Systems and methods for predictive analyses with machine learning systems
A method includes receiving, by one or more processors, a dataset including transition data and factor data. The method includes generating a feature for a machine learning model based on the transition data, generating, via input of at least the feature into the machine learning model, one or more data objects indicative of a transition prediction for a transition from the first stage to the second stage, the machine learning model having been trained: with data sources including training factor data having information other than a chemical constituent of blood, and to output information associated with a transition prediction. The method further includes initiating performance of one or more remedial or analytical actions in response to generating the one or more data objects indicative of the transition prediction.
1 . A computer-implemented method comprising:
receiving, by one or more processors of a computing system, a dataset including transition data and factor data, the transition data being associated with progression of a condition from a first stage to a second stage, the factor data being associated with one or more factors that influence progression of the condition from the first stage to the second stage;
generating, by the one or more processors and by a data analysis module, a feature for a machine learning model based on the transition data, the data analysis module comprising:
a feature extractor that identifies the transition data and the factor data based on analyses of structured text documents in the dataset and unstructured documents in the dataset to generate the feature;
a structured document analyzer that processes the structured text documents from the dataset for input into the feature extractor; and
an unstructured document analyzer that prepares the unstructured documents from the dataset for processing by a natural language processor for input into the feature extractor;
generating, by the one or more processors and via input of at least the feature into the machine learning model, one or more data objects indicative of a transition prediction for a transition from the first stage to the second stage, the machine learning model having been trained with: (i) data sources including training factor data having information other than a chemical constituent of blood, (ii) previously-generated features extracted by the feature extractor, and (iii) previously-generated transition predictions for the transition from the first stage to the second stage, to output information associated with a transition prediction; and
initiating, by the one or more processors, performance of one or more remedial or analytical actions in response to generating the one or more data objects indicative of the transition prediction.
2 . The computer-implemented method of claim 1 , wherein the data sources on which the machine learning model has been trained include at least one of structured data sources or unstructured data sources, including documents storing dietary information, visit summaries, therapy overviews, discharge summaries, or medical adherence information.
3 . The computer-implemented method of claim 2 , wherein the data sources were analyzed using a natural language processing technique, the natural language processing technique including one or more of sentiment analysis, named entity recognition, part of speech tagging, or emotion detection.
4 . The computer-implemented method of claim 1 , wherein the data sources include clinical narratives, the clinical narratives having been analyzed with a sentiment analysis algorithm.
5 . The computer-implemented method of claim 1 , wherein the machine learning model is a first machine learning model and the feature is a first feature, the computer-implemented method further comprising:
generating, by the one or more processors and via input of at least a second feature into a second machine learning model, one or more data objects indicative of at least one characteristic of an identified cluster, the second machine learning model having been trained with data sources including factor data having information other than a chemical constituent of blood.
6 . The computer-implemented method of claim 5 , wherein the one or more remedial or analytical actions are initiated based on the transition prediction and the identified cluster.
7 . The computer-implemented method of claim 5 , wherein the second machine learning model is a clustering model that employs hierarchical or k-means techniques.
8 . The computer-implemented method of claim 1 , wherein the transition prediction is a transition from a first stage of chronic kidney disease to a second stage of chronic kidney disease.
9 . A system, comprising:
one or more processors; and
one or more non-transitory computer-readable media storing instructions that, when executed by the one or more processors, cause the one or more processors to perform operations comprising:
receiving a dataset including transition data and factor data, the transition data being associated with progression of a condition from a first stage to a second stage, the factor data being associated with one or more factors that influence progression of the condition from the first stage to the second stage;
generating, by a data analysis module, a feature for a machine learning model based on the transition data, the data analysis module comprising:
a feature extractor that identifies the transition data and the factor data based on analyses of structured text documents in the dataset and unstructured documents in the dataset to generate the feature;
a structured document analyzer that processes the structured text documents from the dataset for input into the feature extractor; and
an unstructured document analyzer that prepares the unstructured documents from the dataset for processing by a natural language processor for input into the feature extractor;
generating, via input of at least the feature into the machine learning model, one or more data objects indicative of a transition prediction for a transition from the first stage to the second stage, the machine learning model having been trained with: (i) data sources including training factor data having information other than a chemical constituent of blood, (ii) previously-generated features extracted by the feature extractor, and (iii) previously-generated transition predictions for the transition from the first stage to the second stage, to output information associated with a transition prediction; and
initiating performance of one or more remedial or analytical actions in response to generating the one or more data objects indicative of the transition prediction.
10 . The system of claim 9 , wherein the data sources on which the machine learning model has been trained include at least one of structured data sources or unstructured data sources, including documents storing dietary information, visit summaries, therapy overviews, or medical adherence information.
11 . The system of claim 10 , wherein the data sources were analyzed using a natural language processing technique, the natural language processing technique including one or more of sentiment analysis, named entity recognition, part of speech tagging, or emotion detection.
12 . The system of claim 9 , wherein the data sources include clinical narratives, the clinical narratives having been analyzed with a sentiment analysis algorithm.
13 . The system of claim 9 , wherein the machine learning model is a first machine learning model and the feature is a first feature, the operations further comprising:
generating, via input of at least a second feature into a second machine learning model, one or more data objects indicative of at least one characteristic of an identified cluster, the second machine learning model having been trained with data sources including the training factor data having information other than a chemical constituent of blood.
14 . The system of claim 13 , wherein the one or more remedial or analytical actions are initiated based on the transition prediction and the identified cluster.
15 . The system of claim 13 , wherein the second machine learning model is a clustering model that employs hierarchical or k-means techniques.
16 . The system of claim 13 , wherein the transition prediction is a transition from a first stage of chronic kidney disease to a second stage of chronic kidney disease.
17 . One or more non-transitory computer-readable media storing instructions that, when executed by one or more processors, cause the one or more processors to perform operations comprising:
receiving a dataset including transition data and factor data, the transition data being associated with progression of a condition from a first stage to a second stage, the factor data being associated with one or more factors that influence progression of the condition from the first stage to the second stage;
generating, by a data analysis module, a feature for a machine learning model based on the transition data, the data analysis module comprising:
a feature extractor that identifies the transition data and the factor data based on analyses of structured text documents in the dataset and unstructured documents in the dataset to generate the feature;
a structured document analyzer that processes the structured text documents from the dataset for input into the feature extractor; and
an unstructured document analyzer that prepares the unstructured documents from the dataset for processing by a natural language processor for input into the feature extractor;
generating, via input of at least the feature into the machine learning model, one or more data objects indicative of a transition prediction for a transition from the first stage to the second stage, the machine learning model having been trained with: (i) data sources including training factor data having information other than a chemical constituent of blood, (ii) previously-generated features extracted by the feature extractor, and (iii) previously-generated transition predictions for the transition from the first stage to the second stage, to output information associated with a transition prediction; and
initiating performance of one or more remedial or analytical actions in response to generating the one or more data objects indicative of the transition prediction.
18 . The one or more non-transitory computer-readable media of claim 17 , wherein the data sources were analyzed using a natural language processing technique, the natural language processing technique including one or more of sentiment analysis, named entity recognition, part of speech tagging, or emotion detection.
19 . The one or more non-transitory computer-readable media of claim 17 , wherein the data sources include clinical narratives, the clinical narratives having been analyzed with a sentiment analysis algorithm.
20 . The one or more non-transitory computer-readable media of claim 17 , wherein the model is a first model and the feature is a first feature, the operations further comprising:
generating, via input of at least a second feature into a second model, one or more data objects indicative of at least one characteristic of an identified cluster, the second machine learning model having been trained with data sources including the training factor data having information other than a chemical constituent of blood.