System and method for determining data quality of digital imaging and communication (DICOM) data in medicy

Abstract

A system (100) and method (300) are provided for determining a data quality of DICOM data (412, 416, 420). Information related to a data ingestion (402) of the DICOM data (412, 416, 420) from the radiological device (102) to a database (110) may be received (302). A respective data quality of the DICOM data (412, 416, 420) for a set of data ingestion stages (404, 406, 408, 410) of the data ingestion (402) may be determined (304) using a set of data verification rules (414, 418, 422) corresponding to a set of data ingestion stages (404, 406, 408, 410) of the data ingestion (402) from the radiation device (102) to the database (110) for the DICOM data (412, 416, 420). A user interface (700, 800) may be generated (306) that displays information relating to the respective data quality of the DICOM data (412, 416, 420) for the set of data ingestion phases (404, 406, 408, 410) of the data ingestion (402).

Description

Technical Field

[0001] The present disclosure generally relates to systems and methods for determining data quality of DICOM data. More specifically, the present disclosure relates to systems and methods for determining data quality of DICOM data at a set of data ingestion phases of data ingestion of the DICOM data from a radiology device to a database. Background Art

[0002] In modern healthcare systems, data is increasingly important. This can be equipment data, patient data, treatment data, imaging scan data, and other sources of healthcare-related data. The quality and usefulness of data are important for efficient, cost-effective, and accurate healthcare decisions. Data quality can refer to a measure of the data condition that identifies the suitability of data for a particular purpose. For example, the data quality of data can identify the accuracy, completeness, comprehensiveness, consistency, timeliness, uniqueness, validity, etc. of the data. In a medical context, a system of a healthcare institution (e.g., a hospital, clinic, nursing facility, etc.) can perform data ingestion of DICOM data from a radiological device that generates DICOM data to a database that stores DICOM data. The data quality of DICOM data may vary at different data ingestion stages of data ingestion, and / or may be affected at different data ingestion stages of data ingestion. For example, at the data generation stage, DICOM data may have low data quality based on incorrect configuration of a radiological device that is configured to generate DICOM data. As another example, during the data extraction stage, the data quality of DICOM data may be affected based on network connectivity issues of a healthcare institution. As another example, during the data processing stage, DICOM data may have reduced data quality based on incorrect configuration of a data ingestion application that processes DICOM data. A healthcare facility may have a large number of radiology equipment. Personnel at the hospital facility may attempt to assess the data quality of DICOM data from radiology equipment. However, this task may prove to be impossible, inefficient, error-prone, etc. Summary of the invention

[0003] This summary introduces concepts that are described in more detail in the detailed description. It should not be used to identify essential features of the claimed subject matter, nor should it be used to limit the scope of the claimed subject matter.

[0004] In one aspect, a method may include: receiving information related to data ingestion of Digital Imaging and Communications in Medicine (DICOM) data from a radiology device to a database; determining corresponding data quality of the DICOM data at the set of data ingestion stages of the data ingestion of the DICOM data from the radiology device to the database using a set of data validation rules corresponding to a set of data ingestion stages of the data ingestion of the DICOM data from the radiology device to the database, wherein the data validation rules include one or more of a first data validation rule that is common to multiple imaging modalities of a radiology device, a second data validation rule that is specific to the imaging modality of the radiology device, and a third data validation rule that is specific to a data ingestion application associated with the data ingestion; and generating a user interface based on determining the corresponding data quality of the DICOM data at the set of data ingestion stages of the data ingestion, the user interface displaying information related to the corresponding data quality of the DICOM data at the set of data ingestion stages of the data ingestion.

[0005] In another aspect, a device may include a memory configured to store instructions; and one or more processors configured to execute instructions to: receive information related to data ingestion of Digital Imaging and Communications in Medicine (DICOM) data from a radiology device to a database; determine a corresponding data quality of the DICOM data at the set of data ingestion stages of the data ingestion of the DICOM data from the radiology device to the database using a set of data validation rules corresponding to a set of data ingestion stages of the data ingestion of the DICOM data from the radiology device to the database, wherein the data validation rules include a first data validation rule that is common to multiple imaging modalities of the radiology device, a second data validation rule that is specific to the imaging modality of the radiology device, and a third data validation rule that is specific to a data ingestion application associated with the data ingestion; and generate a user interface based on determining the corresponding data quality of the DICOM data at the set of data ingestion stages of the data ingestion, the user interface displaying information related to the corresponding data quality of the DICOM data at the set of data ingestion stages of the data ingestion.

[0006] On the other hand, a non-transitory computer-readable medium may store instructions that, when executed by one or more processors, cause the one or more processors to perform the following operations: receive information related to data ingestion of Digital Imaging and Communications in Medicine (DICOM) data from a radiology device to a database; determine the corresponding data quality of the DICOM data at the set of data ingestion stages of the data ingestion of the DICOM data from the radiology device to the database using a set of data validation rules corresponding to a set of data ingestion stages of the data ingestion of the DICOM data from the radiology device to the database, wherein the data validation rules include a first data validation rule that is common to multiple imaging modalities of the radiology device, a second data validation rule that is specific to the imaging modality of the radiology device, and a third data validation rule that is specific to a data ingestion application associated with the data ingestion; and generate a user interface based on determining the corresponding data quality of the DICOM data at the set of data ingestion stages of the data ingestion, the user interface displaying information related to the corresponding data quality of the DICOM data at the set of data ingestion stages of the data ingestion.

[0007] Some embodiments of the present disclosure provide technical improvements in the field of radiology technology and to equipment associated with acquisition, acquisition, and use of DICOM data by radiology equipment by determining data quality of DICOM data at a set of data acquisition stages of data acquisition of DICOM data. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 is a diagram of an example system for determining data quality of DICOM data.

[0009] Figure 2 yes Figure 1 An illustration of example components of a device.

[0010] Figure 3 is an illustration of a flow chart of an example process for determining data quality of DICOM data using a set of validation rules corresponding to a set of data ingestion phases of data ingestion of the DICOM data from a radiology device into a database.

[0011] Figure 4 is an illustration of an example process for determining data quality of DICOM data using a set of validation rules corresponding to a set of data ingestion phases of data ingestion of the DICOM data from a radiology device into a database.

[0012] Figure 5 is a flow chart of an example process for determining data quality of DICOM data.

[0013] Figure 6is a diagram of example validation rules for determining data quality for DICOM.

[0014] Figure 7 is an illustration of an example user interface for displaying data quality of DICOM data associated with a radiology device.

[0015] Figure 8 is an illustration of an example user interface for displaying corresponding data quality of DICOM data corresponding to a validation rule.

[0016] Fig. 9 is a flow chart of an example process for determining the root cause of data quality not meeting a threshold.

[0017] Fig.10 is a flow chart of an example process for performing actions to improve data quality of DICOM data.

[0018] Fig.11 is a diagram of an example process for training an artificial intelligence (AI) model. DETAILED DESCRIPTION

[0019] Embodiments of the present disclosure will now be described by way of example with reference to the accompanying drawings.

[0020] As described above, the task of assessing data quality across a large number of radiology devices may prove impossible, inefficient, error-prone, etc. The problem is further exacerbated by the fact that radiology devices generate DICOM data that has significant variability depending on the imaging modality, manufacturer, model, etc. of the respective radiology devices.

[0021] Some embodiments of the present disclosure provide a data quality platform that can receive information related to data ingestion of DICOM data from a radiology device to a database, and use a set of data validation rules corresponding to a set of data ingestion stages of the data ingestion of the DICOM data from the radiology device to the database to determine the corresponding data quality of the DICOM data at the set of data ingestion stages of the data ingestion. In addition, the data quality platform can be configured to generate a user interface that displays information related to the corresponding data quality of the DICOM data at the set of data ingestion stages of the data ingestion.

[0022] Some embodiments of the present disclosure provide technical improvements in the field of radiology technology and technical improvements to equipment associated with the acquisition, ingestion and use of DICOM data of radiology equipment by determining the data quality of DICOM data at a set of data ingestion stages of data ingestion of DICOM data. In the event that the data quality does not meet a threshold, some embodiments of the present disclosure provide identification of the root cause of the data quality not meeting the threshold and / or actions to improve the data quality of the DICOM data.

[0023] In this way, some embodiments of the present disclosure improve the security, efficacy, and reliability of DICOM data by ensuring adequate data quality, eliminating or reducing the need for human intervention in assessing data quality, accelerating the serviceability of radiology equipment, improving transparency in data ingestion systems, improving identification of root causes of issues using data lineage, automatically retaining DICOM data when data quality does not meet a threshold, automatically identifying issues associated with data ingestion, reducing the time required to connect, integrate, and verify radiology equipment, and improving the use of AI models trained using DICOM data.

[0024] Figure 1 is a diagram of an example system 100 for determining data quality of DICOM data. Figure 1 As shown, system 100 may include a radiology device 102 , a data source 104 , a data ingestion system 106 , a data ingestion application 108 , a database 110 , a data quality platform 112 , an AI model 114 , a user device 116 , and a network 118 .

[0025] The radiation device 102 may be configured to generate DICOM data using an imaging modality. For example, the radiation device 102 may be an ultrasound device, an X-ray device, an electrocardiogram (ECG) device, a computed tomography (CT) device, a magnetic resonance imaging (MRI) device, a positron emission tomography (PET) device, etc. The radiation device 102 may be associated with a specific manufacturer identified by a manufacturer identifier, may be associated with a specific model identified by a model identifier, etc. DICOM data may include a header and a data set. The data set may include data elements defined by the DICOM standard. For example, a data element may include a patient identifier, name, gender, age, procedure type, device type, procedure description, procedure code, pixel data of a medical image generated by the radiation device 102, etc. The data element may be associated with a corresponding tag.

[0026] The data source 104 may be configured to provide information associated with a patient, information associated with the radiation device 102 , information associated with the data ingestion system 106 , etc. For example, the data source 104 may be a server, a database, etc.

[0027] The data ingestion system 106 may be configured to perform data ingestion of DICOM data from the radiology device 102 to the database 110. For example, the data ingestion system 106 may be a data hub, a data lake, a data warehouse, a database, etc. The data ingestion application 108 may be configured to process the DICOM data associated with the data ingestion from the radiology device 102 to the database 110. For example, the data ingestion application 108 may be a data extraction application configured to perform data extraction using the DICOM data. As another example, the data ingestion application 108 may be a data enhancement application configured to perform data enhancement using the DICOM data. As another example, the data ingestion application 108 may be a data enrichment application configured to perform data enrichment using the DICOM data. Data ingestion may include a set of data ingestion phases. For example, data ingestion may include a data extraction phase, in which the data ingestion application 108 performs data extraction using the DICOM data. As another example, data ingestion may include a data enhancement phase, in which the data ingestion application 108 performs data enhancement using the DICOM data. As another example, data ingestion may include a data enrichment phase, wherein the data ingestion application 108 performs data enrichment using the DICOM data. It should be understood that data ingestion may include any number of phases. Further, it should be understood that multiple data ingestion applications 108 may perform respective data enrichments on the DICOM data. According to an embodiment, the data ingestion system 106 may store information identifying the DICOM data at each respective data ingestion phase of the data ingestion phase. The DICOM data may be different at each respective data ingestion phase. For example, various data elements of the DICOM data may have adjusted values, added values, modified values, removed values, enhanced values, etc., depending on the data ingestion application 108 processing the DICOM data associated with the data ingestion.

[0028] The database 110 may be configured to store DICOM data. For example, the database 110 may be a hierarchical database, a network database, a relational database, etc. The database 110 may store information related to data ingestion of the DICOM data. For example, the database 110 may store DICOM data, a DICOM data identifier that identifies the DICOM data, a radiological device identifier that identifies the radiological device 102 that generates the DICOM data, a manufacturer identifier that identifies the manufacturer of the radiological device 102, a model identifier that identifies the model of the radiological device 102, a site identifier that identifies the site associated with the radiological device 102, a time identifier that identifies the time associated with the DICOM data (e.g., the time of data ingestion, the time of data generation, etc.), etc. According to an embodiment, the database 110 may store information identifying the DICOM data at each corresponding stage in the data ingestion stage.

[0029] The data quality platform 112 may be configured to receive information related to data ingestion of DICOM data from the radiology device 102 to the database 110. Furthermore, the data quality platform 112 may be configured to determine corresponding data quality of the DICOM data at the set of data ingestion stages of the data ingestion using a set of data validation rules corresponding to a set of data ingestion stages of the data ingestion of the DICOM data from the radiology device 102 to the database 110. Furthermore, the data quality platform 112 may be configured to generate a user interface that displays information related to the corresponding data quality of the DICOM data at the set of data ingestion stages of the data ingestion.

[0030] The AI ​​model 114 may be configured to receive DICOM data as input, determine the data quality of the DICOM data, and provide the data quality of the DICOM data as output. For example, the AI ​​model 114 may be a deep neural network (DNN), a convolutional neural network (CNN), a fully convolutional network (FCN), a recurrent neural network (RCN), a Bayesian network, a graphical probability model, a K-nearest neighbor classifier, a decision forest, a maximum margin method, etc.

[0031] The user device 116 may be configured to display information related to the corresponding data quality of the DICOM data at the set of data acquisition phases of data acquisition. For example, the user device 116 may be a smart phone, a laptop computer, a desktop computer, a wearable device, etc.

[0032] The network 118 may be configured to allow communication between the radiology equipment 102, the data source 104, the data ingestion system 106, the database 110, the data quality platform 112, and / or the user device 116. For example, the network 118 may be a cellular network (e.g., a fifth generation (5G) network, a long term evolution (LTE) network, a third generation (3G) network, a code division multiple access (CDMA) network, etc.), a public land mobile network (PLMN), a local area network (LAN), a wide area network (WAN), a metropolitan area network (MAN), a telephone network (e.g., a public switched telephone network (PSTN)), a private network, an ad hoc network, an intranet, the Internet, a fiber-based network, the like, and / or a combination of these or other types of networks.

[0033] Figure 1 The number and arrangement of devices of system 100 shown in FIG. 1 are provided as examples. In practice, system 100 may include additional devices, fewer devices, different devices, or different Figure 1Additionally or alternatively, one or more devices of system 100 (eg, one or more devices) may perform one or more functions described as being performed by another set of devices of system 100.

[0034] Figure 2 yes Figure 1 2 is a diagram of example components of a device 200. The device 200 may correspond to a radiology device 102, a data source 104, a data ingestion system 106, a database 110, a data quality platform 112, and / or a user device 116. Figure 2 As shown, device 200 may include a bus 202 , a processor 204 , a memory 206 , a storage component 208 , an input component 210 , an output component 212 , and a communication interface 214 .

[0035] The bus 202 includes components that allow communication between components of the device 200. The processor 204 can be implemented in hardware, firmware, or a combination of hardware and software. The processor 204 can be a central processing unit (CPU), a graphics processing unit (GPU), an accelerated processing unit (APU), a microprocessor, a microcontroller, a digital signal processor (DSP), a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), or another type of processing component.

[0036] Processor 204 may include one or more processors that can be programmed to perform functions. Memory 206 may include random access memory (RAM), read-only memory (ROM), and / or another type of dynamic or static storage device (e.g., flash memory, magnetic memory, and / or optical memory) that stores information and / or instructions for use by processor 204.

[0037] The storage component 208 may store information and / or software related to the operation and use of the device 200. For example, the storage component 208 may include a hard disk (e.g., a magnetic disk, an optical disk, a magneto-optical disk, and / or a solid-state disk), a compact disk (CD), a digital versatile disk (DVD), a floppy disk, a cassette, a magnetic tape, and / or another type of non-transitory computer-readable medium and a corresponding drive.

[0038] Input components 210 may include components that allow device 200 to receive information, such as via user input (e.g., a touch screen display, a keyboard, a keypad, a mouse, buttons, switches, a camera, and / or a microphone). Additionally or alternatively, input components 210 may include sensors for sensing information (e.g., a global positioning system (GPS) component, an accelerometer, a gyroscope, and / or an actuator). Output components 212 may include components that provide output information from device 200 (e.g., a display, a speaker for outputting sound at an output sound level, and / or one or more light emitting diodes (LEDs)).

[0039] The communication interface 214 may include a transceiver-like component (e.g., a transceiver and / or a separate receiver and transmitter) that enables the device 200 to communicate with other devices, such as via a wired connection, a wireless connection, or a combination of a wired connection and a wireless connection. The communication interface 214 may allow the device 200 to receive information from another device and / or provide information to another device. For example, the communication interface 214 may include an Ethernet interface, an optical interface, a coaxial interface, an infrared interface, a radio frequency (RF) interface, a universal serial bus (USB) interface, a Wi-Fi interface, a cellular network interface, etc.

[0040] Device 200 can perform one or more processes described herein. Device 200 can perform these processes based on processor 204 executing software instructions stored by non-transitory computer-readable media such as memory 206 and / or storage component 208. Computer-readable media can be defined as non-transitory memory devices in this article. Memory devices can include memory space within a single physical storage device or memory space distributed across multiple physical storage devices.

[0041] The software instructions may be read into the memory 206 and / or storage component 208 from another computer-readable medium or from another device via the communication interface 214. The software instructions stored in the memory 206 and / or storage component 208, when executed, may cause the processor 204 to perform one or more processes described herein. Additionally or alternatively, hardwired circuitry may be used in place of or in combination with software instructions to perform one or more processes described herein. Thus, the specific implementations described herein are not limited to any specific combination of hardware circuitry and software.

[0042] Figure 2 The number and arrangement of components shown in FIG. 2 are provided as examples. In practice, device 200 may include additional components, fewer components, different components, or different components. Figure 2Additionally or alternatively, one set of components (eg, one or more components) of the device 200 may perform one or more functions described as being performed by another set of components of the device 200.

[0043] Figure 3 is an illustration of a flow chart of an example process 300 for determining data quality of DICOM data using a set of validation rules corresponding to a set of data ingestion phases of data ingestion of the DICOM data from a radiology device to a database. The data quality platform 112 may be configured to perform the operations of the process 300 .

[0044] like Figure 3 As shown, process 300 may include: receiving information related to data ingestion of DICOM data from a radiology device to a database (operation 302). According to an embodiment, the information related to data ingestion may include DICOM data. Additionally or alternatively, the information related to data ingestion may include DICOM data at each corresponding data ingestion stage in the data ingestion stage. Additionally or alternatively, the information related to data ingestion may include information identifying data ingestion rules associated with data ingestion performed by the data ingestion system 106. Additionally or alternatively, the information related to data ingestion may include information identifying a set of data ingestion stages of data ingestion. Additionally or alternatively, the information related to data ingestion may include information identifying a set of data ingestion applications 108 that process DICOM data. Additionally or alternatively, the information related to data ingestion may include information identifying corresponding rules for a set of data ingestion applications 108 that process DICOM data.

[0045] According to an embodiment, the data quality platform 112 may receive information related to data ingestion of DICOM data at predetermined intervals. For example, the data quality platform 112 may receive information related to data ingestion of DICOM data every ten minutes, every hour, every twenty-four hours, etc. Additionally or alternatively, the data quality platform 112 may receive information related to data ingestion of DICOM data in substantially real time. In this document, "substantially real time" may refer to receiving DICOM data (e.g., radiology equipment 102 generates DICOM data, data ingestion system 106 performs data ingestion, database 110 stores DICOM data, etc.) within a threshold time (e.g., ten seconds, one minute, etc.) of a reference time. Additionally or alternatively, the data quality platform 112 may receive information related to data ingestion of DICOM data based on a request from a user device 116.

[0046] like Figure 3As further shown in , process 300 may include determining corresponding data quality of the DICOM data at the set of data ingestion stages of the data ingestion using a set of validation rules corresponding to a set of data ingestion stages of the data ingestion (operation 304).

[0047] Data quality may be a representation of the actual state of the DICOM data compared to a predetermined state. According to an embodiment, data quality may be represented as a data quality metric. For example, a data quality metric may be the accuracy of the DICOM data, the completeness of the DICOM data, the comprehensiveness of the DICOM data, the consistency of the DICOM data, the timeliness of the DICOM data, the uniqueness of the DICOM data, the validity of the DICOM data, etc.

[0048] The data quality platform 112 may determine the data quality of the DICOM data at the data ingestion phase of the data ingestion using data validation rules corresponding to the data ingestion phase. The data ingestion may include a set of data ingestion phases (e.g., a single data ingestion phase, ten data ingestion phases, one hundred data ingestion phases, etc.). The data quality platform 112 may determine the corresponding data quality of the DICOM data at the set of data ingestion phases. For example, the data quality platform 112 may determine the data quality of the DICOM data at one or more data ingestion phases of the data ingestion phase. The data quality platform 112 may determine the data quality of the DICOM data at the specific data ingestion phase using specific data validation rules corresponding to the specific data ingestion phase. The data quality platform 112 may determine the specific data validation rules to be applied at the specific data ingestion phase based on information related to the data ingestion.

[0049] Figure 4 4 is a diagram of an example process 400 for determining data quality of DICOM data using a set of validation rules corresponding to a set of data ingestion phases of data ingestion of the DICOM data from a radiology device to a database. Figure 4As shown, data ingestion 402 may include a set of data ingestion phases, such as data ingestion phase 404, data ingestion phase 406, data ingestion phase 408, and data ingestion phase 410. In data ingestion phase 404, the radiology device 102 may provide DICOM data 412 to the data ingestion system 106. The data quality platform 112 may determine the data quality of the DICOM data 412 using data validation rules 414 corresponding to the data ingestion phase 404. In data ingestion phase 406, the data ingestion system 106 may input the DICOM data 412 into the data ingestion application 108, and receive DICOM data 416 output by the data ingestion application 108. The data quality platform 112 may determine the data quality of the DICOM data 416 using data validation rules 418 corresponding to the data ingestion phase 406. In data ingestion phase 408, the data ingestion system 106 may receive DICOM data 420 output by the data ingestion application 108. The data quality platform 112 may determine the data quality of the DICOM data 420 using data validation rules 422 corresponding to the data ingestion phase 408. In the data ingestion phase 410, the data ingestion system 106 may provide the DICOM data 420 to the database 110. The data quality platform 112 may determine the data quality of the DICOM data 420 using data validation rules 422 corresponding to the data ingestion phase 410.

[0050] The data quality platform 112 may receive the DICOM data, perform data validation using data validation rules, and determine data quality of the DICOM data based on performing the data validation. Figure 5 is a flow chart of an example process 500 for determining data quality of DICOM data. Figure 6 is a diagram of example validation rules 600 for determining data quality for DICOM.

[0051] like Figure 5 As shown, process 500 may include receiving DICOM data (operation 502 ). For example, data quality platform 112 may receive DICOM data corresponding to a particular data ingestion phase from database 110 or data ingestion system 106 .

[0052] like Figure 5 As further shown in FIG. 5 , process 500 may include performing data validation using the DICOM data and data validation rules that are common to multiple imaging modalities of the radiation device (operation 504). The data validation rules may be common to multiple imaging modalities of the radiation device 102. For example, the data validation rules may correspond to whether a particular data element (e.g., a study identifier, a patient identifier, a patient gender, etc.) has a value in the DICOM data. Figure 6As shown in , the data quality platform 112 can determine the data quality of the DICOM data using data validation rules 602 that are common to multiple imaging modalities of the radiology devices 102. For example, the data quality platform 112 can use the data validation rules 602 when determining the data quality of the DICOM data of a first radiology device 102 associated with a first imaging modality, and can also use the data validation rules 602 when determining the data quality of the DICOM data of a second radiology device 102 associated with a second imaging modality different from the first imaging modality.

[0053] like Figure 5 As further shown in , process 500 may include: performing data validation (operation 506) using the DICOM data and data validation rules specific to the imaging modality of the radiology device 102. The data validation rules may be specific to the imaging modality of the radiology device 102 that generates the DICOM data. For example, the data validation rules may correspond to parameters or configurations of the radiology device 102. As an example, if the data validation rules correspond to an ultrasound imaging modality, the data quality platform 112 may determine whether the duration of the ultrasound imaging is within a specific time range defined by a minimum time and a maximum time. If the data quality platform 112 determines that the duration is not within the specific time range, the data quality platform 112 may not validate the DICOM data. As another example, if the data validation rules correspond to a CT imaging modality, the data quality platform 112 may determine whether an acquisition parameter (e.g., a current value, a voltage value, etc.) is within a range defined by a minimum value and a maximum value. If the data quality platform 112 determines that the acquisition parameter is not within the specific time range, the data quality platform 112 may not validate the DICOM data. Figure 6 As shown, the data quality platform 112 can use data validation rules 604 specific to the first imaging modality to determine data quality, use data validation rules 610 specific to the second imaging modality to determine data quality, and use nth data validation rules 616 specific to the nth imaging modality to determine data quality. For example, the data quality platform 112 can use data validation rules 604 only for radiation devices 102 of the first imaging modality, can use data validation rules 610 only for radiation devices 102 of the second imaging modality, and so on.

[0054] like Figure 5 As further shown in FIG. 5 , process 500 may include performing data validation using the DICOM data and data validation rules specific to the data ingestion application (operation 508). For example, the data validation rules may correspond to a particular data ingestion application 108. As an example, the data quality platform 112 may determine whether a score of an output image output by a data ingestion application 108 configured to perform data enrichment is within a range defined by a minimum value and a maximum value. Figure 6As shown, the data quality platform 112 can determine data quality using data validation rules 606 specific to a first data ingestion application 108 associated with a first imaging modality, determine data quality using data validation rules 608 specific to an m-th data ingestion application 108 associated with the first imaging modality, determine data quality using data validation rules 612 specific to a first data ingestion application 108 associated with a second imaging modality, determine data quality using data validation rules 614 specific to a p-th data ingestion application 108 associated with the second imaging modality, determine data quality using data validation rules 618 associated with a first data ingestion application 108 associated with an n-th imaging modality, and determine data quality using data validation rules 620 associated with a q-th data ingestion application 108 associated with an n-th imaging modality. For example, the data quality platform 112 can use data validation rules 606 associated with only a first data ingestion application 108 associated with a first imaging modality. According to an embodiment, the data ingestion application 108 can generate new data elements for DICOM data related to medical data usage. The data quality platform 112 can determine whether the scope of the new data element is within a specific range.

[0055] like Figure 5 As further shown in , process 500 may include: performing anomaly detection using the DICOM data (operation 510). For example, the data quality platform 112 may perform anomaly detection using anomaly detection techniques such as supervised anomaly detection techniques, semi-supervised anomaly detection techniques, unsupervised anomaly detection techniques, statistical techniques, density-based techniques, subspace techniques, correlation-based techniques, neural networks, Bayesian networks, and the like. By performing anomaly detection, the data quality platform 112 may determine whether the DICOM data deviates from previous DICOM data, determine whether the DICOM data includes rarely occurring values, determine whether the DICOM data is inconsistent with previous DICOM data, and the like. As an example, the data quality platform 112 may determine whether the repartitioning of the exam duration is consistent across time. The data quality platform 112 may detect an anomaly if the distribution of values ​​of a data element of the DICOM data for a particular radiology device 102 deviates by a large amount.

[0056] According to an embodiment, the data quality platform 112 can determine the data quality using the AI ​​model 114. For example, the data quality platform 112 can input DICOM data into the AI ​​model 114 and determine the data quality of the DICOM data based on the output of the AI ​​model 114.

[0057] like Figure 5As further shown, process 500 may include determining data quality of the DICOM data (operation 512). For example, the data quality platform 112 may determine the data quality of the DICOM data based on performing operation 504, operation 506, operation 508, or operation 510. According to an embodiment, the data quality of the DICOM data may correspond to a collective data set. For example, the data quality platform 112 may determine the corresponding data quality of the data elements in the data set, and determine the collective data quality of the data set based on the corresponding data quality. Additionally or alternatively, the data quality may correspond to a specific data element in the data set. For example, the data quality platform 112 may determine the data quality of a specific data element in the data set, and determine the specific data quality as the data quality.

[0058] In this manner, the data quality platform 112 can determine the corresponding data quality of one or more data elements of the DICOM data at one or more data ingestion stages in a set of data ingestion stages of data ingestion from the radiology device 102 to the database 110. Thus, in this manner, the embodiments herein improve the security, efficacy, and reliability of DICOM data by ensuring adequate data quality, eliminate or reduce the need for human intervention in assessing data quality, accelerate the serviceability of radiology devices, improve transparency in data ingestion systems, improve identification of root causes of issues using data lineage, automatically retain DICOM data if data quality does not meet a threshold, automatically identify issues associated with data ingestion, reduce the time required to connect, integrate, and verify radiology devices, and improve the use of AI models trained using DICOM data.

[0059] like Figure 3 As further shown in FIG. 3 , process 300 may include generating a user interface that displays information related to the corresponding data quality of the DICOM data at the set of data ingestion stages of the data ingestion (operation 306). The data quality platform 112 may generate a user interface and provide the user interface to the user device 116 to allow the user device 116 to display information related to the corresponding data quality of the DICOM data at the set of data ingestion stages of the data ingestion from the radiology device 102 to the database 110.

[0060] According to an embodiment, the information related to data quality may be the status of the data quality. For example, the status may be "normal", "passed", "accepted", "valid", "high", "medium", "low", "failed", "warning", etc. Additionally or alternatively, the information related to data quality may be a discrete value of the data quality. For example, the discrete value may be "100%", "90%", "7", "0.5", etc. Additionally or alternatively, the information related to data quality may be an indicator of a specific data acquisition phase associated with the DICOM data. Additionally or alternatively, the information related to data quality may be DICOM data. Additionally or alternatively, the information related to data quality may be a DICOM data identifier that identifies the DICOM data. Additionally or alternatively, the information related to data quality may be a label of a data element that causes the data quality to not meet a threshold. Additionally or alternatively, the information related to data quality may be a radiological device identifier that identifies the radiological device 102 that generates the DICOM data. Additionally or alternatively, the information related to data quality may be a manufacturer identifier that identifies the manufacturer of the radiological device 102. Additionally or alternatively, the information related to data quality may be a model identifier identifying a model of the radiology device 102. Additionally or alternatively, the information related to data quality may be a site identifier identifying a site associated with the radiology device 102. Additionally or alternatively, the information related to data quality may be a time identifier identifying a time associated with the DICOM data (e.g., a time of data ingestion, a time of data generation, etc.), etc. Additionally or alternatively, the information related to data quality may be an identifier of a data validation rule applied to the DICOM data. Additionally or alternatively, the information related to data quality may be an identifier of a data ingestion application 108 that processes the DICOM data.

[0061] Figure 7 is a diagram of an example user interface 700 for displaying data quality of DICOM data associated with a radiology device. Figure 7As shown, the user interface 700 can display data 702 corresponding to a group of radiology devices 102. The data 702 can include device identifiers 704 of the radiology devices 102, site identifiers 706 identifying sites associated with the radiology devices 102, manufacturer identifiers 708 identifying manufacturers of the radiology devices 102, model identifiers 710 identifying models of the radiology devices 102, last association identifiers 712 identifying last associations of the radiology devices 102, last data ingest identifiers 714 identifying last data ingestions of the radiology devices 102, and status identifiers 716 of data quality of DICOM data of the radiology devices 102. The user interface 700 can include fields 718 that allow an operator to filter the data 702 by specific radiology devices 102. The user interface 700 can include fields 718 that allow an operator to filter the data 702 by specific sites.

[0062] Figure 8 8 is a diagram of an example user interface 800 for displaying the corresponding data quality of DICOM data corresponding to a validation rule. Figure 8 As shown, the user interface 800 can display display data 802 corresponding to the corresponding data quality of the DICOM data corresponding to the validation rules of the radiology device 102. As shown, the user interface 800 can display data validation rules 804, the status 806 of the DICOM data with respect to the data validation rules, the integrity metric 808 of the DICOM data with respect to the data validation rules, the consistency metric 810 of the DICOM data with respect to the data validation rules, etc.

[0063] Fig. 9 is a flow chart of an example process 900 for determining a root cause for data quality not meeting a threshold. The data quality platform 112 can be configured to perform the operations of process 900 .

[0064] like Fig. 9As shown, process 900 may include: determining whether the data quality of the DICOM data meets the threshold (operation 902). According to an embodiment, the data quality platform 112 may compare the corresponding data quality of the DICOM data at one or more data ingestion stages of data ingestion with the corresponding threshold, and determine whether the corresponding data quality meets the corresponding threshold. For example, the data quality platform 112 may compare the value of the data quality of the DICOM data with the threshold, and determine whether the value of the data quality meets the threshold. As a specific example, the data quality platform 112 may compare the value of "90%" with the threshold of "70%". In this case, "meeting the threshold" may mean that the value of the data quality is greater than or greater than or equal to the threshold. As another example, the data quality platform 112 may compare the state of the data quality with the threshold state. As a specific example, the data quality platform 112 may compare the state "high" with the threshold state "medium". In this case, "meeting the threshold" may mean that the state of the data quality is greater than or greater than or equal to the threshold state.

[0065] like Fig. 9 As further shown in FIG. 9 , if the data quality meets the threshold (operation 902—Yes), process 900 may include generating a user interface that displays information indicating that the data quality meets the threshold (operation 904). For example, the data quality platform 112 may generate a user interface that displays a status of the DICOM data that indicates that the DICOM data meets the threshold. Additionally, the data quality platform 112 may display the corresponding status of the DICOM data with respect to various data validation rules that indicate that the DICOM data meets the threshold.

[0066] like Fig. 9 As further shown in , if the data quality does not meet the threshold (operation 902—No), process 900 may include: performing data lineage based on determining that the data quality of the DICOM data does not meet the threshold (operation 906), and determining the root cause of the data quality not meeting the threshold (operation 908).

[0067] According to an embodiment, the data quality platform 112 may determine a data ingestion stage associated with the DICOM data having a data quality that does not meet a threshold. According to an embodiment, the data quality platform 112 may perform backward data lineage to determine the root cause of the DICOM data not meeting the threshold. As an example, the data quality platform 112 may perform backward data lineage to determine that the root cause of the DICOM data not meeting the threshold is an incorrect configuration of the radiology device 102. As another example, the data quality platform 112 may perform backward data lineage to determine that the root cause of the DICOM data not meeting the threshold is an incorrect configuration of the data ingestion application 108 that processes the DICOM data. According to an embodiment, the data quality platform 112 may perform forward data lineage to determine the data ingestion application 108 that may be affected by the DICOM data not meeting the threshold. The data quality platform 112 may prevent the data ingestion application 108 from accessing the DICOM data having a data quality that does not meet the threshold. The data quality platform 112 may determine DICOM data associated with a downstream data ingestion stage in a data ingestion stage associated with DICOM data having data quality that does not meet a threshold, and perform an action using the DICOM data (e.g., marking the DICOM data as suspect, determining data quality of the DICOM data, etc.).

[0068] According to an embodiment, the data quality platform 112 can generate a graph structure based on information related to data ingestion. For example, the data quality platform 112 can generate a graph structure based on information associated with the data ingestion system 106, information associated with the data ingestion application 108, etc. The graph structure may include a set of nodes corresponding to DICOM data at various data ingestion stages of data ingestion, and edges connecting the set of nodes and identifying the relationship between the set of nodes. For example, an edge can identify a data ingestion application 108 that processes DICOM data at various data ingestion stages. The data quality platform 112 can use the graph structure to perform data lineage. According to another embodiment, the data quality platform 112 can perform log parsing to perform data lineage.

[0069] like Fig. 9 As further shown in , process 900 may include generating a user interface that displays a root cause for data quality not meeting a threshold (operation 910). For example, data quality platform 112 may generate a user interface that displays the root cause and provide the user interface to user device 116 for display. Additionally, data quality platform 112 may generate the user interface to include information identifying potential actions to be performed to improve the data quality of the DICOM data. In this manner, an operator of user device 116 may identify the root cause and perform an action to improve the data quality of the DICOM data.

[0070] Fig.10 is a flow chart of an example process 1000 for performing actions to improve data quality of DICOM data. The data quality platform 112 may be configured to perform the operations of the process 1000 .

[0071] like Fig.10 As shown, process 1000 may include: determining the root cause of data quality not meeting a threshold (operation 1002), and generating a user interface that displays the root cause of data quality not meeting the threshold (operation 1004). For example, the data quality platform 112 may be combined with the above Fig. 9 The root cause of the data quality not meeting the threshold is determined in a manner similar to the manner described in operation 902 of the present invention, and can be combined with the above Fig. 9 A user interface is generated in a manner similar to that described in operation 908 , and the user interface displays the root cause of the data quality not meeting the threshold.

[0072] like Fig.10 As further shown in FIG. 1 , process 1000 may include receiving user input via a user interface selecting to improve data quality of the DICOM data (operation 1006 ), and performing actions to improve data quality of the DICOM data (operation 1008 ).

[0073] The data quality platform 112 may generate a user interface to display information identifying the root cause of the data quality not meeting the threshold, and display information allowing the operator to select to improve the data quality of the DICOM data. For example, if the root cause of the data quality not meeting the threshold is that the configuration of the radiation device 102 is incorrect, the user interface may display information allowing the operator to select to adjust the configuration of the radiation device 102 and / or use the adjusted configuration of the radiation device to perform additional imaging on the patient. As another example, if the root cause of the data quality not meeting the threshold is that the configuration of the data ingestion application 108 is incorrect, the user interface may display information allowing the operator to select to adjust the configuration of the data ingestion application 108 and / or use the data ingestion application 108 with the adjusted configuration to perform updated data enhancement. As another example, if the root cause of the data quality not meeting the threshold is that the value of the data element is omitted, the data quality platform 112 may display information allowing the operator to select to enter the value of the data element. As another example, if the root cause of the data quality not meeting the threshold is that the format of the value of the data element is incorrect, the data quality platform 112 may display information allowing the operator to select to reformat the value of the data element. Based on the user selection, the data quality platform 112 can perform actions to improve the data quality of the DICOM data. For example, the data quality platform 112 can adjust the configuration of the radiology device 102 (e.g., acquisition parameters, settings, etc.). As another example, the data quality platform 112 can cause the radiology device 102 to perform additional scans on the patient. As another example, the data quality platform 102 can prevent the radiology device 102 from performing additional scans on the patient. As another example, the data quality platform 112 can adjust the configuration of the data ingestion application 108. As another example, the data quality platform 112 can prevent the data ingestion application 108 from processing the DICOM data during data ingestion. As another example, the data quality platform 112 can input the value of the data element. As another example, the data quality platform 112 can reformat the value of the data element. As another example, the data quality platform 112 can send a notification to another device. As another example, the data quality platform 112 can retain the DICOM data, such as by preventing the DICOM data from being displayed, accessed by another device, sent to another device, etc.

[0074] In this way, the embodiments herein improve the security, efficacy, and reliability of DICOM data by ensuring adequate data quality, eliminating or reducing the need for human intervention in assessing data quality, accelerating the serviceability of radiology equipment, improving transparency in data ingestion systems, improving identification of root causes of issues using data lineage, automatically retaining DICOM data when data quality does not meet a threshold, automatically identifying issues associated with data ingestion, reducing the time required to connect, integrate, and authenticate radiology equipment, and improving the use of AI models trained using DICOM data.

[0075] Fig.11 1 is a diagram of an example process 1100 for training an artificial intelligence (AI) model. The data quality platform 112 may generate, store, train, and / or use an AI model 114. According to an embodiment, the data quality platform 112 may include the AI ​​model 114 and / or instructions associated with the AI ​​model 114. For example, the data quality platform 112 may include instructions for generating the AI ​​model 114, training the AI ​​model 114, using the AI ​​model 114, and the like. According to another embodiment, a system or device different from the data quality platform 112 may be used to generate and / or train the AI ​​model 114. For example, the system or device may include instructions for generating the AI ​​model 114 and / or instructions for training the AI ​​model 114. The system or device may provide the resulting trained AI model 114 to the data quality platform 112 for use.

[0076] like Fig.11 As shown in , according to an embodiment, the process 1100 may include a training phase 1102, a deployment phase 1108, and a monitoring phase 1114. In the training phase 1102, at operation 1106, the process 1100 may include: receiving and processing training data 1104 to generate a trained AI model 114 for determining the data quality of DICOM data. The training data 1104 may include a plurality of training data sets associated with a plurality of radiology devices 102. Each training data set in the plurality of training data sets may be associated with a radiology device 102. An example training data set may include DICOM data and an associated data quality of the DICOM data, which may be provided as an input to train the AI ​​model 114.

[0077] The training data 1104 may be generated, received, or otherwise obtained from internal and / or external sources. For example, DICOM data may be obtained from a radiology device 102, a data source 104, a data ingestion system 106, and / or a database 110. The training data may identify a data ingestion application 108 that processes DICOM data at various stages of data ingestion.

[0078] In general, the AI ​​model 114 may include a set of variables (e.g., nodes, neurons, filters, etc.) that are fine-tuned (e.g., weighted, biased, etc.) to different values ​​via the application of the training data 1104. According to an embodiment, the training process at operation 1106 may employ supervised, unsupervised, semi-supervised, and / or reinforcement learning processes to train the AI ​​model 114. According to an embodiment, a portion of the training data 1104 may be retained during training and / or used to validate the trained AI model 114.

[0079] For supervised learning process, training data 1104 may include a label or score that can facilitate the training process by providing ground truth. For example, a label or score may indicate the data quality of DICOM data. Training may be performed by feeding a training data set including DICOM data and associated data quality from training data 1104 into AI model 114. AI model 114 may have variables that are set to initialization values ​​(e.g., randomly, based on Gaussian noise, based on pre-trained values, etc.). AI model 114 may output the data quality of DICOM data based on DICOM data input to AI model 114. The output may be compared with a corresponding label or score (e.g., ground truth) indicating data quality, which may then be back-propagated through AI model 114 to adjust the value of the variable. The process may be repeated for multiple samples, at least until the determined loss or error is below a predefined threshold. According to an embodiment, some training data in training data 1104 may be retained and used to further verify or test the trained AI model 114.

[0080] For an unsupervised learning process, the training data 1104 may not include pre-assigned labels or scores to assist in the learning process. Instead, the unsupervised learning process may include clustering, classification, and the like to identify naturally occurring patterns in the training data 1104. As an example, DICOM data may be clustered into groups based on identified similarities and / or patterns. K-means clustering or K-nearest neighbors may also be used, which may be supervised or unsupervised. A combination of K-nearest neighbor and unsupervised clustering techniques may also be used. For semi-supervised learning, a combination of training data 1104 with pre-assigned labels or scores and training data 1104 without pre-assigned labels or scores may be used to train the AI ​​model 114.

[0081] When reinforcement learning is employed, an agent (e.g., an algorithm) may be trained to make decisions about data quality through trial and error from training data 1104. For example, based on making a decision, the agent may then receive feedback (e.g., a positive reward if the prediction is above a predetermined threshold), adjust its next decision to maximize the reward, and repeat until the loss function is optimized.

[0082] After being trained, the trained AI model 114 may be stored and subsequently applied by the data quality platform 112 during a deployment phase 1108. For example, during the deployment phase 1108, the trained AI model 114 executed by the data quality platform 112 may receive input data 1110. The input data 1110 may include DICOM data generated by the radiology device 102 and processed by the data ingestion application 108 of the data ingestion system 106. The AI ​​model 114 may provide data quality of the DICOM data as output data 1112.

[0083] The monitoring data 1116 may include data identifying the data quality of the DICOM data determined by the operator. During process 1118, the monitoring data 1116 may be analyzed together with the predicted output data 1112 and the input data 1110 to determine the accuracy of the trained AI model 114. According to an embodiment, based on the analysis, the process 1000 may return to the training phase 1102, where at operation 1106, the values ​​of one or more variables of the model may be adjusted to improve the accuracy of the AI ​​model 114.

[0084] The above example process 1000 is provided as an example only and may include more Fig.11 Additional, fewer, different, or differently arranged aspects of the ones depicted.

[0085] Fig.11 The training, deployment, and monitoring associated with a trained AI model 114 for determining data quality of DICOM data are described. According to an embodiment, one or more other trained AI models 114 may be applied, such as a trained AI model 114 for performing data validation, for performing anomaly detection, for determining the root cause of data quality not meeting a threshold, etc. Each of the trained AI models 114 may include the training AI models described above for Fig.11 Similar training, deployment, and / or monitoring phases are described for the trained AI model 114 in , however, the specific types of training data, input data, output data, and monitoring data may be different.

[0086] The embodiments of the present disclosure shown in the drawings and described above are exemplary embodiments only and are not intended to limit the scope of the appended claims, including any equivalents included within the scope of the claims. Various modifications are possible and will be apparent to those skilled in the art. It is intended that any combination of non-mutually exclusive features described herein is within the scope of the invention. That is, the features of the embodiments may be combined with any appropriate aspect described above, and the optional features of any one aspect may be combined with any other appropriate aspect. Similarly, the features listed in the dependent claims may be combined with non-mutually exclusive features of other dependent claims, particularly where the dependent claims are dependent on the same independent claim. In some jurisdictions that require single claim dependencies, these dependencies may have been used as a practice, but this should not be taken to mean that the features in the dependent claims are mutually exclusive.

Claims

1. A method (300), comprising: receiving (302) information related to data ingestion (402) of Digital Imaging and Communications in Medicine (DICOM) data (412, 416, 420) from a radiology device (102) to a database (110); determining (304) validation of the DICOM data (412, 416, 420) at the set of data ingestion stages (404, 406, 408, 410) of the data ingestion (402) from the radiology device (102) to the database (110) using a set of data validation rules (414, 418, 422) corresponding to a set of data ingestion stages (404, 406, 408, 410) of the data ingestion (402) of the DICOM data (412, 416, 420) corresponding data quality, wherein the data validation rules (414, 418, 422) include one or more of a first data validation rule (602) that is common to multiple imaging modalities of a radiology device (102), a second data validation rule (604, 610, 616) that is specific to an imaging modality of the radiology device (102), and a third data validation rule (606, 608, 612, 614, 618, 620) that is specific to a data ingestion application (108) associated with the data ingestion (402); and A user interface (700, 800) is generated (306) based on determining the corresponding data qualities of the DICOM data (412, 416, 420) at the set of data ingestion stages (404, 406, 408, 410) of the data ingestion (402), the user interface displaying information related to the corresponding data qualities of the DICOM data (412, 416, 420) at the set of data ingestion stages (404, 406, 408, 410) of the data ingestion (402).

2. The method (300) of claim 1, further comprising: performing (504) a first data validation using the DICOM data (412, 416, 420) and the first data validation rules (602) that are common to the plurality of imaging modalities of the radiology device (102), Wherein determining (304) the corresponding data quality of the DICOM data (412, 416, 420) in the set of data ingestion stages (404, 406, 408, 410) of the data ingestion (402) includes: determining (304) the corresponding data quality of the DICOM data (412, 416, 420) in the set of data ingestion stages (404, 406, 408, 410) of the data ingestion (402) based on performing (504) the first data validation.

3. The method (300) of claim 2, further comprising: performing (506) a second data validation using the DICOM data (412, 416, 420) and the second data validation rules (604, 610, 616) specific to the imaging modality of the radiology device (102), Wherein determining (304) the corresponding data quality of the DICOM data (412, 416, 420) in the set of data ingestion stages (404, 406, 408, 410) of the data ingestion (402) includes: determining (304) the corresponding data quality of the DICOM data (412, 416, 420) in the set of data ingestion stages (404, 406, 408, 410) of the data ingestion (402) based on performing (506) the second data validation.

4. The method (300) according to claim 3, further comprising: performing (508) a third data validation using the DICOM data (412, 416, 420) and the third data validation rules (606, 608, 612, 614, 618, 620) specific to the data ingestion application (108) associated with the data ingestion (402), Wherein determining (304) the corresponding data quality of the DICOM data (412, 416, 420) in the set of data ingestion stages (404, 406, 408, 410) of the data ingestion (402) includes: determining (304) the corresponding data quality of the DICOM data (412, 416, 420) in the set of data ingestion stages (404, 406, 408, 410) of the data ingestion (402) based on performing the third data validation.

5. The method (300) of claim 1, further comprising: determining (902) that data quality in the corresponding data quality does not meet a threshold; performing (906) data lineage based on determining that the data quality does not meet the threshold; as well as A root cause of the data quality not meeting the threshold is determined (908) based on performing the data lineage.

6. The method (300) of claim 5, further comprising: The user interface (700, 800) is generated (910) to display information identifying the root cause.

7. The method (300) of claim 1, wherein the data validation rules (414, 418, 422) include the first data validation rules (602) that are common to multiple imaging modalities of a radiological device (102), the second data validation rules (604, 610, 616) that are specific to the imaging modality of the radiological device (102), and the third data validation rules (606, 608, 612, 614, 618, 620) that are specific to a data ingestion application (108) associated with the data ingestion (402).

8. A device (200), comprising: a memory (206) configured to store instructions; and One or more processors (204) configured to execute the instructions to: receiving (302) information related to data ingestion (402) of Digital Imaging and Communications in Medicine (DICOM) data (412, 416, 420) from a radiology device (102) to a database (110); A method for determining (304) validation of the DICOM data (412, 416, 420) at the set of data ingestion stages (404, 406, 408, 410) of the data ingestion (402) from the radiology device (102) to the database (110) using a set of data validation rules (414, 418, 422) corresponding to a set of data ingestion stages (404, 406, 408, 410) of the data ingestion (402) of the DICOM data (412, 416, 420) from the radiology device (102) to the database (110) 20), wherein the data validation rules (414, 418, 422) include a first data validation rule (602) that is common to multiple imaging modalities of a radiology device (102), a second data validation rule (604, 610, 616) that is specific to the imaging modality of the radiology device (102), and a third data validation rule (606, 608, 612, 614, 618, 620) that is specific to a data ingestion application (108) associated with the data ingestion (402); and A user interface (700, 800) is generated (306) based on determining the corresponding data qualities of the DICOM data (412, 416, 420) at the set of data ingestion stages (404, 406, 408, 410) of the data ingestion (402), the user interface displaying information related to the corresponding data qualities of the DICOM data (412, 416, 420) at the set of data ingestion stages (404, 406, 408, 410) of the data ingestion (402).

9. The device (200) of claim 8, wherein the one or more processors (204) are further configured to: performing (504) a first data validation using the DICOM data (412, 416, 420) and the first data validation rules (602) that are common to the plurality of imaging modalities of the radiology device (102), Wherein, when determining the corresponding data quality of the DICOM data (412, 416, 420) at the set of data ingestion stages (404, 406, 408, 410) of the data ingestion, the one or more processors are also configured to determine the corresponding data quality of the DICOM data (412, 416, 420) at the set of data ingestion stages (404, 406, 408, 410) of the data ingestion based on performing (504) the first data validation.

10. The device (200) of claim 9, wherein the one or more processors (204) are further configured to: performing (506) a second data validation using the DICOM data (412, 416, 420) and the second data validation rules (604, 610, 616) specific to the imaging modality of the radiology device (102), Wherein, when determining (304) the corresponding data quality of the DICOM data (412, 416, 420) in the set of data ingestion stages (404, 406, 408, 410) of the data ingestion (402), the one or more processors (204) are also configured to determine (304) the corresponding data quality of the DICOM data (412, 416, 420) in the set of data ingestion stages (404, 406, 408, 410) of the data ingestion (402) based on performing (506) the second data validation.

11. The device (200) of claim 10, wherein the one or more processors (204) are further configured to: performing (508) a third data validation using the DICOM data (412, 416, 420) and the third data validation rules (606, 608, 612, 614, 618, 620) specific to the data ingestion application (108) associated with the data ingestion (402), Wherein, when determining (304) the corresponding data quality of the DICOM data (412, 416, 420) at the set of data ingestion stages (404, 406, 408, 410) of the data ingestion, the one or more processors (204) are also configured to determine (304) the corresponding data quality of the DICOM data (412, 416, 420) at the set of data ingestion stages (404, 406, 408, 410) of the data ingestion based on performing (508) the third data validation.

12. The device (200) of claim 8, wherein the one or more processors (204) are further configured to: determining (902) that data quality in the corresponding data quality does not meet a threshold; performing (906) data lineage based on determining that the data quality does not meet the threshold; and A root cause of the data quality not meeting the threshold is determined (908) based on performing the data lineage.

13. The device (200) of claim 12, wherein the one or more processors (204) are further configured to: The user interface (700, 800) is generated (910) to display information identifying the root cause.

14. The device (200) of claim 8, wherein the one or more processors (204) are further configured to: determining (902) that data quality in the corresponding data quality does not meet a threshold, The information related to the corresponding data quality of the DICOM data (412, 416, 420) includes DICOM tags of data elements of the DICOM data (412, 416, 420) that cause the data quality to fail to meet the threshold.

15. A non-transitory computer-readable medium (206) storing instructions that, when executed by one or more processors (204), cause the one or more processors (204) to: receiving (302) information related to data ingestion (402) of Digital Imaging and Communications in Medicine (DICOM) data (412, 416, 420) from a radiology device (102) to a database (110); determining (304) validation of the DICOM data (412, 416, 420) at the set of data ingestion stages (404, 406, 408, 410) of the data ingestion (402) from the radiology device (102) to the database (110) using a set of data validation rules (414, 418, 422) corresponding to a set of data ingestion stages (404, 406, 408, 410) of the data ingestion (402) of the DICOM data (412, 416, 420) from the radiology device (102) to the database (110) , 416, 420), wherein the data validation rules include a first data validation rule (602) that is common to multiple imaging modalities of a radiology device (102), a second data validation rule (604, 610, 616) that is specific to the imaging modality of the radiology device (102), and a third data validation rule (606, 608, 612, 614, 618, 620) that is specific to a data ingestion application (108) associated with the data ingestion (402); and A user interface (700, 800) is generated (306) based on determining the corresponding data qualities of the DICOM data (412, 416, 420) at the set of data ingestion stages (404, 406, 408, 410) of the data ingestion (402), the user interface displaying information related to the corresponding data qualities of the DICOM data (412, 416, 420) at the set of data ingestion stages (404, 406, 408, 410) of the data ingestion (402).

Images