Systems, methods, computer medium, and program products for maintaining data integrity in health analysis platform by evaluating and modifying physiological measurements based on filtered health care data

Abstract

Systems, methods, computer medium, and program products are disclosed for maintaining data integrity in a computerized health analysis platform. For example, a method includes (i) filtering existing health care data by first determining or extracting a first subset of data of a dataset such that the first subset is concentrated in a common health-related attribute (s), (ii) generating a reference measurement range from the extracted first subset, and (iii) determining measurement units for laboratory test data lacking measurement units or exhibiting mismarking errors based on the reference measurement range. For example, a first subset of the data sets represents measurements of the physiological parameter (s) of the entity. For example, the laboratory test data differs from the first subset or existing healthcare data used to determine the first subset. After the measurement unit is determined, a data structure representing the measurement unit is generated and stored.

Description

TECHNICAL FIELD

[0001] The present specification generally relates to systems and methods for maintaining data integrity in a health analytics platform by evaluating and modifying physiological measurements against filtered health data. BACKGROUND

[0002] Generally, a user’s health can be evaluated by measuring one or more physiological characteristics of the user and comparing the measured physiological characteristics to health references. For example, the health references can correspond to or be derived from existing health data, including historical data, prior clinical trial data, real-time data, and other existing health data. Thus, having accurate measurements of existing health data can improve the quality of the evaluation. SUMMARY

[0003] Implementations in accordance with the present disclosure include a system for maintaining data integrity in a computerized health analytics platform. The system includes at least one processor and a memory subsystem communicatively coupled to the at least one processor. The memory subsystem stores instructions that, when executed by the at least one processor, cause the at least one processor to perform operations including: (i) accessing one or more first data structures comprising a plurality of first data sets pertaining to a plurality of entities, (ii) determining a first subset of the first data sets based on one or more health-related attributes of the plurality of entities, (iii) generating a reference measurement range for one or more physiological parameters of the plurality of entities of the first subset of the first data sets in one or more measurement units, (iv) accessing one or more second data structures comprising one or more second data sets pertaining to one or more target entities, (v) determining, based on the reference measurement range, measurement units for which one or more measurements are missing or exhibit mislabeling errors with respect to the measurement units, (vi) generating a third data structure representing the measurement units, and (vii) storing the third data structure in a hardware storage device. Each of the first data sets in the plurality of first data sets represents measurements of one or more physiological parameters of the plurality of entities and one or more health-related attributes of the plurality of entities, wherein a machine that transmitted the data structure is determined to be authorized to access the data structure by the system accessing the data structure. The one or more second data sets represent one or more measurements of one or more physiological parameters of the one or more target entities, wherein the one or more target entities exhibit the one or more health-related attributes. The one or more measurements of the one or more target entities are missing measurement units or exhibit mislabeling errors with respect to the measurement units.

[0004] Implementations in accordance with this disclosure include methods for maintaining data integrity in a computerized health analysis platform. The method includes (i) accessing, by an electronic device, one or more first data structures comprising a plurality of first data sets pertaining to a plurality of entities, (ii) determining, by the electronic device, a first subset of the first data sets based on one or more health-related attributes of the plurality of entities, (iii) generating, by the electronic device, a reference measurement range for one or more physiological parameters of the plurality of entities of the first subset of the first data sets in one or more measurement units, (iv) accessing, by the electronic device, one or more second data structures comprising one or more second data sets pertaining to one or more target entities, (v) determining, by the electronic device, measurement units of one or more measurement results that lack measurement units or exhibit mislabeling errors with respect to measurement units based on the reference measurement range, (vi) generating, by the electronic device, a third data structure representing the measurement units, and (vii) storing, by the electronic device, the third data structure in a hardware storage device. Each of the first data sets in the first data sets represents measurement results of one or more physiological parameters of the plurality of entities and one or more health-related attributes of the plurality of entities, wherein a machine that transmitted the data structure is authorized to access the data structure by a system that accesses the data structure. The one or more second data sets represent one or more measurement results of one or more physiological parameters of the one or more target entities, wherein the one or more target entities exhibit the one or more health-related attributes. The one or more measurement results of the one or more target entities lack measurement units or exhibit mislabeling errors with respect to measurement units.

[0005] Other embodiments of this aspect include corresponding computer systems, apparatus, and computer programs recorded on one or more computer storage devices, each configured to perform the actions or operations described herein. A computer system can be configured to perform particular actions by virtue of having software, firmware, hardware, or a combination thereof installed on the system that in operation causes the system to perform the actions. One or more computer programs can be configured to perform particular actions by virtue of including instructions that, when executed by data processing apparatus, cause the apparatus to perform the actions.

[0006] The details of one or more embodiments of the subject matter of this specification are set forth in the accompanying drawings and the description below. Other features, aspects, and advantages of the subject matter will become apparent from the description, the drawings, and the claims. BRIEF DESCRIPTION OF DRAWINGS

[0007] Figure 1 is a diagram of an example data integrity maintenance system.

[0008] Figure 2 is an example implementation of software or algorithms utilized by the electronic device.

[0009] Figure 3 is a flowchart of an example process for determining a unit of measurement for laboratory test data.

[0010] Figure 4 is a flowchart of an example process for determining a unit of measurement for laboratory test data and modifying the laboratory test data.

[0011] Figure 5 is a diagram of an example computer system.

[0012] Figure 6 is a frequency curve graph illustrating a relationship between a frequency of a physiological measurement and a unit of measurement.

[0013] The same reference numbers and designations in the various drawings indicate the same elements. DETAILED DESCRIPTION

[0014] Accurate laboratory test data is important in various aspects of healthcare. For example, in a clinical context, accurate laboratory test results are critical for diagnosing and monitoring diseases because they provide key information about a patient’s health condition. For example, blood test data can include important physiological parameter(s) indicative of a disease or indication such as diabetes, cardiovascular disease, and infection. Accurate laboratory test data is also important in a clinical trial, where accurate laboratory test data is used as reference data for assessing the effectiveness and safety of a treatment. Furthermore, in a research context, accurate laboratory test data forms the basis for achieving an accurate understanding, correct diagnosis, and advancement of medical research. If inaccurate, laboratory test data can lead to misdiagnosis, ineffective treatment, and misleading research results, which in turn can endanger patient safety and hinder medical progress.

[0015] However, laboratory test data can exhibit data inconsistency or errors. For example, laboratory test data including measurements (e.g., for physiological parameters) can lack a unit of measurement or exhibit mislabeling errors.

[0016] For example, inconsistency can occur frequently due to various reasons including methods, samples, units of measurement, changes in in-vitro diagnostic devices, different data entry standards, human errors, sensor inaccuracies, data format differences, etc.

[0017] In particular, when the units of measurement for data entries are missing or mislabeled (e.g., the units of measurement are incorrectly labeled or incorrectly recorded), this can lead to significant errors in data interpretation and analysis. For example, if a laboratory report incorrectly labels cholesterol levels in milligrams per deciliter (mg / dL) as grams per liter (g / L), this can lead to a misunderstanding of a patient’s health condition and potentially lead to inappropriate treatment decisions.

[0018] Accordingly, when a healthcare assessment, treatment decision, or medical research relies on such laboratory test data that lacks units of measurement or exhibits mislabeling errors, this can lead to misdiagnosis, ineffective treatment, and misleading research results, which in turn can compromise patient safety and hinder medical progress.

[0019] Implementations in accordance with the present disclosure address the above data issues by: [1] filtering existing healthcare data (e.g., historical data, prior clinical trial data, real-time data, and other existing healthcare data) by first determining or extracting a first subset of data such that the first subset of data focuses on a common health-related attribute(s), [2] generating a reference measurement range with the extracted first subset of data, and [3] determining units of measurement for laboratory test data that lacks units of measurement or exhibits mislabeling errors based on the reference measurement range.

[0020] For example, a data integrity maintenance system can filter such existing healthcare data by first determining or extracting a first subset of data based on a health-related attribute(s), at least. As an example, a first subset of time series data can be determined or extracted based on one or more health-related attributes that match attributes of laboratory test data being evaluated (e.g., for measurement data accuracy). The health-related attributes can include an indication of a disease, a medical condition other than an indication of a disease, a same medication use, a same medical treatment, or a gender, among others. Such determination or extraction of a first subset of time series data can be advantageous because ranges of physiological measurement data can be tailored based on commonalities of the health-related attribute(s). Accordingly, by generating a reference measurement range based on these commonalities of the health-related attribute(s) between existing healthcare data and laboratory data, laboratory data can be more accurately evaluated based on such a reference measurement range. For example, the reference measurement range can be used to determine whether laboratory test data lacks units of measurement or exhibits mislabeling errors and to determine correct units of measurement for corresponding laboratory test data.

[0021] After determining or extracting the first subset of data based on the health-related attribute(s), a reference measurement range can be generated. For example, generating the reference measurement range includes [1] comparing the physiological parameters of the first subset of data to each other, and [2] converting different measurement units of the physiological parameters of the first subset into one or more measurement units. By filtering the first subset of data to have the reference measurement range in the form of one or more measurement units, data uniformity and data comparability are ensured. For example, when the reference measurement range is generated in one or more single units (e.g., multiple reference ranges for cholesterol levels in mg / dL, g / L, etc.), such reference measurement range not only ensures data uniformity, but also enables simplified calculations and reduction of errors (e.g., when comparing the reference measurement range with different data or laboratory test data).

[0022] After generating the reference measurement range, the measurement unit(s) for the laboratory test data that lack measurement units or exhibit mislabeling errors can be determined based on the reference measurement range, which is customized based on the health-related attribute(s). Further, before determining the measurement units, a determination can be made as to whether the laboratory data lack measurement units or exhibit mislabeling errors. For example, based on a comparison of the measurement values of the laboratory data that share a common health-related attribute(s) with the reference measurement range, it can be determined whether the laboratory test data lack measurement units or exhibit mislabeling errors. Once the measurement units are determined for the corresponding laboratory test data that lack measurement units or exhibit labeling errors, an indication or notification can be generated on the display of the user interface, as well as a frequency curve graph representing the relationship between the measurement frequency and the measurement units.

[0023] Further, data quality of the laboratory data can be improved by preventing processing of incomplete or erroneous laboratory data. Data quality of the laboratory data can be improved by incorporating the determined measurement unit(s) into the missing measurement units of the corresponding laboratory data or replacing each measurement unit of the corresponding laboratory data that exhibits a labeling error with the determined measurement unit(s).

[0024] Further, based on the improvement in data quality, the accuracy of the laboratory data can lead to reliable health assessments, effective treatment decisions, and promotion of medical research. For example, such accurate laboratory data can be used as reference data for: clinical trials to evaluate efficacy and safety of treatments, current health monitoring and diagnosis to assess patient conditions, and research purposes.

[0025] Further, the embodiments described herein can also reduce the amount of computer resources consumed in processing health data. For example, when generating a health assessment, a computer system encountering low quality data can generate results with errors and / or inconsistencies that are not suitable for use. As a result, the computer system can reprocess the data multiple times (e.g., based on manual feedback from a user) until a satisfactory result is achieved. These repeated operations can increase the amount of computing resources (e.g., CPU utilization), memory resources, storage resources, etc. consumed during the health assessment processing. The embodiments described herein can be used to automatically identify and correct labeling errors and / or inconsistencies, thereby [1] reducing the likelihood that health data is reprocessed due to low quality, and [2] reducing the consumption of computer resources.

[0026] Figure 1 An example data integrity maintenance system 100 is shown. In particular, the data integrity maintenance system 100 can [1] filter existing health data, [2] generate a reference measurement range from a first subset of extracted data, and [3] determine a measurement unit for laboratory test data that is missing a measurement unit or exhibits a mislabeling error based on the reference measurement range.

[0027] The data integrity maintenance system 100 can include an electronic device 120 and a sensor device 110 communicatively coupled to each other (e.g., via one or more wired or wireless communication links 150). Generally, the data integrity maintenance system 100 accesses a data structure (e.g., health-related data, such as existing health data or laboratory test data stored in a data store, such as a database module 122, or otherwise accessible to the electronic device 120, e.g., through a server), and determines the data integrity (e.g., data quality) of the data structure by processing methods according to implementations described in this disclosure. Further, in some implementations, the data integrity maintenance system 100 obtains sensor data about a user using the sensor device 110, and processes the sensor data using the electronic device 120 to determine one or more biomarkers indicative of a medical condition of the user.

[0028] Generally, the electronic device 120 can include any number of devices configured to receive, process, and transmit data. Examples of the electronic device 120 include a client computing device (e.g., a desktop or laptop computer), a server computing device (e.g., a server computer or a cloud computing system), a mobile computing device (e.g., a cellular phone, a smartphone, a tablet computer, a personal data assistant, a laptop computer with networking capabilities), a wearable computing device (e.g., a smartphone or a headset), and other computing devices capable of receiving, processing, and transmitting data. In some implementations, the electronic device 120 can include computing devices that operate using one or more operating systems (e.g., Microsoft Windows, Apple macOS, Linux, Unix, Google Android, and Apple iOS, among others) and one or more architectures (e.g., x86, PowerPC, and ARM, among others).

[0029] The sensor device 110 includes one or more sensors 112 configured to obtain measurements related to a user’s physiology, a user’s behavior, and / or any other characteristic of a user. For example, the sensor device 110 can include or correspond to a wearable device (e.g., a smartwatch), a smartphone, a medical monitoring system, a laboratory instrument, and / or the like. As an example, the sensor device can include one or more sensors 112 configured to obtain physiological parameters including vital signs such as glucose level, heart rate, blood pressure, respiration rate, or temperature, among others. For example, the one or more sensors can be an optical sensor (e.g., PPG), a pulse pressure sensor (PP), a pressure sensor, an electrocardiogram (ECG), a bioimpedance sensor, a galvanic skin response sensor, an eye pressure measurement / contact sensor, an accelerometer, a gyroscope, a pressure sensor, an acoustic sensor, an electromechanical motion sensor, and / or an electromagnetic sensor. Further, for example, when the sensor device takes the form of a laboratory instrument, it can also measure physiological parameters or conduct blood tests such as analyzing blood glucose levels, cholesterol, and other biomarkers.

[0030] Further, the sensor device 110 includes a communication module 116 configured to transmit and / or receive data with respect to the electronic device 120. As an example, the communication module 116 can include one or more receivers, transmitters, and / or transceivers. In some embodiments, the communication module 116 can communicate with the electronic device 120 via one or more wired links (e.g., serial links, Ethernet links, and / or the like) and / or wireless links (e.g., Wi-Fi links, Bluetooth links, and / or the like).

[0031] In general, the electronic device 120 is configured to receive sensor data (e.g., physiological parameter data, such as clinical parameter(s), etc.) obtained by the sensor apparatus 110 and process the sensor data to determine one or more biomarkers representative of a medical condition of the user. In addition, the electronic device 120 is configured to present information about the biomarkers and any other information to the user and / or another user (e.g., a healthcare provider).

[0032] In Figure 1 the electronic device 120 is illustrated as a single component. In practice, however, the electronic device 120 can be implemented on one or more computing devices (e.g., individual computing devices including at least one processor, such as a microprocessor or microcontroller, etc.). As an example, the electronic device 120 can be a single computing device, such as a single smartphone, etc. As another example, the electronic device 120 can include multiple computing devices connected via a network (e.g., the Internet, a local area network, etc.) and can maintain and operate components of the electronic device 120 on some or all of the computing devices. For example, the electronic device 120 can include several computing devices and components of the electronic device 120 can be distributed on one or more of these computing devices.

[0033] In addition, the electronic device 120 is illustrated as a component that is separate from the sensor apparatus 110. However, while the electronic device 120 can be a separate component from the sensor apparatus 110, the electronic device 120 can also include the sensor apparatus 110, be coupled with the sensor apparatus 110, or be adjacent to the sensor apparatus 110 (e.g., in a housing). For example, the electronic device 120 can be a wearable device that includes the sensor apparatus 110, is coupled with the sensor apparatus 110, or is adjacent to the sensor apparatus 110.

[0034] As Figure 1 shown, the electronic device 120 includes a database module 122, a communication module 124, a processing module 126, and a user interface module 128. The operational modules can be provided as one or more computer-executable software modules, hardware modules, or a combination thereof. For example, one or more of the operational modules can be implemented as a software code block having instructions that cause one or more processors to perform the operations described herein. Additionally or alternatively, one or more of the operational modules can be implemented in an electronic circuit, such as a programmable logic circuit, a field-programmable logic array (FPGA), or an application-specific integrated circuit (ASIC), etc.

[0035] The communication module 124 is configured to transmit and / or receive data with respect to the sensor device 110. As an example, the communication module 124 can include one or more receivers, transmitters, and / or transceivers. In some implementations, the communication module 124 can communicate with the sensor device 110 (e.g., via the communication module 116) via one or more wired links (e.g., serial links, Ethernet links, etc.) and / or wireless links (e.g., Wi-Fi links, Bluetooth links, etc.).

[0036] The database module 122 maintains information related to the operation of the data integrity maintenance system 100.

[0037] As an example, the database module 122 can store input data 122a that is used as input for determining one or more biomarkers indicative of a user’s health. For example, the input data 122a can include at least some of the sensor data generated by the sensor device 110.

[0038] As another example, the database module 122 can store output data 122b generated by the electronic device 120. As an example, the output data 122b can include one or more metrics or biomarkers generated by the electronic device 120 based on the input data 122a.

[0039] Further, the database module 122 can store processing rules 122c that specify how data in the database module 122 can be processed for the operations described herein.

[0040] As an example, the processing rules 122c can include one or more rules that specify how to format, parse, and process the input data 122a to determine one or more corresponding metrics or biomarkers with respect to a user.

[0041] As another example, the processing rules 122c can include one or more rules that specify conditions under which to present data to a user (e.g., using the user interface module 128) and the manner in which to present the data.

[0042] As another example, the processing rules 122c can include one or more rules that specify the manner in which to store data (e.g., using the database module 122) for future retrieval and / or processing.

[0043] Example data processing techniques are described in further detail below.

[0044] The processing module 126 processes data stored or otherwise accessible to the electronic device 120. For example, the processing module 126 can be used to perform one or more of the operations described herein (e.g., by executing a processing rule 122c on input data 112a to generate output data 122b).

[0045] The user interface module 128 is configured to present information to and / or receive input from a user. As an example, the user interface module 128 can include one or more display devices (e.g., a display screen, a touchscreen, etc.) configured to present a user interface (e.g., a graphical user interface (GUI)) that enables a user to interact with the electronic device 120 and / or the sensor device 110. Example interactions include viewing data, transferring data from one component to another, and / or issuing commands to the electronic device 120 and / or the sensor device 110. Commands can include, for example, any user instructions to perform a particular operation or task on one or more of the electronic device 120 and / or the sensor device 110. In some implementations, the user interface module can also present information to the user audibly (e.g., using one or more speakers) and / or via tactile feedback (e.g., using one or more haptic generators, such as vibration generation).

[0046] In some implementations, a software application can be used to facilitate the performance of the tasks described herein. As an example, an application can be installed on the electronic device 120. Further, a user can interact with the application to input data and / or commands to the electronic device 120 and to review data generated by the electronic device 120.

[0047] Figure 2 is a processor-based electronic device (e.g., Figure 1 the electronic device 120 of the data integrity maintenance system 100 of Figure 5 an example implementation 200 of software or algorithms utilized by the computing device (which can also be a server) of the system 500 of

[0048] The example implementation 200 illustrates a data store 210 and a measurement unit determination software 220.

[0049] The data store 210 can include or correspond to a data store of an electronic device (which can also be a server). For example, the data store 210 can be the database module 122 of the electronic device 120 and one or more storage devices 530 of the computing device of the system 500 (which can also be a server). The data store 210 can be in data communication with the electronic device (which can also be a server).

[0050] The data store 210 can include one or more first data structures 212 and one or more second data structures 216. The first data structures 212 can include or correspond to existing healthcare data (e.g., historical data, prior clinical trial data, real-time data, and other existing healthcare data). For example, such existing healthcare data can be used by the electronic device to generate a reference measurement range that is used to [1] determine whether a second data set (e.g., laboratory test data) of the second data structures 216 is missing a measurement unit or exhibits a mislabeling error, and [2] determine a correct measurement unit for corresponding laboratory test data. For example, the first data structures 212 can include first data sets for entities (e.g., individuals, users, patients, or subjects, etc.). The respective first data sets represent measurements of one or more physiological parameters 213 of the entities and one or more health-related attributes 214 of the entities. For example, the one or more physiological parameters can represent clinical parameters that are continuously collected at regularly spaced intervals. For example, the one or more physiological parameters can represent one or more vital signs such as glucose level, heart rate, blood pressure, respiratory rate, or temperature, etc. For example, the one or more health-related attributes can include at least one of a disease indication, a medical condition other than a disease indication, a same medication use, a same medical treatment, and a gender, etc.

[0051] The second data structures 216 can include or correspond to laboratory test data that is subject to a data integrity evaluation (e.g., to evaluate whether measurements are missing a measurement unit or exhibit a mislabeling error). For example, the second data structures 216 can include second data sets for target entities. For example, the respective second data sets represent measurements of one or more physiological parameters 217 of the target entities and one or more health-related attributes 218 of the target entities.

[0052] In some implementations, the first data structures 212 and the second data structures 216 can be stored in different data stores. For example, the first data structures 212 can be stored in a separate server (e.g., the computing device of the system 500 (which can also be a server)) while the second data structures 216 can be stored in a data store of the electronic device (e.g., assuming that the example electronic device does not take the form of a server).

[0053] Furthermore, at least some of the data filtering or unit detection can be implemented as corresponding software programs executable on an electronic device. The software program may include machine-readable instructions, which can be stored in memory (such as...). Figure 1 Database module 122, storage 520, Figure 5 The measurement unit determination software 220 contains (one or more) storage devices 530 and, when executed by a processor, causes a processor-based electronic device to execute instructions of a software program. As shown, the measurement unit determination software 220 may include a data filtering tool 230 and / or a unit determination tool 240. In some implementations, the measurement unit determination software 220 may include more or fewer tools. In some implementations, some of the tools may be combined, some of the tools may be broken down into more tools, or combinations thereof. In some implementations, the measurement unit determination software 220 may run on a server (e.g., the computing device of system 500, which may also be a server) or on both a server and an electronic device.

[0054] In some implementations, the data filtering tool 230 may take the form of a different software than the measurement unit determination software 220 and run on a server, while the unit determination tool 240 may take the form of the measurement unit determination software 220 and run on an electronic device that communicates data with the server. Other variations are possible where the data filtering tool 230 and the reference measurement range generation tool 232 are separate software and run on an electronic device, a server, or a combination thereof.

[0055] Data filtering tool 230 includes a reference measurement range generation tool 232. For example, data filtering tool 230 may first filter such existing healthcare data by identifying or extracting a first subset of data based on one or more health-related attributes 214 that match one or more health-related attributes 218 of the target entity. After identifying or extracting the first subset of data based on (one or more) health-related attributes 214 and 218, reference measurement range generation tool 232 may be used to generate reference measurement ranges. For example, generating reference measurement ranges includes [1] comparing the physiological parameters of the first subset of data to each other, and [2] converting the different units of measurement of the physiological parameters of the first subset into one or more units of measurement. Figure 3 and Figure 4 More detailed processing is described in the example processing for 300 and 400.

[0056] The unit determination tool 240 can determine the measurement unit(s) for the second data set (e.g., laboratory data) of the second data structure 216 that lacks a measurement unit or exhibits mislabeling errors based on the reference measurement range. Further, the determination of the measurement unit(s) can be preceded by a determination of whether the second data set of the second data structure 216 lacks a measurement unit or exhibits mislabeling errors. In Figure 3 and Figure 4 More detailed processing is described in the example processes 300 and 400.

[0057] Example processes

[0058] Figure 3 is a flowchart of an example process 300 for determining measurement units for laboratory test data. In particular, the example process 300 [1] filters existing healthcare data (e.g., historical data, prior clinical trial data, real-time data, and other existing healthcare data, etc.) by first determining or extracting a first subset of data such that the first subset of data focuses on a common health-related attribute(s), [2] generates a reference measurement range from the first subset of data, and [3] determines measurement units for the laboratory test data that lack a measurement unit or exhibit mislabeling errors based on the reference measurement range. As described in this disclosure, the process 300 can be implemented by a processor-based system, such as the data integrity maintenance system 100 and the system 500, and in conjunction with the example implementation 200.

[0059] At 302, one or more first data structures including a first data set are accessed. For example, a processor-based electronic device (e.g., the electronic device 120 of the data integrity maintenance system 100 or the computing device of the system 500) can be used to access the one or more first data structures. For example, the one or more first data structures can correspond to existing healthcare data (e.g., historical data, prior clinical trial data, real-time data, other existing healthcare data, etc.) stored in a data store (e.g., the data store 210). For example, the data store can correspond to a database module 122 of the electronic device 120 or a storage device (e.g., one or more storage devices 530 of the computing device including a server) of the computing device. The one or more first data structures can include a plurality of first data sets for a plurality of entities (e.g., individuals, users, patients, or subjects, etc.) and one or more health-related attributes for the plurality of entities. The first data sets can represent measured values for one or more physiological parameters for the plurality of entities. For example, the one or more physiological parameters can represent one or more vital signs, such as a glucose level, a heart rate, a blood pressure, a respiratory rate, or a temperature, etc. Figure 5

[0060] ​At 304, a first subset of the first data set is determined based on the one or more health-related attributes. For example, the electronic device can extract or determine the first subset of the first data set based on the one or more health-related attributes. For example, the one or more health-related attributes can include at least one of a disease indication, a medical condition other than the disease indication, a same medication usage, a same medical treatment, and a gender, and the like. For example, the first subset can represent measured values of one or more physiological parameters of a plurality of entities sharing a common health-related attribute(s). For example, the one or more physiological parameters of the first subset can represent outcome variables that measure an effect of a same medication or a same medical treatment. For example, for illustrative purposes, the entities corresponding to the first subset can correspond to a group of people who are taking or have taken a same medication or who are receiving or have received a same medical treatment. For example, the plurality of entities of the first subset can correspond to a group of people exhibiting a same disease indication.

[0061] At 306, one or more reference measurement ranges are generated. For example, the electronic device can generate the one or more reference measurement ranges by comparing values (e.g., measured values) of the one or more physiological parameters of the first subset of the first data set to each other and converting different measurement units of the one or more physiological parameters of the first subset to one or more measurement units. For example, for illustrative purposes, when the physiological parameter corresponds to a blood glucose level and the health-related attribute is a certain medication taken by patients (corresponding to the first subset), then within the first subset, there can be many different values (e.g., measured values) with many different measurement units such as mmol / L (millimoles per liter) or mg / dL (milligrams per deciliter). The electronic device can convert (or normalize) these different values with different measurement units to one or more single measurement units and generate the reference measurement ranges. For example, for blood glucose levels, all values can be normalized to mmol / L or mg / dL and the reference measurement ranges can be generated in units of mmol / L, mg / dL, or both.

[0062] At 308, one or more second data structures are accessed. For example, the electronic device can be configured to access the one or more second data structures. For example, the one or more second data structures include one or more second data sets pertaining to one or more target entities (e.g., target patients, target users, or target subjects, and the like). For example, the one or more second data sets can correspond to laboratory data or target healthcare data that is subject to a data integrity assessment.

[0063] For example, the one or more second datasets represent measured values of one or more physiological parameters of a target entity that exhibit or share the same one or more health-related attributes as the first subset of the first dataset from which the reference measurement range is generated. For example, for illustrative purposes, the target entity corresponding to the one or more second datasets can correspond to a group of people who take or have taken the same medication or who receive or have received the same medical treatment. For example, the physiological parameter of the one or more target entities can correspond to a blood glucose level, and the health-related attribute can be a particular medication taken by the target entity (where the particular medication is the same medication taken or has been taken by the entities of the first subset of the first dataset).

[0064] For example, the one or more second datasets regarding the one or more target entities can be subject to an assessment of data completeness. For example, the measured value(s) of the physiological parameter of the target entity can be missing or exhibit mislabeling errors, and the corresponding data, such as the one or more second datasets of laboratory data, can be subject to the assessment of data completeness. For example, the determination of whether the measured value(s) are missing or whether there are mislabeling errors can be made at 310 as part of the measured unit(s) determination step.

[0065] At 310, the measured unit(s) of the one or more measurements of the physiological parameter of the target entity that are missing measured units or exhibit mislabeling errors are determined. For example, the measured unit(s) can be determined based on the reference measurement range.

[0066] For example, determining the measured unit(s) can include: [1] first comparing one or more values of one or more measurements of the physiological parameter of the target entity to the reference measurement range, [2] determining that one or more measurements exhibit mislabeling errors or are missing measured units based on the comparison, and [3] determining the correct measured unit(s) of the one or more measurements.

[0067] In some implementations, a score can be generated based on a comparison of one or more measurement values to a reference measurement range. Thereafter, such a score can be compared to a threshold score to determine that one or more measurements lack a measurement unit or exhibit a mislabeling error. For example, measurement values of a target entity for a laboratory test data can be compared to reference measurement ranges having respective measurement units, and each comparison (of a reference measurement range in respective reference units) can generate a score. Once a score is generated, the score can be compared to a threshold score, and the electronic device can determine whether one or more measurement values lack a measurement unit or exhibit a mislabeling error.

[0068] In some implementations, determining that one or more measurements exhibit a mislabeling error can include determining a frequency of one or more values of the measurement(s) that fall within a reference measurement range, and determining a mislabeling error based on the frequency. For example, for illustrative purposes, when a physiological parameter is a blood glucose level and a reference range is 70-100 mg / dL, blood glucose values of a target entity can form a frequency cluster that falls within the reference range or within a threshold (e.g., ±20% of the maximum or minimum value of the reference range) of the reference range (as depicted by the frequency of one or more measurements 602 in FIG. 6B). Thus, values that fall outside of the reference range or the threshold (e.g., ±20% of the maximum or minimum value of the reference range) can be considered data points that exhibit a mislabeling error. Figure 6

[0069] In some implementations, a machine learning classifier can be used to determine a measurement unit of one or more measurements that lack a measurement unit or exhibit a mislabeling error with respect to a measurement unit. For example, a machine learning classifier can be trained based on a z-score probability and one or more of frequency features extracted from one or more first data structures (e.g., existing healthcare data such as historical data). The frequency features include a frequency of units by study, a frequency of units by site, and a frequency of units by subject. For example, the frequency features are described in Table 1 below. For example, a logistic regression model can be used to predict a binary outcome indicating a likelihood that one or more measurements are measured in one or more respective measurement units based on the z-score probability derived from a reference measurement range and the frequency features. For example, a probability outcome from the classifier can serve as another threshold for detecting mislabeled units, which helps to reduce a false positive rate.

[0070] Feature Description z-score probability Inferred probability of value being in the reference measurement range for the labeled unit Unit frequency by study Proportion of records labeled with the same unit within the same study, range = [0, 1] Unit frequency by site Proportion of records labeled with the same unit within the same site, range = [0, 1] Unit frequency by subject Proportion of records labeled with the same unit within the same subject, range = [0, 1]

[0071] Table 1: Features used in machine learning classifier for unit determination or prediction

[0072] ​At 312, a third data structure representing the determined unit of measure is generated. For example, the electronic device can format the data into a standardized data format, such as JavaScript Object Notation (JSON) or Extensible Markup Language (XML) format. The standardized data format can be other suitable data format suitable for storage in a data store or hardware storage device.

[0073] At 314, the third data structure is stored in a hardware storage device. For example, the data structure can be stored in a data store (e.g., data store 210, database module 122 of electronic device 120, one or more storage devices 530, etc.).

[0074] In some implementations, instead of or in addition to storing the data structure in a hardware storage device, the data structure can be output to a user interface (e.g., using user interface module 128). For example, in the case that the data structure is output for user display without storing the data structure, the data structure can be generated at step 312 in other formats suitable for display. For example, in the case that the data structure is output for user display after storing the data structure, the data structure can be converted from a storage format to an appropriate display format and output.

[0075] In some implementations, the third data structure can be provided or transmitted to a computerized health analysis platform.

[0076] Figure 4 is a flowchart of an example process 400 for determining a unit of measure for laboratory test data and modifying the laboratory test data. In particular, the example process 400 [1] determines a unit of measure for laboratory test data that is missing a unit of measure or exhibits a mislabeling error, and [2] modifies one or more second data sets (of the second data structure). As described in this disclosure, the process 400 can be implemented by a processor-based system, such as the data integrity maintenance system 100 and the system 500, and in conjunction with the example implementation 200 and the example process 300.

[0077] At 402, one or more second data structures are accessed. As the techniques used in step 402 can be the same as the techniques used in step 308 of the example process 300, the techniques are not repeated here. Figure 3

[0078] At 404, a unit of measure(s) for one or more measurements of a physiological parameter of a target entity that is missing a unit of measure or exhibits a mislabeling error is determined. For example, the unit of measure(s) can be determined based on a reference measurement range. As the techniques used in step 404 can be the same as the techniques used in step 310 of the example process 300, the techniques are not repeated here. Figure 3 ​The techniques used in step 310 of FIG. 3B are the same as those used in step 310 of FIG. 3A, and thus are not repeated here.

[0079] At 406, a third data structure is generated that represents the determined units of measurement. The techniques used in step 406 can be the same as those used in step 406 of FIG. 4A, and thus are not repeated here. Figure 3 The techniques used in step 312 of FIG. 3B are the same as those used in step 312 of FIG. 3A, and thus are not repeated here.

[0080] At 408, the units of measurement are output for display (e.g., using the user interface module 128) on a user interface. For example, a plot representing the first subset of time series data can be generated for display on the user interface. For example, an indication of whether one or more measurement results exhibit mislabeling errors or lack of units of measurement is generated on a display of the user interface. For example, a frequency plot representing a relationship between the frequency of one or more measurement results and the plurality of units of measurement is generated on a display of the user interface.

[0081] For example, Figure 6 is a frequency plot 600 illustrating a relationship between the frequency of one or more measurement results 602 and a plurality of units of measurement 604. For example, the x-axis depicts the plurality of units of measurement 605, and the y-axis depicts actual measured values of a physiological parameter. For example, the shaded regions 606 represent a plurality of reference measurement ranges in a plurality of units. Further, for example, the first column 608 illustrates a set of points or measurement values that lack units of measurement. For example, the implementations described in this disclosure (including the processes 300 and 400) have determined, based on the reference measurement ranges of the respective units of measurement as shown by the respective shaded regions in the shaded regions 606, what the units of measurement should be for these measurement values that lack units of measurement in the first column 608. On the other hand, when data points [1] are not within the primary shaded region (e.g., the region with the highest frequency) to which most of the data points belong, but instead fall within a different shaded region within the same unit of measurement category on the x-axis, these data points can exhibit mislabeling errors. In this case, for example, the units of measurement corresponding to the different shaded region in which these data points (outside of the primary shaded region) were captured can be the correct units of measurement for these data points.

[0082] In some implementations, when one or more measurement results exhibit mislabeling errors or lack of units of measurement, the data points corresponding to the one or more measurement results can be subjected to review and flagged as records or data points that require review or validation.

[0083] At 410, one or more second data sets of the second data structure are modified. For example, the second data sets can be modified based on the determined measurement units. For example, the modification includes [1] incorporating the determined measurement units into one or more measurement results that lack measurement units, or [2] replacing respective measurement units of one or more measurement results that exhibit mislabeling errors with the determined measurement units. Through the modification, data quality of the one or more second data sets can be improved by preventing processing of incomplete or erroneous second data sets. In some implementations, such modification can be made after the labeled records or data points that need to be checked or verified have been checked or verified.

[0084] In some implementations, the modified second data sets or the second data structure can be provided or transmitted to the computerized health analytics platform.

[0085] Example computer system

[0086] Figure 5 An example computing system in accordance with implementations of the present disclosure is depicted. The system 500 can be used for any of the operations described with respect to the various implementations discussed herein. The system 500 can be included in, used by, in communication with, or correspond to the electronic device 120. Further, the system 500 can include, be used by, or be in communication with the sensor device 110. The system 500 can include one or more processors 510, a memory 520, one or more storage devices 530, and one or more input / output (I / O) devices 560, which can be controlled by one or more I / O interfaces 540. The various components 510, 520, 530, 540, or 560 can be interconnected by at least one system bus 550, which can enable communication of data among the various modules and components of the system 500.

[0087] The processor(s) 510 can be configured to process instructions for execution within the system 500. The processor(s) 510 can include single-threaded processor(s), multi-threaded processor(s), or both. The processor(s) 510 can be configured to process instructions stored in the memory 520 or on the storage device(s) 530. The processor(s) 510 can include hardware-based processor(s) each including one or more cores. The processor(s) 510 can include general-purpose processor(s), special-purpose processor(s), or both.

[0088] Memory 520 can store information within system 500. In some implementations, memory 520 includes one or more computer-readable media. Memory 520 can include any type of long-term, short-term, or both physical storage of information. In some examples, memory 520 can be used to store program instructions for execution by one or more of the processors 510. Memory 520 can include only memories, or only storage components, or both, and can take any

[0089] Storage device(s) 530 can be configured to provide mass storage for system 500. In some implementations, storage device(s) 530 can include one or more computer- readable media. Storage device(s) 530 can include removable media, such as floppy diskettes, ZIP® disks, compact diskettes (CDs) or digital versatile disks (DVDs), or hard disk drives. Storage device(s) 530 can include drives, such as optical drives, hard drives, or removable media drives. Storage device(s) 530 can include one or more of the same types of media used by memory 520 or one or more different types of media.

[0090] One or both of memory 520 or storage device(s) 530 can include one or more computer-readable storage media (CRSM). CRSM can include one or more of volatile memory, non-volatile memory, flash memory, or any other storage media. CRSM can provide storage of computer-readable instructions describing the data structures, processes, applications, programs, other modules, or other data for system 500. In some implementations, CRSM can include data storage that provides storage of computer-readable instructions or other information in a non-transitory format. CRSM can be incorporated into system 500 or external to system 500. CRSM can include read only memory, random access memory, or both. One or more CRSM suitable for embodying computer program instructions and data can include any type of non-volatile storage, including, but not limited to: semiconductor memory devices, such as EPROM, EEPROM, and flash memory devices; magnetic disks, such as internal hard disks and removable disks; magneto-optical disks; and CD-ROM and DVD-ROM disks. In some examples, processors 510 and memory 520 can be supplemented by, or incorporated into, one or more application-specific integrated circuits (ASICs).

[0091] System 500 can include one or more I / O devices 560. I / O device(s) 560 can include one or more input devices, such as a keyboard, a mouse, a pen, a game controller, a touch input device, an audio input device (e.g., a microphone), a gesture input device, a haptic input device, an image or video capture device (e.g., a camera), or other devices. In some examples, I / O device(s) 560 can also include one or more output devices, such as a display, LED(s), an audio output device (e.g., a speaker), a printer, and a haptic output device. I / O device(s) 560 can be physically incorporated into one or more computing devices of system 500, or can be external with respect to one or more computing devices of system 500.

[0092] System 500 can include one or more I / O interfaces 540 to enable components or modules of system 500 to control, interact with, or otherwise communicate with I / O device(s) 560. I / O interface(s) 540 can enable information to be communicated in or out of system 500 or between components of system 500 through serial, parallel, or other types of communication. For example, I / O interface(s) 540 can conform to a version of the RS-232 standard for serial ports, or a version of the IEEE 1284 standard for parallel ports. As another example, I / O interface(s) 540 can be configured to provide connectivity through a universal serial bus (USB) or Ethernet. In some examples, I / O interface(s) 540 can be configured to provide serial connectivity conforming to a version of the IEEE 1394 standard.

[0093] I / O interface(s) 540 can also include one or more network interfaces that enable communication between computing devices in system 500, or between system 500 and other network-connected computing systems. Network interface(s) can include one or more network interface controllers (NICs) or other types of transceiver devices configured to send and receive communications over one or more networks using any network protocol.

[0094] The computing devices of system 500 can communicate with each other or with other computing devices using one or more networks. Such networks can include public networks such as the Internet, private networks such as an institutional or personal intranet, or any combination of private and public networks. Networks can include any type of wired or wireless network, including but not limited to local area networks (LANs), wide area networks (WANs), wireless local area networks (WWANs), wireless LANs (WLANs), and mobile telecommunications networks (e.g., 3G, 4G, Edge, etc.) and the like. In some implementations, communications between computing devices can be encrypted or otherwise secured. For example, communications can employ one or more public or private encryption keys, passwords, digital certificates, or other credentials supported by a security protocol such as any version of the Secure Sockets Layer (SSL) or Transport Layer Security (TLS) protocol.

[0095] System 500 can include any number of any type of computing device. The computing device(s) can include, but are not limited to, personal computers, smart phones, tablet computers, wearable computers, implantable computers, mobile gaming devices, e-book readers, automobile computers, desktop computers, laptop computers, notebook computers, game consoles, home entertainment devices, network computers, server computers, mainframe computers, distributed computing devices (e.g., cloud computing devices), microcomputers, system on a chip (SoC), and system in a package (SiP), and the like. Although examples herein can describe the computing device(s) as physical device(s), implementations are not so limited. In some examples, a computing device can include one or more of a virtual computing environment, hypervisor, emulation, or virtual machine executing on one or more physical computing devices. In some examples, two or more computing devices can include a cluster, cloud, farm, or other grouping of multiple devices that coordinate operation to provide load balancing, failover support, parallel processing capability, shared storage resources, shared networking capability, or other aspects.

[0096] This specification uses the term“configured to” in connection with systems and computer program components. For a system of one or more computers to be configured to perform particular operations or actions, that system has installed on it software, firmware, hardware, or a combination of them that in operation causes the system to perform the operations or actions. For one or more computer programs to be configured to perform particular operations or actions, that one or more programs include instructions that in operation cause a data processing apparatus to perform the operations or actions.

[0097] Embodiments of the subject matter and the functional operations described in this specification can be implemented in digital electronic circuitry, in tangibly-embodied computer software or firmware, in computer hardware, including the structures disclosed in this specification and their structural equivalents, or in combinations of one or more of them. Embodiments of the subject matter described in this specification can be implemented as one or more computer programs, i.e., one or more modules of computer program instructions encoded on a tangible non-transitory storage medium for execution by, or to control the operation of, data processing apparatus. The computer storage medium can be a machine-readable storage device, a machine-readable storage substrate, a random or serial access memory device, or a combination of one or more of them. Alternatively or additionally, the program instructions can be encoded on an artificially generated propagated signal, e.g., a machine-generated electrical, optical, or electromagnetic signal, that is generated to encode information for transmission to suitable receiver apparatus for execution by a data processing apparatus.

[0098] The term“data processing apparatus” refers to data processing hardware and encompasses all kinds of apparatus, devices, and machines for processing data, including by way of example a programmable processor, a computer, or multiple processors or computers. The apparatus can also be, or further include, special purpose logic circuitry, e.g., an FPGA (field programmable gate array) or an ASIC (application specific integrated circuit). The apparatus can optionally include, in addition to hardware, code that creates an execution environment for computer programs, e.g., code that constitutes processor firmware, a protocol stack, a database management system, an operating system, or a combination of one or more of them.

[0099] A computer program (which can also be referred to or described as a program, software, a software application, an app, a module, a software module, a script, or code) can be written in any form of programming language, including compiled or interpreted languages, or declarative or procedural languages; and it can be deployed in any form, including as a stand-alone program or as a module, component, subroutine, or other unit suitable for use in a computing environment. A program may, but need not, correspond to a file in a file system. A program can be stored in a portion of a file that holds other programs or data (e.g., one or more scripts stored in a markup language document), in a single file dedicated to the program in question, or in multiple coordinated files (e.g., files that store one or more modules, sub programs, or portions of code). A computer program can be deployed to be executed on one computer or on multiple computers that are located at one site or distributed across multiple sites and

[0100] In this specification, the term“database” is used broadly to refer to any collection of data: the data need not be structured in any particular way, or structured at all, and need not be stored using a database management system. Thus, for example, an index database can include a plurality of collections of data, each of which can be organized and accessed differently.

[0101] The processes and logic flows described in this specification can be performed by one or more programmable computers executing one or more computer programs to perform functions by operating on input data and generating output. The processes and logic flows can also be performed by special purpose logic circuitry, e.g., an FPGA or an ASIC, or by a combination of special purpose logic circuitry and one or more programmed computers.

[0102] Computers suitable for the execution of a computer program can be based on general or special purpose microprocessors or any other kind of central processing unit. Generally, a central processing unit will receive instructions and data from a read-only memory or a random access memory or both. The essential elements of a computer are a central processing unit for performing or executing instructions

[0103] Computer-readable media suitable for storing computer program instructions and data include all forms of non-volatile memory, media and memory devices, including by way of example semiconductor memory devices, e.g., EPROM, EEPROM, and flash memory devices; magnetic disks, e.g., internal hard disks or removable disks; magneto-optical disks; and CD-ROM and DVD-ROM disks.

[0104] The term“memory subsystem” can include one or more memories, where each memory can be a computer-readable medium. The memory subsystem can encompass a memory hardware unit (e.g., a hard drive or disk) that stores data or instructions in software form. Alternatively or additionally, the memory subsystem can include data or instructions hardwired into the processing circuitry.

[0105] To provide for interaction with a user, embodiments of the subject matter described in this specification can be implemented on a computer having a display device, e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor, for displaying information to the user and a keyboard and a pointing device, e.g., a mouse or a trackball, by which the user can provide input to the computer. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback, e.g., visual feedback, auditory feedback, or tactile feedback; and input from the user can be received in any form, including acoustic, speech, or tactile input. In addition, a computer can interact with a user by sending documents to and receiving documents from a device that is used by the user; for example, by sending web pages to a web browser on a user’s device in response to requests received from the web browser. Also, a computer can interact with a user by sending text messages or other forms of message to a personal device, e.g., a smartphone that is running a messaging application, and receiving responsive messages from the user in return.

[0106] Embodiments of the subject matter described in this specification can be implemented in a computing system that includes a back-end component, e.g., as a data server, or that includes a middleware component, e.g., an application server, or that includes a front-end component, e.g., a client computer having a graphical user interface, a web browser, or an app through which a user can interact with an implementation of the subject matter described in this specification, or any combination of one or more such back-end, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication, e.g., a communication network. Examples of communication networks include a local area network (LAN) and a wide area network (WAN), e.g., the Internet.

[0107] The computing system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. In some embodiments, a server transmits data, e.g., an HTML page, to a user device, e.g., for purposes of displaying data to and receiving user input from a user interacting with the device that is

[0108] While this specification contains many specifics, these should not be construed as limitations on the scope of any invention, but rather as descriptions of particular implementations of particular embodiments thereof. Certain features that are described in this specification in the context of separate embodiments can also be implemented in combination in a single embodiment. Conversely, various features that are described in the context of a single embodiment can also be implemented separately or in any suitable subcombination. Moreover, although features can be described above as acting in certain combinations and even initially claimed as such, one or more features from a claimed combination can in some cases be excised from the combination and the claimed combination can be directed to a subcombination or variation of a subcombination.

[0109] Similarly, while operations are depicted in the drawings and recited in the claims in a particular order, this should not be understood as requiring such order nor

[0110] Particular embodiments of the subject matter have been described. Other embodiments are within the scope of the following claims. For example, the acts recited in the claims can be performed in a different order and still achieve desirable results. As one example, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve the desired results. In some instances, multitasking and parallel processing can be advantageous.

Claims

1. A system for maintaining data integrity in a computerized health analytics platform, the system comprising: at least one processor; and a memory subsystem communicatively coupled to the at least one processor, the memory subsystem storing instructions that, when executed by the at least one processor, cause the at least one processor to perform operations comprising: accessing one or more first data structures comprising: a plurality of first data sets regarding a plurality of entities, wherein each of the first data sets represents measurements of one or more physiological parameters of the plurality of entities, wherein a machine transmitting the data structure determines that the system accessing the data structure is authorized to access the data structure, and one or more health-related attributes of the plurality of entities; determining a first subset of the first data sets based on the one or more health-related attributes of the plurality of entities; generating a reference measurement range for the one or more physiological parameters of a plurality of entities of the first subset of the first data sets in one or more measurement units; accessing one or more second data structures comprising: one or more second data sets regarding one or more target entities, wherein the one or more second data sets represent one or more measurements of the one or more physiological parameters of the one or more target entities, wherein the one or more target entities exhibit the one or more health-related attributes, and wherein the one or more measurements of the one or more target entities lack measurement units or exhibit mislabeling errors regarding the measurement units; determining measurement units for the one or more measurements that lack the measurement units or exhibit the mislabeling errors regarding the measurement units based on the reference measurement range; generating a third data structure representing the measurement units; and storing the third data structure in a hardware storage device.

2. The system of claim 1, wherein, the operations further comprising providing the third data structure to the computerized health analytics platform.

3. The system of claim 1, wherein, the one or more health-related attributes comprise at least one of a disease indication, a medical condition other than the disease indication, a same medication use, a same medical treatment, and a gender.

4. The system of claim 1, wherein, the one or more physiological parameters represent clinical parameters collected continuously at regularly spaced intervals.

5. The system of claim 1, wherein, generating the reference measurement range comprises: comparing measurement values of one or more physiological parameters of the first subset of the first data sets to each other; and converting different measurement units of the one or more physiological parameters of the first subset to the one or more measurement units.

6. The system of claim 5, wherein, determining measurement units for the one or more measurements that lack the measurement units or exhibit the mislabeling errors regarding the measurement units comprises: comparing one or more values of the one or more measurements to the reference measurement range; and determining that the one or more measurements exhibit the mislabeling error or lack the measurement unit based on the comparison.

7. The system of claim 6, wherein, determining that the one or more measurements exhibit the mislabeling error includes: determining a frequency of one or more values of the one or more measurements that fall within the reference measurement range, and determining the mislabeling error based on the frequency.

8. The system of claim 6, wherein, determining the measurement unit includes: generating a score based on a comparison of one or more values of the one or more measurements to the reference measurement range.

9. The system of claim 8, wherein, the one or more measurements exhibit the mislabeling error, and wherein determining the measurement unit includes: determining that the one or more measurements exhibit the mislabeling error based on a comparison of the score to a threshold score.

10. The system of claim 1, wherein, the operations further include: outputting the measurement unit to a display of a user interface.

11. The system of claim 10, wherein, the operations further include: generating an indication on the display of the user interface of whether one or more measurements exhibit the mislabeling error or lack the measurement unit.

12. The system of claim 11, wherein, the operations further include: generating a frequency plot representing a relationship between a frequency of the one or more measurements and a plurality of measurement units.

13. The system of claim 1, wherein, determining the measurement unit for the one or more measurements that lack the measurement unit or exhibit a mislabeling error with respect to the measurement unit includes: determining the measurement unit with a machine learning classifier, wherein the machine learning classifier is trained based on one or more of a z-score probability and a frequency feature, wherein the frequency feature includes a unit frequency per study, a unit frequency per site, and a unit frequency per subject, and wherein determining the measurement unit with a machine learning classifier includes predicting a binary outcome based on the z-score probability derived from the reference measurement range and the frequency feature with a logistic regression model, the binary outcome indicating a likelihood that the one or more measurements were measured in one or more respective measurement units.

14. The system of claim 1, wherein, the operations further include: modifying the one or more second data sets based on the third data structure, wherein modifying the one or more second data sets improves data integrity by preventing processing of incomplete or erroneous second data sets, and wherein the measurement unit is (i) incorporated into the one or more measurements that lack the measurement unit or (ii) replaces a respective measurement unit of the one or more measurements that exhibit the mislabeling error.

15. A method for maintaining data integrity in a computerized health analytics platform, comprising: accessing, by an electronic device, one or more first data structures, the one or more first data structures including: a plurality of first data sets for a plurality of entities, wherein each first data set of the first data sets represents measurements of one or more physiological parameters of the plurality of entities, and one or more health-related attributes of the plurality of entities; determining, by the electronic device, a first subset of the first data sets based on one or more health-related attributes of the plurality of entities; generating, by the electronic device and in one or more measurement units, a reference measurement range for a physiological parameter of the plurality of entities of the first subset of the first data sets; accessing, by the electronic device, one or more second data structures, the second data structures comprising: one or more second data sets for one or more target entities, wherein the one or more second data sets represent one or more measurements of one or more physiological parameters of the one or more target entities, wherein the one or more target entities exhibit the one or more health-related attributes, and wherein the one or more measurements lack measurement units or exhibit mislabeling errors with respect to the measurement units; determining, by the electronic device, measurement units for the one or more measurements that lack the measurement units or exhibit the mislabeling errors with respect to the measurement units; generating, by the electronic device, a third data structure representing the measurement units; and storing, by the electronic device, the third data structure in a hardware storage device.

16. The method of claim 15, wherein, determining measurement units for the one or more measurements that lack the measurement units or exhibit mislabeling errors with respect to the measurement units comprises: comparing one or more values of the one or more measurements to the reference measurement range; and determining, based on the comparison, that the one or more measurements exhibit the mislabeling errors or lack the measurement units.

17. The method of claim 16, wherein, determining that the one or more measurements exhibit the mislabeling errors comprises: determining a frequency of the one or more measurements falling within the reference measurement range; and determining the mislabeling errors based on the frequency.

18. The method of claim 16, wherein, determining the measurement units comprises: generating a score based on a comparison of one or more values of the one or more measurements to the reference measurement range, and wherein determining the measurement units comprises: determining that the one or more measurements lack the measurement units based on comparing the score to a threshold score.

19. The method of claim 15, further comprising: generating, by the electronic device, an indication on a display of a user interface of whether one or more measurements exhibit the mislabeling errors or lack the measurement units.

20. One or more non-transitory computer-readable media storing instructions that, when executed by at least one processor, cause the at least one processor to perform: accessing one or more first data structures, the one or more first data structures comprising: a plurality of first data sets for a plurality of entities, wherein each of the first data sets represents measurements of one or more physiological parameters of the plurality of entities, and one or more health-related attributes of the plurality of entities. determining a first subset of the first data set based on one or more health-related attributes of the plurality of entities; generating a reference measurement range for a physiological parameter of a plurality of entities of the first subset of the first data set in one or more measurement units; accessing one or more second data structures comprising: one or more second data sets on one or more target entities, wherein the one or more second data sets represent one or more measurements of one or more physiological parameters of the one or more target entities, wherein the one or more target entities exhibit the one or more health-related attributes, and wherein the one or more measurements lack a measurement unit or exhibit a mislabeling error with respect to the measurement unit; determining a measurement unit for the one or more measurements that lack the measurement unit or exhibit the mislabeling error with respect to the measurement unit; generating a third data structure representing the measurement unit; and storing the third data structure in a hardware storage.

21. A computer program product having stored instructions which, when executed by at least one processor, cause the at least one processor to perform the method of any one of claims 15-19.

Images