=Paper=
{{Paper
|id=Vol-3630/paper14
|storemode=property
|title=Patient trajectory visualization for FHIR healthcare data: A use case on melanoma patients
|pdfUrl=https://ceur-ws.org/Vol-3630/LWDA2023-paper14.pdf
|volume=Vol-3630
|authors=Meijie Li,Wolfgang Galetzka,Bahadir Eryilmaz,Georg Christian Lodde,Elisabeth Livingstone,Jörg Schlötterer,Christin Seifert
|dblpUrl=https://dblp.org/rec/conf/lwa/LiGELLSS23
}}
==Patient trajectory visualization for FHIR healthcare data: A use case on melanoma patients==
https://ceur-ws.org/Vol-3630/LWDA2023-paper14.pdf
Patient trajectory visualization for FHIR healthcare
data: A use case on melanoma patients
Meijie Li1,* , Wolfgang Galetzka2 , Bahadir Eryilmaz1 , Georg Christian Lodde4 ,
Elisabeth Livingstone4 , Jörg Schlötterer5,6 and Christin Seifert5
1
University of Duisburg-Essen, Forsthausweg 2, 47057 Duisburg, Germany
2
IMIBE, Unversity Hospital Essen, Hufelandstraße 55, 45147 Essen, Germany
4
Department of Dermatology, University Hospital Essen, 45147 Essen, Germany
5
University of Marburg, Hans-Meerwein-Straße 6, 35032 Marburg, Germany
6
University of Mannheim, Area Information Systems, L15 1-6, 68161 Mannheim, Germany
Abstract
Fast Healthcare Interoperability Resources (FHIR) is gaining popularity as a standard framework for
the exchange of electronic health record (EHR) data. Despite the advantages of FHIR, it is difficult
for clinicians to understand the data in EHR. To support clinicians in accessing data about a patient,
we created a pipeline that extracts, transforms, and visualizes patient data from FHIR. We employ a
web-based timeline visualization that shows all clinical data recorded for the patient over their disease
trajectory. This can help clinicians to use the patient data more efficiently and to get a clear picture of
the patient’s disease progress and physical condition more quickly, which could help them to develop the
best treatment plan for their patients. The source code with an example synthetic, but realistic patient is
available at https://github.com/rtg-wispermed/Patient_trajectory_public.
Keywords
Clinical data visualization, Electronic Health Records, Patient history visualization, FHIR
1. Introduction
Since the enactment of the Health Information Technology for Economic and Clinical Health
(HITECH) Act of 2009 [1], the adoption of electronic health records (EHR) in hospitals has
grown rapidly [2]. The EHR contains data for each patient, including diagnoses, lab tests and
results, prescriptions, clinical notes and radiology images [1].
FHIR1 (Fast Healthcare Interoperability Resource) is a standard for health care data exchange.
While FHIR allows for comprehensible capture of patient information, the resulting structure is
rather complex and difficult to comprehend. Thus, access to patient data essentially requires
LWDA 2023 Workshops: BIA, DB, IR, KDML and WM. Marburg, Germany, 09.-11. October 2023
*
Corresponding author.
$ meijie.li@uni-due.de (M. Li); wolfgang.galetzka@uk-essen.de (W. Galetzka); bahadir.eryilmaz@uk-essen.de
(B. Eryilmaz); georg.lodde@uk-essen.de (G. Lodde); elisabeth.livingstone@uk-essen.de (E. Livingstone);
joerg.schloetterer@uni-marburg.de (J. Schlötterer); christin.seifert@uni-marburg.de (C. Seifert)
https://wispermed.com/author/meijie-li/ (M. Li)
0009-0004-3037-9207 (M. Li); 0009-0002-8743-4751 (B. Eryilmaz); 0000-0002-3678-0390 (J. Schlötterer);
0000-0002-6776-3868 (C. Seifert)
© 2023 Copyright © 2023 by the paper’s authors. Copying permitted only for private and academic purposes. In: M. Leyer, Wichmann, J. (Eds.): Proceedings of
the LWDA 2023 Workshops: BIA, DB, IR, KDML and WM. Marburg, Germany, 09.-11. October 2023, published at http://ceur-ws.org
CEUR
Workshop
CEUR Workshop Proceedings (CEUR-WS.org)
Proceedings
http://ceur-ws.org
ISSN 1613-0073
1
http://www.hl7.org/fhir, last accessed 01.06.2023
CEUR
ceur-ws.org
Workshop ISSN 1613-0073
Proceedings
tools that i) process FHIR data, ii) select and transform it for the use case at hand, and iii) display
the relevant data in a user-friendly manner.
Visualizations allow users to understand information more effectively, thereby improving
judgment and decision-making performance [3]. This is also true for healthcare data, where
an efficient display of information is critical for clinicians to understand patient histories [4].
There has been a significant amount of research work on healthcare data visualization [5, 6].
Among them, Cousins and Kahn [7] first introduced the concept of graphical clinical timelines;
the LifeLines project [8, 9] and the Health Timeline project [4] further developed the idea of
visualizing patient data on the basis of timelines. And in a later evaluation study based on the
Health Timeline project [10], it was demonstrated that visualizing clinical data chronologically
can effectively help physicians improve their understanding of clinical data and help them
identify complex patterns from the data. Prior work on visualizations mostly focused on the
design of the presentation and graphics of the data visualization and did not support the pipeline
to FHIR encoded EHR data.
In this paper, we present a proof-of-concept for a visualization pipeline that obtains patient
data from EHR in FHIR and apply it to a use case of melanoma patients. In place of the long-term
and complex development of visualization tools, we utilize the publicly available, web-based
charting framework AnyChart2 .
2. Approach
Our pipeline consists of four steps (cf. Figure 1): i) patient selection, ii) relevance judgment of
attributes by domain experts, iii) extraction of relevant attributes, and iv) visualization.
Figure 1: Pipeline. Patients are selected based on exclusion and inclusion criteria and exported to a
temporary FHIR database. Based on expert judgments of attribute relevancy, the data is filtered and
transformed to a simple JSON file, which is used as input for the patient trajectory visualization.
First, the patient cohort of interest is selected and exported from the hospital EHR database
based on defined inclusion and exclusion criteria. This export is done using high-level FHIR
APIs, such as FHIR-PYrate3 or fhircrackr4 . The primary stakeholder of our visualization are clin-
icians, each with specific information needs about their patients. EHRs contain comprehensive
information about patients, including information that is not relevant for a first assessment. To
identify the relevant information for the visualization, we let clinicians rate the available
information. Clinicians rate each available attribute on a 3-point Likert scale (0: not important,
2
https://github.com/AnyChart, last accessed 12.07.2023
3
https://github.com/UMEssen/FHIR-PYrate, last accessed 07.06.2023
4
https://github.com/POLAR-fhiR/fhircrackr, last accessed 07.06.2023
1: moderately important, 2: very important). We then selected the attributes that at least one
clinician considered very important. In the whole process, domain experts only need to do the
relevance judgment of attributes once for all the patients in one patient cohort.
The selected attributes are then extracted from FHIR and transformed into a simple
JSON file. First, the resources were flattened, i.e. we changed the nested structure of the
original FHIR resource such that each field in the newly transformed JSON only contains atomic
values to make it easier readable and usable. Secondly, we removed unnecessary and redundant
information based on the patient ID and date. Lastly, in cases where information stems from
the same event but is distributed between multiple resources, the information is merged.
For the visualization of the extracted attributes as a timeline, we use the AnyChart
JavaScript visualization library2 . The AnyChart library provides interactive functionality de-
signed specifically for temporal data and supports both, time point events (a specific point
in time) and time range events. We map the patient information accordingly, e.g., a physical
examination is encoded as a time point event, and a medication plan with start and end date to
a time range event. AnyChart supports zoom, to adjust the time range showing either a global
view or focusing on a particular time period only.
3. Use Case: Melanoma Patients
In this section, we will describe the application of our approach to a specific use case. Due
to the confidentiality of patient data, we use a synthetic, but realistic patient to showcase the
visualization.
We use the FHIR-based melanoma database from the University Hospital in Essen. As
selection criteria, we used the primary tumor diagnosis to only include melanoma patients:
Condition.category = |C0677930 (Primary Neoplasm) AND Condition.code=
C43.*. Each file contains the complete tumor documentation for a patient and other relevant
information (e.g. laboratory observations and progress notes). Overall, our cohort includes all
possible patients with the start of treatment between 2001 and 2023, with the majority of the
cases being from 2014 to 2023 (about 90%). In total, there are 1899 melanoma patients (46.18%
female, 53.77% male) in the data set. The vast majority is from the North Rhine-Westphalia.
An example of the relevance judgment of patient attributes is shown in Table 1 (see Appendix A
for the full list). Two experienced dermato-oncologists have rated the attributes based on the
national guidelines for melanoma treatment5 . We selected the attributes which at least one
clinician rated as very important for inclusion in the patient trajectory visualization. An example
of JSON output based on these selected attributes can be seen in Appendix B.
The visualization of the patient trajectory shows general patient information along with the
timeline. Figure 2, top, shows the overview including all attributes selected in the relevance
judgment step. It presents core diagnostic and treatment information for the patient. To further
facilitate clinicians to select the information of interest without being distracted by other
information, we also added a filter function, to show only one special attribute or one special
combination of different attributes (cf. Figure 2, bottom).
5
http://www.leitlinienprogrammonkologie.de/leitlinien/melanom, last accessed 31.08.2023
Table 1
Example relevance judgment of attributes (excerpt, translated to English)
Attributes Clinician 1 Clinician 2 Included
Careplan
Time 2 1 ✓
Type (very general: e.g., surgery, best supportive 1 1
care, radiotherapy)
Intention (palliative, curative) 2 2 ✓
Category (specific, e.g. treatment of relapses) 2 1 ✓
...
Legend:
not important 0 0 moderately important 1 1 very important 2 2 ✓: selected
Figure 2: Timeline of an example patient showing the all selected out attributes (cropped due to space
constraints). Bottom: Showing only disease progress. Patient meta information (name etc. omitted).
4. Summary
This work presents a method to visualize FHIR healthcare data as patient trajectory which aims
to support clinicians in accessing information from EHR (available in FHIR) and in obtaining a
comprehensive and concise picture of the patient’s disease progress and physical condition. The
overall framework is extensible and applicable to other patient cohorts and use cases. However
it also has some potential limitations, e.g., relevant attributes might need to be updated due to
the changing of guidelines of treatment and ICD versions, and the choice of web technology,
some browsers may not support JavaScript. An evaluation of the usability, usefulness and
impact of this visualization is planned for future work.
References
[1] G. S. Birkhead, M. Klompas, N. R. Shah, Uses of electronic health records for public health
surveillance to advance public health, Annual review of public health 36 (2015) 345–359.
[2] J. Henry, Y. Pylypchuk, T. Searcy, V. Patel, et al., Adoption of electronic health record
systems among us non-federal acute care hospitals: 2008–2015, ONC data brief 35 (2016)
2008–2015.
[3] K. Eberhard, The effects of visualization on judgment and decision-making: a systematic
literature review, Management Review Quarterly 73 (2023) 167–214.
[4] A. A. Bui, D. R. Aberle, H. Kangarloo, Timeline: visualizing integrated patient records,
IEEE Transactions on Information Technology in Biomedicine 11 (2007) 462–473.
[5] V. L. West, D. Borland, W. E. Hammond, Innovative information visualization of electronic
health record data: a systematic review, Journal of the American Medical Informatics
Association 22 (2015) 330–339.
[6] Q. Wang, R. S. Laramee, Ehr star: the state-of-the-art in interactive ehr visualization, in:
Computer Graphics Forum, volume 41, Wiley Online Library, 2022, pp. 69–105.
[7] S. B. Cousins, M. G. Kahn, The visual display of temporal information, Artificial intelligence
in medicine 3 (1991) 341–357.
[8] C. Plaisant, R. Mushlin, A. Snyder, J. Li, D. Heller, B. Shneiderman, Lifelines: using
visualization to enhance navigation and analysis of patient records, in: The craft of
information visualization, Elsevier, 2003, pp. 308–312.
[9] C. Cheng, Y. Shahar, A. Puerta, D. Stites, Navigation and visualization of abstractions of
time-oriented clinical data, Section on Medical Informatics Technical Report No. SMI-97
688 (1997).
[10] A. Ledesma, N. Bidargaddi, J. Strobel, G. Schrader, H. Nieminen, I. Korhonen, M. Ermes,
Health timeline: an insight-based study of a timeline visualization of clinical data, BMC
medical informatics and decision making 19 (2019) 1–14.
A. Full List of Relevance Judgments
Table 2: Relevance judgment of all attributes (translated to English)
Attributes Clinician 1 Clinician 2 Included
Patient
Gender 2 2 ✓
Birth date 2 2 ✓
Deceased time 2 2 ✓
Careplan
Time 2 1 ✓
Type (very general: e.g., surgery, best support- 1 1
ive care, radiotherapy)
Intention (palliative, curative) 2 2 ✓
Category (specific, e.g., Treatment of relapses) 2 1 ✓
Medication
Name 2 2 ✓
Start time 2 2 ✓
End time 2 2 ✓
Status (e.g., stopped/completed) 2 1 ✓
Reason of status (e.g., Abort due to progress) 2 2 ✓
Intention of Medication (e.g., Palliative / Cu- 2 2 ✓
rative)
Quantity (how many times did they get the 2 1 ✓
medication)
Condition
ICD Tumor and Location code 2 2 ✓
Tumor Location on the body 1 1
Multiple Locations 1 1
L2Obs–Progression w.r.t Time 2 2 ✓
Comorbidities 1 1
Tumour stage
Pathological TNM staging (pTNM) 2 2 ✓
Version of pTNM 1 2 ✓
T, N, M Stage separately (pTNM) 1 2 ✓
Residual-State (pTNM) 0 1
Sentinel Lymph nodes positive (pTNM) 2 2 ✓
Sentinel Lymph nodes examined (pTNM) 1 1
Regional Lymph nodes positive (pTNM) 2 1 ✓
Regional Lymph nodes examined (pTNM) 1 1
Clinical TNM staging (cTNM) 1 1
Version of cTNM 1 1
T, N, M Stage separately (cTNM) 1 1
Oncogenes
Results of genetic analysis, if genetic findings 2 2 ✓
frequency and name of the mutation
Progress
Result of the regular check up (e.g., No tumor 2 2 ✓
detection, Questionable findings, Progress,
Decline, etc.)
Procedures, Radiotherapy
Intention (e.g., adjuvant, curative) 2 1 ✓
Status (completed or not) 2 1 ✓
Reason of status 2 1 ✓
Details of on how it was performed(e.g., with 2 1 ✓
Cyberknife)
Procedures, Surgery
Ops-Code (e.g., Code for removal of Lymph- 1 1
Node)
Residual state 1 1
Procedures, Examinations
Type of examination (e.g., Ultrasound) 2 2 ✓
Reason for examinations (e.g., Toxicity Assess- 1 1
ment)
Primary properties
Ulceration of primary 1 2 ✓
Tumor thickness (in mm) 1 1
Regression 0 0
Re-excision 1 1
Transcapsular (Capsular Breakthrough) 1 2 ✓
Mitosis rate 1 1
%PD1 1 1
Metastases properties
Location 2 1 ✓
Proof (tumor detection, no tumor detection) 2 1 ✓
Type of proof (e.g., Imaging) 2 1 ✓
Legend:
0: Not important
1: Moderately important
2: Very important
✓: selected
B. Example JSON File
Example data extract from a synthetic, but realistic patient (excerpt due to space constraints)
{
"patient_info": {
"resourceType": "patient",
"id": "01",
"gender": "female",
"birthDate": "1961-05-30",
"deceasedDateTime": "2019-07-21"
},
"stages": [
{
"patid": "Patient/01",
"dt_record": "2013-12",
"cat_version": null,
"tnm_stage": "III",
"tstage": "2 B",
"nstage": "0",
"p_or_c": "p",
"val_print": "Version: k.A., pIII T2 BN0M k.A. ",
"dt_end": "2013-12-15"
},
...
],
"examinations": [
{
"patid": "Patient/01",
"dt_record": "2014-02-06",
"cat_examination_type": [
"Laboratory procedure",
"physical exam"
],
"cat_reasons": [
"Initial presentation",
"Treatment planning"
]
},
...
],
"radiotherapy": [
{
"patid": "Patient/01",
"dt_start": "2016-06",
"dt_end": "2016-06-29",
"cat_intention": [
"palliative",
"adjuvant"
],
"cat_status": "completed",
"cat_reason_end": "regular ending"
},
...
],
...
}