locating and/or determining the condition of a wear part in an image of an operating implement associated with heavy equipment is disclosed. The method involves capturing at least one image of the operating implement during operation of the heavy equipment, the image including a plurality of pixels each having an intensity value. The method also involves selecting successive pixel subsets within the plurality of pixels, and processing each pixel subset to determine whether pixel intensity values in the pixel subset meet a matching criterion indicating a likelihood that the pixel subset corresponds to the wear part. The matching criterion is based on processing a labeled set of training images during a training exercise prior to capturing the at least one image of the operating implement.