FHIR Radiology-Image Workflow

--- title: FHIR Radiology-Image Workflow tags: fhir-project --- <font color="red">FHIR</font> Radiology-Image Workflow === ###### tags: `fhir-project` ## Table of Contents [TOC] ## PPT Version  <iframe src="https://www.slideshare.net/slideshow/embed_code/key/4nVZlPdrTlpa72" width="100%" height="500" frameborder="0" marginwidth="0" marginheight="0" scrolling="no" style="border:1px solid #CCC; border-width:1px; margin-bottom:5px; max-width: 100%;" allowfullscreen> </iframe> <div style="margin-bottom:5px"> <strong> <a href="https://www.slideshare.net/VictoriaTjia1/fhir-medical-imaging-exam-workflowmisac-wg3-pptx" title="FHIR Medical Imaging Exam Workflow-MISAC WG3 .pptx" target="_blank">FHIR Medical Imaging Exam Workflow-MISAC WG3 .pptx</a> </strong> from <strong><a href="https://www.slideshare.net/VictoriaTjia1" target="_blank">Victoria Tjia</a></strong> </div> ## 1. Background ### 1.1. Current Radiology Workflow in Hospital *<center> ![](https://i.imgur.com/6eGk1VQ.png) Fig 1. General radiology image exam workflow </center>* Medical imaging is an important tool for diagnosis and treatment planning. For example in pulmonology clinic scenario, chest X-ray and CT Scan plays an important role in identifying potential pulmonary diseases. Pulmonology is a medical specialty that deals with the diseases of the pulmonary system, particularly those involving the upper respiratory tract, lungs, and bronchial tubes. Doctors who specialize in this field are called pulmonologists. When a patient come to do consultation with pulmonologist, most of the patient must have problem with their respiratory, e.g. chest pain, asthma, etc. As shown in the fig. 1, a pulmonology consultation starts with: **Step 1.** Patient arrives at the hospital and proceeds with the registration. **Step 2.** When a patient enters the consultation room, pulmonologists will start with discussion of the patient’s medical history, so the pulmonologist will fully understand the patient’s condition. Pulmonologists may also perform physical tests to ascertain the patient’s breathing. **Step 3.** To gather additional information needed to accurately identify potential pulmonary diseases, the pulmonologist may request additional tests, such as chest X-rays and CT scans. Pulmonologists will order the X-ray and CT scan in the Hospital Information System (HIS). **Step 4.** Next, the image exam will be scheduled based on the available time. The schedule itself is executed on a Radiology Information System, known as RIS. **Step 5.** Patient will then move to the radiology department receptionist to check in. **Step 6.** Before starting the imaging exam, a radiographer will typically take some steps to ensure that the patient's basic information is properly connected to the modality (e.g. X-Ray, CT scan, etc.). Once the preparation is complete, the patient will enter the room and undergo the image exam. **Step 7.** The medical image generated from modality will be stored into Picture Archiving and Communication System (PACS). **Step 8.** It’s time for the radiologist to review the image result. If an abnormality is detected on a medical image, a radiologist may draw an annotation to help highlight. Annotations can take the form of circles, arrows, or other marks that are overlaid on the image itself. **Step 9.** After drawing the annotation, the radiologist will further provide a more detailed description of the annotation, such as its size, location, and shape. **Step 10.** Next, the radiologist will write a diagnostic report, which is the final summary of the radiologist's findings and conclusion. The report itself is usually stored in the RIS. In the scenario where none of the abnormality is found on the medical image result, the radiologist might skip step 8 and 9 and write the diagnostic report directly in this step. **Step 11.** Once the X-rays or CT scan report is out, pulmonologists will be able to recommend a treatment plan, which may include medication, pulmonary rehabilitation, smoking cessation, dietary planning, and other therapeutic treatments. In the end of the consultation, the patient will bring a medical image film and paper-based radiology report. --- ### 1.2. Challenge in Radiology Image Sharing Between Hospitals Picture Archiving and Communication System (PACS) is a system for managing and archiving medical images created in radiology department. More specifically, PACS is used in radiology as a system for filing and sending images that come from specialized imaging/ radiological equipment such as Plain X-Ray, CT, or MRI. Hospitals that already have constructed PACS, all devices are connected via a computer network to the PACS server. And the imaging results of the devices would be stored on the server. Consequently, radiologists can use PACS viewer and analysis applications to query and access the medical images that were stored in PACS server. However, PACS follows on Digital Imaging and COmmunication in Medicine (DICOM) standard to support the process, transmit, store, retrieve, print and display of medical imaging data. Most of IT system developers are not familiar with DICOM specified protocol and data format, also conventional DICOM standard is not appropriated for sharing images across medical department. It is a huge overhead for those developers to handle DICOM standard images, presentation object, and structure report. And next problem, although security specifications has addressed in DICOM standard, most of PACS used in hospital do not support those security specifications. PACS usually confined in internal network inside hospital and cannot be accessed by others outside of the hospital. This will make doctors of a hospital unable to access the past imaging records of patients from other hospitals. The imaging viewer system used to access the patient's imaging results were also not able to save the results of the analysis to a server, making it difficult for doctors in making clinical reports. Recently, both DICOM and HL7 have published web specifications, DICOM web and HL7 Fast Healthcare Interoperability Resources (FHIR). FHIR is a standard for exchanging healthcare information electronically. To ensure that medical images have the same standard in each hospital, medical images can be uploaded to the FHIR server according to the standard. This enables hospitals to access a patient’s past medical image, by using an understandable international standard, FHIR standard, easily. To integrate the DICOM Web System, the hospital’s PACS system, and the FHIR server, we are planning to develop a web solution for medical images that can be uploaded to the FHIR server for a standard image sharing. ## 2. Purpose The main purposes of this article are to; - Providing a guidance in the process of creating medical image exam workflow through the new FHIR standard - Ensure that medical images have the same standard in each hospitals. Medical images can be uploaded to the FHIR server according to the standard, this enables hospitals to access a patient’s past medical images by using an understandable international standard, FHIR standard, easily - Improve radiologists’ performance in practice by developing a standard web solution for medical images and annotation sharing ## 3. Methodology ### 3.1. Conventional Imaging Workflows Conventional medical imaging workflows require the integration of information systems. As shown in Fig. 2, standardized data is created in the HIS, RIS (Radiological Information System), and PACS, and then shared across systems based on IHE Integration Profiles.  *<center> ![](https://i.imgur.com/3a46pFG.png) Fig 2. Conventional medical imaging workflow </center>* Figure 2 lists the six fundamental systems that must be integrated within a typical imaging workflow. The six systems deal with ordering, scheduling, imaging, storing, inspecting, and reporting. In accordance with the IHE Radiology SWF and Reporting Workflow (RWF) profiles, system integration requires support for the DICOM protocol (modality work list, image storage, image query/retrieve) as well as the HL7 V2 protocol (ordering and image availability notification). The system handles imaging reports usually via PDF file, the HL7 Clinical Document Architecture (CDA), or DICOM Structure Reporting (SR) for the storage of reports in the report repository or PACS. It is also important to provide support for IHE XDS or XDS-I profiles using ebXML (Electronic Business using eXtensible Markup Language) web services for the sharing of images and reports among hospitals.  ### 3.2. FHIR and DICOMweb Imaging Workflows *<center> ![](https://i.imgur.com/DgdrSHw.png) Fig 3. Proposed medical imaging workflow </center>* The proposed framework includes two web servers (FHIR and DICOMweb servers) and four clients (Order Placer, Order Filler, Modality, and Viewer). All of the clients and servers use the HL7 FHIR and DICOMweb protocols to facilitate integration, and both servers utilize HTTP RESTful APIs for the sharing of data, which is formatted as XML or JSON and stored on the FHIR server. Data sharing by the DICOMweb server can be in XML, JSON format, or binary data format. Our adoption of RESTful web solutions makes it easy for most information technology (IT) developers to understand the system.  ## 4. Project Planning - A system where radiologist can analyze medical image, from viewing the image, mark annotation, input finding report, until input diagnostic report. In a reversible flow, clinician can also review the diagnostic report while referenced to its finding report, annotation, and image - A system where patient can retrieve their medical image and diagnostic report ## References - [Creating a Medical Imaging Workflow Based on FHIR, DICOMweb, and SVG](https://link.springer.com/article/10.1007/s10278-021-00522-6) - [FHIR Structured Report](https://hackmd.io/@victoriatjia/fhir-sr) - [Liver Observation (LI-RADS)](https://www.acr.org/-/media/ACR/Files/Clinical-Resources/LIRADS/Chapter-7-The-LIRADS-observation.pdf) - [Liver Diagnostic Report Code (CT/MRI LI-RADS)](https://www.acr.org/-/media/ACR/Images/Clinical-Resources/RADS/LI-RADS/LI-RADSv2018.JPG) - [Liver Hepatocellular Carcinoma Finding report template](https://documents.cap.org/protocols/Liver.HCC_4.3.0.0.REL_CAPCP.pdf) - [AI in assessment of hepatocellular carcinoma treatment response](https://www.researchgate.net/publication/350523686_Artificial_intelligence_in_assessment_of_hepatocellular_carcinoma_treatment_response) - Tumor size/volume periodically change as the time passes calculation (benefit for Patient to know their future progress). Compare to AI automatic locate the tumor position, patient will be more interested in this topic https://qibawiki.rsna.org/index.php/Profiles - https://journal.ahima.org/page/dense-breasts-and-coding-mammography