Enhanced batch analysis and device performance determination with user interface
View Patent ↗Systems and methods for glucose device batch analyses. An example method includes obtaining analysis datasets reflecting manufacturing information associated with medical devices included in a device batch, the manufacturing data measuring information associated with individual manufacturing steps, and with the analysis datasets being formatted for input into a machine learning model, and with unique identifying information associated with the medical devices being used to aggregate analysis datasets specific to the medical devices. The analysis datasets are provided to the machine learning model, with the machine learning model being trained to output performance information for individual medical devices indicating whether they are in releasable condition. The machine learning model was trained based on ground truth associated with manufactured devices, and the ground truth was derived based on test information associated with the manufactured devices. An interactive user interface is generated which presents summary information associated with the performance information.
1 . A method implemented by a system of one or more processors, the method comprising:
obtaining analysis datasets reflecting manufacturing information associated with a plurality of medical devices included in a device batch, the manufacturing data measuring information associated with individual manufacturing steps of a plurality of manufacturing steps, wherein at least a subset of the analysis datasets are automatically obtained from systems which associated with individual manufacturing stages, and wherein obtaining an individual dataset comprises:
accessing sensor data associated with, at least, visual characteristics of the medical devices, the sensor data being derived from one or more sensors, and
generating placement data indicative of alignment of the medical devices during manufacture, the placement data being generated via analyzing the sensor data and indicating, at least, a metric associated with a distinction in placement of an individual medical device from a desired placement;
formatting the analysis datasets for input into a machine learning model, and aggregating analysis datasets specific to the medical devices based on unique identifying information associated with the medical devices;
determining output via the machine learning model based on the analysis datasets, wherein a forward pass is computed through the machine learning model and wherein the output includes performance information for individual medical devices of the plurality of medical devices,
wherein the machine learning model was trained based on ground truth associated with manufactured devices, the ground truth being derived based on test information associated with the manufactured devices,
and wherein the performance information is indicative of the devices being in releasable condition; and
generating an interactive user interface which presents summary information associated with the performance information.
2 . The method of claim 1 , wherein the devices are glucose devices and wherein the manufacturing data reflects information indicating accurate placement of a membrane configured to detect glucose.
3 . The method of claim 1 , wherein the devices are configured to detect a plurality of analytes, and wherein the manufacturing data is specific to individual analytes.
4 . The method of claim 1 , wherein generating the test information comprises:
determining, for each manufactured device, measures of difference between expected responses to individual analytes of a plurality of analytes and actual responses of the devices; and
generating, based on the measures, a target vector indicating whether the manufactured devices have performance greater than a threshold, wherein the target vector includes a binary value for each manufactured device.
5 . The method of claim 4 , wherein generating the target vector comprises:
assigning, for each analyte of individual manufactured devices, a binary value indicating whether the individual manufactured devices are in specification; and
determining, based on the binary values, whether each manufactured device has performance greater than a threshold.
6 . The method of claim 1 , wherein the interactive user interface responds to user input associated with requesting detailed information, and wherein the interactive user interface updates to reflect graphical representations of the manufacturing information.
7 . The method of claim 1 , wherein the interactive user interface indicates a manufacturing step which was not performed, or which was performed out of bounds as compared to normal performance.
8 . Non-transitory computer storage media storing instructions that when executed by one or more processors, cause the one or more processors to perform operations comprising:
obtaining analysis datasets reflecting manufacturing information associated with a plurality of medical devices included in a device batch, the manufacturing data measuring information associated with individual manufacturing steps of a plurality of manufacturing steps, wherein obtaining analysis datasets comprises:
accessing sensor data associated with, at least, visual characteristics of the medical devices, the sensor data being derived from one or more sensors, and
generating placement data indicative of alignment of the medical devices during manufacture, the placement data being generated via analyzing the sensor data and indicating, at least, a metric associated with a distinction in placement of an individual medical device from a desired placement;
formatting the analysis datasets for input into a machine learning model, and aggregating analysis datasets specific to the medical devices based on unique identifying information associated with the medical devices;
determining output via the machine learning model based on the analysis datasets, wherein the output includes performance information for individual medical devices of the plurality of medical devices,
wherein the machine learning model was trained based on ground truth associated with manufactured devices, the ground truth being derived based on test information associated with the manufactured devices,
and wherein the performance information is indicative of the devices being in releasable condition; and
generating an interactive user interface which presents summary information associated with the performance information.
9 . The computer storage media of claim 8 , wherein the devices are glucose devices and wherein the manufacturing data reflects information indicating accurate placement of a membrane configured to detect glucose.
10 . The computer storage media of claim 8 , wherein the devices are configured to detect a plurality of analytes, and wherein the manufacturing data is specific to individual analytes.
11 . The computer storage media of claim 8 , wherein generating the test information comprises:
determining, for each manufactured device, measures of difference between expected responses to individual analytes of a plurality of analytes and actual responses of the devices; and
generating, based on the measures, a target vector indicating whether the manufactured devices have performance greater than a threshold, wherein the target vector includes a binary value for each manufactured device.
12 . The computer storage media of claim 11 , wherein generating the target vector comprises:
assigning, for each analyte of individual manufactured devices, a binary value indicating whether the individual manufactured devices are in specification; and
determining, based on the binary values, whether each manufactured device has performance greater than a threshold.
13 . The computer storage media of claim 8 , wherein the interactive user interface responds to user input associated with requesting detailed information, and wherein the interactive user interface updates to reflect graphical representations of the manufacturing information.
14 . The computer storage media of claim 8 , wherein the interactive user interface indicates a manufacturing step which was not performed, or which was performed out of bounds as compared to normal performance.
15 . A system comprising one or more processors and non-transitory computer storage media storing instructions that when executed by the one or more processors, cause the processors to perform operations comprising:
obtaining analysis datasets reflecting manufacturing information associated with a plurality of medical devices included in a device batch, the manufacturing data measuring information associated with individual manufacturing steps of a plurality of manufacturing steps, wherein obtaining analysis datasets comprises:
accessing sensor data associated with, at least, visual characteristics of the medical devices, the sensor data being derived from one or more sensors, and
generating placement data indicative of alignment of the medical devices during manufacture, the placement data being generated via analyzing the sensor data and indicating, at least, a metric associated with a distinction in placement of an individual medical device from a desired placement;
formatting the analysis datasets are formatted for input into a machine learning model, and aggregating analysis datasets specific to the medical devices based on unique identifying information associated with the medical devices;
determining output via providing the analysis datasets to the machine learning model based on the analysis datasets, wherein the output includes performance information for individual medical devices of the plurality of medical devices,
wherein the machine learning model was trained based on ground truth associated with manufactured devices, the ground truth being derived based on test information associated with the manufactured devices,
and wherein the performance information is indicative of the devices being in releasable condition; and
generating an interactive user interface which presents summary information associated with the performance information.
16 . The system of claim 15 , wherein the devices are glucose devices and wherein the manufacturing data reflects information indicating accurate placement of a membrane configured to detect glucose.
17 . The system of claim 15 , wherein the devices are configured to detect a plurality of analytes, and wherein the manufacturing data is specific to individual analytes.
18 . The system of claim 15 , wherein generating the test information comprises:
determining, for each manufactured device, measures of difference between expected responses to individual analytes of a plurality of analytes and actual responses of the devices; and
generating, based on the measures, a target vector indicating whether the manufactured devices have performance greater than a threshold, wherein the target vector includes a binary value for each manufactured device.
19 . The system of claim 18 , wherein generating the target vector comprises:
assigning, for each analyte of individual manufactured devices, a binary value indicating whether the individual manufactured devices are in specification; and
determining, based on the binary values, whether each manufactured device has performance greater than a threshold.
20 . The system of claim 15 , wherein the interactive user interface responds to user input associated with requesting detailed information, and wherein the interactive user interface updates to reflect graphical representations of the manufacturing information.