PubMed Health⌕ Search
Home/Archive/PubMed

PubMed · 16779084

Extracting diagnoses from discharge summaries.

William Long
View original source ↗Export metadata ↓

Abstract

We have developed a program for extracting the diagnoses and procedures from the past medical history and discharge diagnoses in the discharge summary of a case and coding these using SNOMED-CT in the UMLS. The program uses a limited amount of natural language processing. Rather, it makes use of the relatively standard structure of the discharge summary, a small dictionary to divide the text into phrases, and the extensive collection of phrases for concepts in the UMLS to do the coding. With this approach the program finds 240 of 250 desired concepts with 19 false positives in 23 discharge summaries.

Explore related subjects

DiagnosisDiseaseElectronic Data ProcessingForms and Records ControlHumansMedical Informatics ApplicationsNatural Language ProcessingPatient DischargeSoftwareSystematized Nomenclature of MedicineUnified Medical Language SystemJournal ArticleResearch Support, N.I.H., Extramural

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

William Long. 2005. Extracting diagnoses from discharge summaries.. https://pubmed.ncbi.nlm.nih.gov/16779084/

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

PubMed2025

[Achievements and Expectations of the Rare Disease Diagnostic Support Program in the Republic of Korea].

OBJECTIVES: The Rare Disease Diagnostic Support Program in the Republic of Korea aims to improve early diagnosis and diagnostic yield for patients with rare diseases, particularly for those residing in non-metropolitan areas, by providing whole genome sequencing (WGS) services through regional medical institutions. This study evaluated the performance of the program, focusing on its clinical utility, including early diagnosis and treatment linkage, and its policy impact related to patient benefits. METHODS: From August 2024, WGS was performed on 410 patients with suspected rare diseases at 23 institutions outside the metropolitan area. A one-stop diagnostic pathway was established to perform sample collection, test referral, report delivery, and genetic counseling within a single clinical flow based on the patient’s location of residence. Sequencing was performed by external laboratories. RESULTS: Among the 410 patients, pathogenic variants were identified in 129 (31.5%), with a turnaround time of 28 days. Of those diagnosed, 78.2% received treatment benefits via national programs such as co-payment exemption and medical expense support programs. Approximately 30% of the patients were eligible for therapeutic intervention, particularly medication or dietary therapy. Family genetic testing of three members identified potential carriers or high-risk groups in 28 households (65.1%). Consent for secondary findings was 99.0%, with clinically significant variants found in 3.9% of cases. CONCLUSIONS: The program demonstrated clinical value by improving diagnostic accessibility, reducing regional disparities, facilitating timely treatment, and supporting preventive care through family risk identification. These findings support the need for sustainable expansion of genome-based diagnostic services in the national health policy.

Diagnosis↗
PubMed2007

A randomized controlled trial of the diagnostic accuracy of internet-based telepathology compared with conventional microscopy.

AIMS: To compare the diagnostic accuracy of internet-based virtual microscopy with conventional light microscopy, in the context of renal biopsies assessed by participants in the UK National Renal Pathology External Quality Assessment (EQA) Scheme. METHODS AND RESULTS: 'Virtual slides' of current EQA cases were made available over the internet to participants in the scheme. The approach permitted what may reasonably be described as a randomized controlled trial of the diagnostic accuracy of this mode of telepathology, in the context of renal pathology. No significant difference in diagnostic accuracy could be detected between the diagnoses proffered on the basis of virtual slides and conventional slides; but using virtual slides took pathologists considerably longer. CONCLUSIONS: This result provides some encouragement for the implementation of such virtual slide-based telepathology systems. However, in failing to detect a difference, equivalence of diagnostic accuracy has not been proved; nor has the relevance of this result to other fields of histopathology been demonstrated. A decision to implement a diagnostic telepathology system requires consideration of numerous factors beyond diagnostic accuracy, including financial, legal, professional and ethical issues.

Diagnosis↗