Using a Large Open Clinical Corpus for Improved ICD-10 Diagnosis Coding.

With the recent advances in natural language processing and deep learning, the development of tools that can assist medical coders in ICD-10 diagnosis coding and increase their efficiency in coding discharge summaries is significantly more viable than before. To that end, one important component in the development of these models is the datasets used to train them. In this study, such datasets are presented, and it is shown that one of them can be used to develop a BERT-based language model that can consistently perform well in assigning ICD-10 codes to discharge summaries written in Swedish. Most importantly, it can be used in a coding support setup where a tool can recommend potential codes to the coders. This reduces the range of potential codes to consider and, in turn, reduces the workload of the coder. Moreover, the de-identified and pseudonymised dataset is open to use for academic users.

De-identifying Norwegian Clinical Text using Resources from Swedish and Danish.

The lack of relevant annotated datasets represents one key limitation in the application of Natural Language Processing techniques in a broad number of tasks, among them Protected Health Information (PHI) identification in Norwegian clinical text. In this work, the possibility of exploiting resources from Swedish, a very closely related language, to Norwegian is explored. The Swedish dataset is annotated with PHI information. Different processing and text augmentation techniques are evaluated, along with their impact in the final performance of the model. The augmentation techniques, such as injection and generation of both Norwegian and Scandinavian Named Entities into the Swedish training corpus, showed to increase the performance in the de-identification task for both Danish and Norwegian text. This trend was also confirmed by the evaluation of model performance on a sample Norwegian gastro surgical clinical text.

Design and Prestudy Assessment of a Dashboard for Presenting Self-Collected Health Data of Patients With Diabetes to Clinicians: Iterative Approach and Qualitative Case Study.

BACKGROUND: Introducing self-collected health data from patients with diabetes into consultation can be beneficial for both patients and clinicians. Such an initiative can allow patients to be more proactive in their disease management and clinicians to provide more tailored medical services. Optimally, electronic health record systems (EHRs) should be able to receive self-collected health data in a standard representation of medical data such as Fast Healthcare Interoperability Resources (FHIR), from patients systems like mobile health apps and display the data directly to their users-the clinicians. However, although Norwegian EHRs are working on implementing FHIR, no solution or graphical interface is available today to display self-collected health data. OBJECTIVE: The objective of this study was to design and assess a dashboard for displaying relevant self-collected health data from patients with diabetes to clinicians. METHODS: The design relied on an iterative participatory process involving workshops with patients, clinicians, and researchers to define which information should be available and how it should be displayed. The assessment is based on a case study, presenting an instance of the dashboard populated with data collected from one patient with diabetes type 1 (in-house researcher) face-to-face by 14 clinicians. We performed a qualitative analysis based on usability, functionality, and expectation by using responses to questionnaires that were distributed to the 14 clinicians at the end of the workshops and collected before the participants left. The qualitative assessment was guided by the Standards for Reporting Qualitative Research. RESULTS: We created a dashboard permitting clinicians to assess the reliability of self-collected health data, list all collected data including medical calculations, and highlight medical situations that need to be investigated to improve the situation of the patients. The dashboard uses a combination of tables, graphs, and other visual representations to display the relevant information. Clinicians think that this type of solution will be useful during consultations every day, especially for patients living in remote areas or those who are technologically interested. CONCLUSIONS: Displaying self-collected health data during consultations is not enough for clinicians; the data reliability has to be assured and the relevant information needs to be extracted and displayed along with the data to ease the introduction during a medical encounter. The prestudy assessment showed that the system provides relevant information to meet clinicians' need and that clinicians were eager to start using it during consultations. The system has been under testing in a medical trial since November 2018, and the first results of its assessment in a real-life situation are expected in the beginning of next year (2020).

How Do Experienced Physicians Access and Evaluate Laboratory Test Results for the Chronic Patient? A Qualitative Analysis.

BACKGROUND: Electronic health records may present laboratory test results in a variety of ways. Little is known about how the usefulness of different visualizations of laboratory test results is influenced by the complex and varied process of clinical decision making. OBJECTIVE: The purpose of this study was to investigate how clinicians access and utilize laboratory test results when caring for patients with chronic illness. METHODS: We interviewed 10 attending physicians about how they access and assess laboratory tests when following up patients with chronic illness. The interviews were audio-recorded, transcribed verbatim, and analyzed qualitatively. RESULTS: Informants preferred different visualizations of laboratory test results, depending on what aspects of the data they were interested in. As chronic patients may have laboratory test results that are permanently outside standardized reference ranges, informants would often look for significant change, rather than exact values. What constituted significant change depended on contextual information (e.g., the results of other investigations, intercurrent diseases, and medical interventions) spread across multiple locations in the electronic health record. For chronic patients, the temporal relations between data could often be of special interest. Informants struggled with finding and synthesizing fragmented information into meaningful overviews. CONCLUSION: The presentation of laboratory test results should account for the large variety of associated contextual information needed for clinical comprehension. Future research is needed to improve the integration of the different parts of the electronic health record.

Presentation of clinical laboratory results: an experimental comparison of four visualization techniques.

OBJECTIVE: To evaluate how clinical chemistry test results were assessed by volunteers when presented with four different visualization techniques. MATERIALS AND METHODS: A total of 20 medical students reviewed quantitative test results from 4 patients using 4 different visualization techniques in a balanced, crossover experiment. The laboratory data represented relevant patient categories, including simple, emergency, chronic and complex patients. Participants answered questions about trend, overall levels and covariation of test results. Answers and assessment times were recorded and participants were interviewed on their preference of visualization technique. RESULTS: Assessment of results and the time used varied between visualization techniques. With sparklines and relative multigraphs participants made faster assessments. With relative multigraphs participants identified more covarying test results. With absolute multigraphs participants found more trends. With sparklines participants more often assessed laboratory results to be within reference ranges. Different visualization techniques were preferred for the four different patient categories. No participant preferred absolute multigraphs for any patient. DISCUSSION: Assessments of clinical chemistry test results were influenced by how they were presented. Importantly though, this association depended on the complexity of the result sets, and none of the visualization techniques appeared to be ideal in all settings. CONCLUSIONS: Sparklines and relative multigraphs seem to be favorable techniques for presenting complex long-term clinical chemistry test results, while tables seem to suffice for simpler result sets.