PubCaseFinder: A Case-Report-Based, Phenotype-Driven Differential-Diagnosis System for Rare Diseases

Abstract

Recently, to speed up the differential-diagnosis process based on symptoms and signs observed from an affected individual in the diagnosis of rare diseases, researchers have developed and implemented phenotype-driven differential-diagnosis systems. The performance of those systems relies on the quantity and quality of underlying databases of disease-phenotype associations (DPAs). Although such databases are often developed by manual curation, they inherently suffer from limited coverage. To address this problem, we propose a text-mining approach to increase the coverage of DPA databases and consequently improve the performance of differential-diagnosis systems. Our analysis showed that a text-mining approach using one million case reports obtained from PubMed could increase the coverage of manually curated DPAs in Orphanet by 125.6%. We also present PubCaseFinder (see Web Resources), a new phenotype-driven differential-diagnosis system in a freely available web application. By utilizing automatically extracted DPAs from case reports in addition to manually curated DPAs, PubCaseFinder improves the performance of automated differential diagnosis. Moreover, PubCaseFinder helps clinicians search for relevant case reports by using phenotype-based comparisons and confirm the results with detailed contextual information.

AMELIE 2 speeds up Mendelian diagnosis by matching patient phenotype & genotype to primary literature

“…The set of 215 patients obtained from the DDD study was used to evaluate AMELIE against Exomiser (14), Phenolyzer (15), Phen-Gen (16), eXtasy (17), and PubCaseFinder (18). The output of all methods, consisting of a list of ranked genes, was subset to the (median) 127 candidate genes that AMELIE used for each patient based on the filtering criteria previously described (Figure 2a).…”

“…Manually curated databases (6)(7)(8)(9)(10) are utilized to more efficiently access extracts of the unstructured knowledge in the primary literature. Automatic gene ranking tools (11)(12)(13)(14)(15)(16)(17)(18) use these databases to prioritize candidate genes in patients' genomes for their ability to explain patient phenotypes. An important feature of many gene ranking tools is the use of phenotype match functions on patient phenotypes and gene/disease-associated phenotypes.…”

Diagnosis support systems for rare diseases: a scoping review

“…Among 19 knowledge-based systems, only 8 studies validated their model on real cases [2,46,54,55,61,67,71,77]. Among the 11 remaining studies, 5 studies considered simulated patients (queries consisted of a list of phenotype concepts).…”

“…However, this process was clearly specified in very few cases. Updates were either based on manual review of new information by experts [54,61] or automatically performed based on case reports retrieval from PubMed [71] or by downloading raw data from Orphanet [70].…”

Learning Contextualized Document Representations for Healthcare Answer Retrieval

“…In a clinical decision support system (CDSS), doctors and healthcare professionals require access to information from heterogeneous sources, such as research papers [16,40], electronic health records Note that the answer is part of a longer document and there is almost no word overlap between query and answer passage. [19], clinical case reports [13], reference works and knowledge base articles. Differential diagnosis is an important task where a doctor seeks to retrieve answers for non-factoid queries about diseases, such as "symptoms of IgA nephropathy" (see Figure 1).…”