Automated semantic annotation of rare disease cases: a case study

Taboada, M.; Rodriguez, Hadriana; Martínez, Diego; Pardo, María Laura; Sobrido, María Jesús

doi:10.1093/database/bav107

Cited by 8 publications

(28 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We also considered other alternatives to extract these entities such as OBO annotator (Taboada et al, 2014) and Bio-Lark CR (Groza et al, 2015a) for extracting phenotype names, and GNormPlus (Wei et al, 2015) and ABNER (Settles, 2005) for extracting protein names. However, most of these systems either did not provide desirable results or were difficult to access or use.…”

Section: Methodsmentioning

confidence: 99%

ProPheno 1.0: An online dataset for accelerating the complete characterization of the human protein-phenotype landscape in biomedical literature

Shahri

Kahanda

2019

Preprint

View full text Add to dashboard Cite

Identifying protein-phenotype relations is of paramount importance for applications such as uncovering rare and complex diseases. One of the best resources that captures the protein-phenotype relationships is the biomedical literature. In this work, we introduce ProPheno, a comprehensive online dataset composed of human protein/phenotype mentions extracted from the complete corpora of Medline and PubMed Central Open Access. Moreover, it includes co-occurrences of protein-phenotype pairs within different spans of text such as sentences and paragraphs. We use ProPheno for completely characterizing the human protein-phenotype landscape in biomedical literature. ProPheno, the reported findings and the gained insight has implications for (1) biocurators for expediting their curation efforts, (2) researches for quickly finding relevant articles, and (3) text mining tool developers for training their predictive models. The RESTful API of ProPheno is freely available at http://propheno.cs.montana.edu.

show abstract

Section: Methodsmentioning

confidence: 99%

ProPheno 1.0: An online dataset for accelerating the complete characterization of the human protein-phenotype landscape in biomedical literature

Shahri

Kahanda

2019

Preprint

View full text Add to dashboard Cite

show abstract

“…Six of the thirteen CDSS included in the analysis use phenotypic and genetic matching as a CDSS functionality [16][17][18][19][20][21] whereas three of these systems additionally make use of clinical data [18,20,21]. Five of the CDSS are based on machine learning and information retrieval [22][23][24][25][26] and two use web search methods [27,28]. Six systems are clinical prototypes [22-24, 26, 28] and seven are full developed [16][17][18][19][20][21].…”

Section: Functionality Development Status Type Of Clinical Data Symentioning

confidence: 99%

“…Six systems are clinical prototypes [22-24, 26, 28] and seven are full developed [16][17][18][19][20][21]. Regarding "Type of clinical data", two systems use literature databases [27,28] and two use clinical data [24,25], while one of these system uses both clinical and literature data [26]. Two CDSS use patient questionnaires [22,23].…”

Section: Functionality Development Status Type Of Clinical Data Symentioning

confidence: 99%

“…This process is timeconsuming and inefficient and tools to find and compare these reports would be helpful. In this review, we identified two fully developed CDSS adopting this idea of patients with RD: FindZebra [27] and a system from Taboada et al [28].…”

Section: Functionality Web Searchmentioning

confidence: 99%

“…Dragusin et al [27] FindZebra uses a larger portfolio of data with 10 different data sources. While FindZebra can be accessed directly online [24], the use of the search engine by Taboada et al [28] requires a download.…”

Section: Functionality Web Searchmentioning

confidence: 99%

See 2 more Smart Citations

Diagnosis of Rare Diseases: A scoping review of clinical decision support systems

Schaaf

Sedlmayr

Schaefer

et al. 2019

Preprint

View full text Add to dashboard Cite

Background Rare Diseases (RD), which are defined as diseases affecting not more than 5 out of 10,000 people, are often severe, chronic, degenerative and life-threating. A main problem is the delay in diagnosis of RD. Clinical Decision Support Systems (CDSS) for RD are software-systems to support physicians in the diagnosis of patients with RD. It would therefore be useful to get a comprehensive overview of which CDSS are available and can be used under what conditions. In this work we provide a review of current CDSS in RD and which functionality and data are used by the CDSS. Methods We searched Pubmed and Cochrane for CDSS in RD published between December 1, 2008 and December 16, 2018. Only English articles, original peer reviewed journals and conference paper describing a clinical prototype or a routine use of CDSS where included. A total of 2076 articles were found and following a screening step 16 articles (describing 13 different CDSS) were considered as relevant for the final analysis. We then described and compared the CDSS using the defined categories “functionality”, “development status”, “type of clinical data” and “system availability”. Results Three types of CDSS for RD were identified: “Machine Learning and Information retrieval”, “Web Search”, and “Phenotypic and genetic matching”. 8 of the 13 reviewed CDSS are publicly available and for use by physicians. The other remaining CDSS are clinical prototypes which have been applied in clinical studies but are not accessible to others. Only one clinical prototype online. The approaches of the CDSS differ depending on what type of clinical data is used. “Machine Learning and information retrieval” can show recommendations for a diagnosis, while Web “search CDSS” will retrieve articles from literature databases (e.g. case reports), which may provide hints for a possible Diagnosis. CDSS in “Phenotypic and genetic matching” can identify similar patients based on genetic or phenotypic data. Conclusions Different CDSS for different purposes have been established and physicians have to decide which CDSS is more accurate for a particular patient case. It remains to be seen which of the CDSS will be used and maintained in the future.

show abstract

PhenoTagger: a hybrid method for phenotype concept recognition using human phenotype ontology

et al. 2021

View full text Add to dashboard Cite

Motivation Automatic phenotype concept recognition from unstructured text remains a challenging task in biomedical text mining research. Previous works that address the task typically use dictionary-based matching methods, which can achieve high precision but suffer from lower recall. Recently, machine learning-based methods have been proposed to identify biomedical concepts, which can recognize more unseen concept synonyms by automatic feature learning. However, most methods require large corpora of manually annotated data for model training, which is difficult to obtain due to the high cost of human annotation. Results In this article, we propose PhenoTagger, a hybrid method that combines both dictionary and machine learning-based methods to recognize Human Phenotype Ontology (HPO) concepts in unstructured biomedical text. We first use all concepts and synonyms in HPO to construct a dictionary, which is then used to automatically build a distantly supervised training dataset for machine learning. Next, a cutting-edge deep learning model is trained to classify each candidate phrase (n-gram from input sentence) into a corresponding concept label. Finally, the dictionary and machine learning-based prediction results are combined for improved performance. Our method is validated with two HPO corpora, and the results show that PhenoTagger compares favorably to previous methods. In addition, to demonstrate the generalizability of our method, we retrained PhenoTagger using the disease ontology MEDIC for disease concept recognition to investigate the effect of training on different ontologies. Experimental results on the NCBI disease corpus show that PhenoTagger without requiring manually annotated training data achieves competitive performance as compared with state-of-the-art supervised methods. Availabilityand implementation The source code, API information and data for PhenoTagger are freely available at https://github.com/ncbi-nlp/PhenoTagger. Supplementary information Supplementary data are available at Bioinformatics online.

show abstract

Automated semantic annotation of rare disease cases: a case study

Cited by 8 publications

References 0 publications

ProPheno 1.0: An online dataset for accelerating the complete characterization of the human protein-phenotype landscape in biomedical literature

ProPheno 1.0: An online dataset for accelerating the complete characterization of the human protein-phenotype landscape in biomedical literature

Diagnosis of Rare Diseases: A scoping review of clinical decision support systems

PhenoTagger: a hybrid method for phenotype concept recognition using human phenotype ontology

Contact Info

Product

Resources

About