Abstract: Various embodiments of the present disclosure provide an automated language model-based training framework for machine learning classifiers. The techniques comprise receiving a data request that includes a search parameter to be applied to a plurality of data elements, generating a search embedding corresponding to the search parameter via an encoder language model, detecting a first subset of data elements among the plurality of data elements that comprise corresponding embeddings that align with the search embedding in accordance with a first threshold, detecting a second subset of data elements among the first subset of data elements that comprise corresponding classification scores that satisfy a second threshold, and transmitting one or more data packets that comprises one or more data elements of the second subset of data elements in response to the data request.

Abstract: Various embodiments of the present disclosure provide methods, apparatus, systems, computing devices, computing entities, and/or the like for processing a constrained query by (i) generating a cross-reference document data object by (a) extracting a plurality of questions from a guideline document, (b) assigning a rank to each of a plurality of passages from an unstructured data object for each of the plurality of questions, (c) generating a plurality of answers for the plurality of questions based on top ranking passages of the plurality of passages for each question to retrieval machine learning model, and (d) combining the plurality of answers, (ii) generating one or more cross-reference embeddings based on the cross-reference document data object, and (iii) training a predictive machine learning model based on the one or more cross-reference embeddings.

Abstract: Various embodiments of the present disclosure provide prompt engineering and iterative, feedback-based generative techniques that improve traditional LLM technology, including extractive LLM techniques.

Abstract: Various embodiments of the present disclosure provide prompt engineering and iterative, feedback-based generative techniques that improve traditional LLM technology, including extractive LLM techniques. The techniques may include selecting one or more simple annotated question-answer pairs for an input data object comprising an input question and an input document from a reference dataset. The techniques may include selecting one or more complex annotated question-answer pairs from the reference dataset. The techniques may include generating a few-shot prompt based on the one or more simple annotated question-answer pairs and the one or more complex annotated question-answer pairs. The techniques may include providing the few-shot prompt to a large language model (LLM) to receive a predictive output for the input question.

Abstract: Various embodiments of the present disclosure provide a user interface and a natural language interface for predictive models. The techniques may include receiving a user interface application programming interface (API) request that indicates an entity feature dataset associated with the entity identifier and/or an event progression model, receiving a model API request via a conversational user interface comprising a natural language query for interacting with the event progression model, receiving a simulated event risk data object for the entity identifier that is generated using the entity feature dataset, the event progression model, and the natural language query, and initiating a rendering of an event progression graphical visualization via the conversational user interface that is based on the simulated event risk data object.

Abstract: Various embodiments of the present disclosure provide a natural language interface for predictive models. The techniques may include generating an event risk data object for an entity identifier based on an entity feature dataset associated with the entity identifier, receiving a natural language query for the event risk data object, generating a structured data object from the natural language query that comprises a structured data format associated with a defined application programming interface (API) for an event progression model, generating a simulated event risk data object for the entity identifier using the event progression model, and initiating the performance of a prediction-based action based on the simulated event risk data object.

Abstract: Various embodiments of the present disclosure provide methods, apparatus, systems, computing devices, computing entities, and/or the like for improving question-answer (QA) machine learning model training based on generating predicted label indicators, generating prediction score indicators and prediction explanation indicators, generating structured label-explanation datasets, generating synthetic QA training datasets, generating a prediction output, and initiating the performance of one or more prediction-based operations based on the prediction output.

Abstract: Systems and methods are described for training and/or using a machine-learning model. A first set of textual data is received. Using a trained machine-learning model that is applied to the first set, a classification of the first set is generated. The trained machine-learning model has been trained based on a subset of textual data that resulted from filtering a set of training textual data. The filtering of the set of training textual data to generate the subset of textual data is based on a comparison between the training textual data and a second set of textual data.

Abstract: Various embodiments of the present disclosure provide machine-learning question resolution techniques for improving question response outputs. The techniques may include receiving a plurality of evidence passages from a document set corresponding to an input question. The techniques may include generating, using a retrieval ensemble model, a plurality of evidence predictions for an evidence passage of the plurality of evidence passages based on the input question. The techniques may include generating, using the retrieval ensemble model, a weighted aggregate prediction for the evidence passage based on the plurality of evidence predictions. The techniques may include selecting, a set of input passages from the plurality of evidence passages based on the weighted aggregate prediction. The techniques may include generating, using a machine learning aggregation model, a question response based on the set of input passages and the input question. The techniques may include providing the question response.

Abstract: Various embodiments of the present disclosure provide methods, apparatus, systems, computing devices, computing entities, and/or the like for improving machine learning model training based on receiving labeled training data objects, generating a normal prediction loss parameter, generating a global classification loss parameter, generating a composite loss parameter, and initiating the performance of one or more prediction-based operations.

Abstract: Various embodiments of the present disclosure provide methods, apparatus, systems, computing devices, computing entities, and/or the like for improving question-answer (QA) machine learning model training based on generating predicted label indicators, generating prediction score indicators and prediction explanation indicators, generating structured label-explanation datasets, generating synthetic QA training datasets, generating a prediction output, and initiating the performance of one or more prediction-based operations based on the prediction output.

Abstract: Systems and methods are disclosed for predicting a next text. A method may include receiving one or more documents, such as a document associated with a healthcare provider. The document is then processed to generate one or more tokens which are representative of the document. The document is then processed with a machine-learning model, such as a topic model, and a topic vector is output for the document. Based at least in a part on this topic vector, the document is then processed by one or more expert machine-learning models, which each output a probability vector. The various probability vectors are then further processed to calculate a total probability vector for the document. Based at least in part on the total probability vector for the document, a text output is selected.

Abstract: The present disclosure is directed towards methods and systems for determining multimodal image edits for a digital image. The systems and methods receive a digital image and analyze the digital image. The systems and methods further generate a feature vector of the digital image, wherein each value of the feature vector represents a respective feature of the digital image. Additionally, based on the feature vector and determined latent variables, the systems and methods generate a plurality of determined image edits for the digital image, which includes determining a plurality of set of potential image attribute values and selecting a plurality of sets of determined image attribute values from the plurality of sets of potential image attribute values wherein each set of determined image attribute values comprises a determined image edit of the plurality of image edits.

Abstract: Aspects of the technology described herein relate to guiding collection of ultrasound data collection using motion and/or orientation data. A first instruction for rotating or tilting the ultrasound imaging device to a default orientation may be provided. Based on determining that the ultrasound imaging device is in the default orientation, a second instruction for translating the ultrasound imaging device to a target position may be provided. Based on determining that the ultrasound imaging device is in the target position, a third instruction for rotating or tilting the ultrasound imaging device to a target orientation may be provided.

Abstract: Methods and apparatuses for providing indications of missing landmarks in ultrasound images are described. Some embodiments are directed to apparatuses comprising a processing device configured to obtain data representing an ultrasound image, and determine whether the ultrasound image is clinically usable, wherein the determining comprises determining whether the ultrasound image lacks one or more landmarks. Determining whether the ultrasound image is clinically usable may further comprise determining a quality value representative of a quality of the ultrasound image and comparing the quality value to a threshold quality value. In some embodiments, landmarks comprise one or more anatomical features, such as a rib, a pleural line and an A line, a liver, and a kidney.

Abstract: Aspects of technology described herein relate to guiding collection of ultrasound data collection using motion and/or orientation data. A directional indicator corresponding to an instruction for moving an ultrasound imaging device relative to a subject may be displayed in an augmented reality display. The direction of the directional indicator in the augmented reality display may be independent of an orientation of the ultrasound imaging device. The augmented reality display may include video captured by a camera that depicts the ultrasound imaging device and a fiducial marker on the ultrasound imaging device. The direction of the directional indicator may be based on the pose of the camera relative to the fiducial marker and the rotation and/or tilt of the ultrasound imaging device relative to the axis of gravity. The direction of the directional indicator may also be based on the pose of the camera relative to the subject.

Abstract: The present disclosure is directed towards methods and systems for determining multimodal image edits for a digital image. The systems and methods receive a digital image and analyze the digital image. The systems and methods further generate a feature vector of the digital image, wherein each value of the feature vector represents a respective feature of the digital image. Additionally, based on the feature vector and determined latent variables, the systems and methods generate a plurality of determined image edits for the digital image, which includes determining a plurality of set of potential image attribute values and selecting a plurality of sets of determined image attribute values from the plurality of sets of potential image attribute values wherein each set of determined image attribute values comprises a determined image edit of the plurality of image edits.

Abstract: The present disclosure is directed towards methods and systems for determining multimodal image edits for a digital image. The systems and methods receive a digital image and analyze the digital image. The systems and methods further generate a feature vector of the digital image, wherein each value of the feature vector represents a respective feature of the digital image. Additionally, based on the feature vector and determined latent variables, the systems and methods generate a plurality of determined image edits for the digital image, which includes determining a plurality of set of potential image attribute values and selecting a plurality of sets of determined image attribute values from the plurality of sets of potential image attribute values wherein each set of determined image attribute values comprises a determined image edit of the plurality of image edits.

Abstract: Aspects of technology described herein relate to guiding collection of ultrasound data collection using motion and/or orientation data. A directional indicator corresponding to an instruction for moving an ultrasound imaging device relative to a subject may be displayed in an augmented reality display. The direction of the directional indicator in the augmented reality display may be independent of an orientation of the ultrasound imaging device. The augmented reality display may include video captured by a camera that depicts the ultrasound imaging device and a fiducial marker on the ultrasound imaging device. The direction of the directional indicator may be based on the pose of the camera relative to the fiducial marker and the rotation and/or tilt of the ultrasound imaging device relative to the axis of gravity. The direction of the directional indicator may also be based on the pose of the camera relative to the subject.

Abstract: Aspects of the technology described herein relate to guiding collection of ultrasound data collection using motion and/or orientation data. A first instruction for rotating or tilting the ultrasound imaging device to a default orientation may be provided. Based on determining that the ultrasound imaging device is in the default orientation, a second instruction for translating the ultrasound imaging device to a target position may be provided. Based on determining that the ultrasound imaging device is in the target position, a third instruction for rotating or tilting the ultrasound imaging device to a target orientation may be provided.