Systems and methods for ai assisted echocardiography

A system for conducting a measurement procedure on a patient, the system comprising:

• at least one measurement tools configured to perform the measurement procedure;
• at least one recording device configured to record measurement data from the patient during the measurement procedure;
• and
• at least one transmitter configured to transmit the measurement data to at least one processor, wherein the processor is configured to:
• receive the measurement data and phenotype data;
• associate the measurement data and phenotype data to determine a patient phenotype associated with one or more disease states;
• generate at least one of:
• a measurement prediction protocol configured to predict measurement data for unpopulated measurement fields;
• and
• a phenotype model configured to associate the patient data with a phenotype associated with one or more disease state,
• predict a probable disease state for the patient undergoing the measurement procedure;
• and
• alert a measurement operator of the predicted measurement data and the probable disease state;
• and
• provide a set of directions to the measurement operator to collect a dataset of relevant measurement data to be collected based on the predicted probable disease state.

A non-transitory computer readable medium that stores a set of instructions that is executable by at least one processor of a computing device to cause the computing device to perform operations for generating an inspection tool sampling plan, the operations comprising:

• (a) retrieving data, from a sparsely populated data source, to form a base dataset, the data source comprising a plurality of patient records including patient mortality data, wherein each patient record comprises_at least one unpopulated data field corresponding to a medical measurement;
• (b) dividing the base dataset into two portions:
• a first portion comprising a training dataset being a defined percentage, X %, of the base dataset;
• and
• a second portion comprising a validation dataset being a defined percentage (100%-X %) of the base dataset;
• (c) analysing the training dataset to jointly model variable relationships using a non-linear function approximation algorithm applied iteratively to the records of the training dataset to obtain a trained model and measurement prediction protocols for populating unpopulated fields in the training dataset;
• (d) using the measurement prediction protocols, computing predicted measurement values for each of the unpopulated data fields;
• (e) imputing the predicted measurement values in the patient records of the training dataset;
• (f) analysing the training dataset based on at least one predefined disease conditions in a known portion of the patient records of the base dataset to form a phenotype model configured to associate patient phenotype data to a probability of a disease condition in the patient records of a trained dataset;
• (g) imputing the predicted measurement values in the records of the validation dataset;
• (h) validating the phenotype model, where the validating comprises analysing the validation dataset using the phenotype model, wherein the records of the validation dataset comprise phenotype data associated with patient data, and determining a validation error comprising a probability of correctly predicting the patient phenotype associated with a disease state probability in the records of the validation dataset;
• (i) repeating Steps (c) to (h) to minimise the validation error and computing a prediction of a probable disease state phenotype for each patient record in the base dataset.