Is YouTube a reliable tool for approaching robotic assisted transabdominal preperitoneal surgery? A critical review of the available resources.

INTRODUCTION: The robotic transabdominal preperitoneal approach (rTAPP) is a relatively recent technique for the treatment of inguinal hernia. To achieve optimal results, the 10 golden rules described must be followed. Surgeons in training often review videos to familiarize themselves with new techniques, YouTube being one of the most used platforms. The objective of this study is to carry out an evaluation of the 10 most viewed videos on YouTube of inguinal hernia repair by transabdominal preperitoneal approach (rTAPP) to determine if the 10 golden rules are met. METHODS: Identify and evaluate the 10 videos with the highest number of views related to rTAPP. Three experienced Surgeons evaluated compliance with the 10 golden rules using a Likert scale. Data were analyzed in Excel (Microsoft) and plotted with Tableau (Tableau Inc). The consistency between evaluators was determined using Cronbach's alpha, considering a value >0.7 acceptable. RESULTS: The average overall evaluation was 3.63 with a range of 2.6 to 4.9. The scores related to compliance with the rules 1, 2, 9, 10 were satisfactory; on the other hand, rules 3, 4, 5, 7 and 8 were weak, particularly rule number 7. Internal consistency was observed between raters with a Cronbach's alpha of 0.98. CONCLUSIONS: The lack of compliance with the 10 golden rules in most of the videos demonstrates that the use of videos (YouTube) is not an adequate resource for learning robot-assisted inguinal hernia cure.

DGH-GO: dissecting the genetic heterogeneity of complex diseases using gene ontology.

BACKGROUND: Complex diseases such as neurodevelopmental disorders (NDDs) exhibit multiple etiologies. The multi-etiological nature of complex-diseases emerges from distinct but functionally similar group of genes. Different diseases sharing genes of such groups show related clinical outcomes that further restrict our understanding of disease mechanisms, thus, limiting the applications of personalized medicine approaches to complex genetic disorders. RESULTS: Here, we present an interactive and user-friendly application, called DGH-GO. DGH-GO allows biologists to dissect the genetic heterogeneity of complex diseases by stratifying the putative disease-causing genes into clusters that may contribute to distinct disease outcome development. It can also be used to study the shared etiology of complex-diseases. DGH-GO creates a semantic similarity matrix for the input genes by using Gene Ontology (GO). The resultant matrix can be visualized in 2D plots using different dimension reduction methods (T-SNE, Principal component analysis, umap and Principal coordinate analysis). In the next step, clusters of functionally similar genes are identified from genes functional similarities assessed through GO. This is achieved by employing four different clustering methods (K-means, Hierarchical, Fuzzy and PAM). The user may change the clustering parameters and explore their effect on stratification immediately. DGH-GO was applied to genes disrupted by rare genetic variants in Autism Spectrum Disorder (ASD) patients. The analysis confirmed the multi-etiological nature of ASD by identifying four clusters of genes that were enriched for distinct biological mechanisms and clinical outcome. In the second case study, the analysis of genes shared by different NDDs showed that genes causing multiple disorders tend to aggregate in similar clusters, indicating a possible shared etiology. CONCLUSION: DGH-GO is a user-friendly application that allows biologists to study the multi-etiological nature of complex diseases by dissecting their genetic heterogeneity. In summary, functional similarities, dimension reduction and clustering methods, coupled with interactive visualization and control over analysis allows biologists to explore and analyze their datasets without requiring expert knowledge on these methods. The source code of proposed application is available at https://github.com/Muh-Asif/DGH-GO.

K-RET: knowledgeable biomedical relation extraction system.

MOTIVATION: Relation extraction (RE) is a crucial process to deal with the amount of text published daily, e.g. to find missing associations in a database. RE is a text mining task for which the state-of-the-art approaches use bidirectional encoders, namely, BERT. However, state-of-the-art performance may be limited by the lack of efficient external knowledge injection approaches, with a larger impact in the biomedical area given the widespread usage and high quality of biomedical ontologies. This knowledge can propel these systems forward by aiding them in predicting more explainable biomedical associations. With this in mind, we developed K-RET, a novel, knowledgeable biomedical RE system that, for the first time, injects knowledge by handling different types of associations, multiple sources and where to apply it, and multi-token entities. RESULTS: We tested K-RET on three independent and open-access corpora (DDI, BC5CDR, and PGR) using four biomedical ontologies handling different entities. K-RET improved state-of-the-art results by 2.68% on average, with the DDI Corpus yielding the most significant boost in performance, from 79.30% to 87.19% in F-measure, representing a P-value of 2.91×10-12. AVAILABILITY AND IMPLEMENTATION: https://github.com/lasigeBioTM/K-RET.

SeEn: Sequential enriched datasets for sequence-aware recommendations.

The recommendation of items based on the sequential past users' preferences has evolved in the last few years, mostly due to deep learning approaches, such as BERT4Rec. However, in scientific fields, recommender systems for recommending the next best item are not widely used. The main goal of this work is to improve the results for the recommendation of the next best item in scientific domains using sequence aware datasets and algorithms. In the first part of this work, we present the adaptation of a previous method (LIBRETTI) for creating sequential recommendation datasets for scientific fields. The results were assessed in Astronomy and Chemistry. In the second part of this work, we propose a new approach to improve the datasets, not the algorithms, to obtain better recommendations. The new hybrid approach is called sequential enrichment (SeEn), which consists of adding to a sequence of items the n most similar items after each original item. The results show that the enriched sequences obtained better results than the original ones. The Chemistry dataset improved by approximately seven percentage points and the Astronomy dataset by 16 percentage points for Hit Ratio and Normalized Discounted Cumulative Gain.

NILINKER: Attention-based approach to NIL Entity Linking.

The existence of unlinkable (NIL) entities is a major hurdle affecting the performance of Named Entity Linking approaches, and, consequently, the performance of downstream models that depend on them. Existing approaches to deal with NIL entities focus mainly on clustering and prediction and are limited to general entities. However, other domains, such as the biomedical sciences, are also prone to the existence of NIL entities, given the growing nature of scientific literature. We propose NILINKER, a model that includes a candidate retrieval module for biomedical NIL entities and a neural network that leverages the attention mechanism to find the top-k relevant concepts from target Knowledge Bases (MEDIC, CTD-Chemicals, ChEBI, HP, CTD-Anatomy and Gene Ontology-Biological Process) that may partially represent a given NIL entity. We also make available a new evaluation dataset designated by EvaNIL, suitable for training and evaluating models focusing on the NIL entity linking task. This dataset contains 846,165 documents (abstracts and full-text biomedical articles), including 1,071,776 annotations, distributed by six different partitions: EvaNIL-MEDIC, EvaNIL-CTD-Chemicals, EvaNIL-ChEBI, EvaNIL-HP, EvaNIL-CTD-Anatomy and EvaNIL-Gene Ontology-Biological Process. NILINKER was integrated into a graph-based Named Entity Linking model (REEL) and the results of the experiments show that this approach is able to increase the performance of the Named Entity Linking model.

Biomedical Relation Extraction With Knowledge Graph-Based Recommendations.

Biomedical Relation Extraction (RE) systems identify and classify relations between biomedical entities to enhance our knowledge of biological and medical processes. Most state-of-the-art systems use deep learning approaches, mainly to target relations between entities of the same type, such as proteins or pharmacological substances. However, these systems are mostly restricted to what they directly identify on the text and ignore specialized domain knowledge bases, such as ontologies, that formalize and integrate biomedical information typically structured as direct acyclic graphs. On the other hand, Knowledge Graph (KG)-based recommendation systems already showed the importance of integrating KGs to add additional features to items. Typical systems have users as people and items that can range from movies to books, which people saw or read and classified according to their satisfaction rate. This work proposes to integrate KGs into biomedical RE through a recommendation model to further improve their range of action. We developed a new RE system, named K-BiOnt, by integrating a baseline state-of-the-art deep biomedical RE system with an existing KG-based recommendation state-of-the-art system. Our results show that adding recommendations from KG-based recommendation improves the system's ability to identify true relations that the baseline deep RE model could not extract from the text. The code supporting this system is available at https://github.com/lasigeBioTM/K-BiOnt.

Prediction of Prostate Cancer Disease Aggressiveness Using Bi-Parametric Mri Radiomics.

Prostate cancer is one of the most prevalent cancers in the male population. Its diagnosis and classification rely on unspecific measures such as PSA levels and DRE, followed by biopsy, where an aggressiveness level is assigned in the form of Gleason Score. Efforts have been made in the past to use radiomics coupled with machine learning to predict prostate cancer aggressiveness from clinical images, showing promising results. Thus, the main goal of this work was to develop supervised machine learning models exploiting radiomic features extracted from bpMRI examinations, to predict biological aggressiveness; 288 classifiers were developed, corresponding to different combinations of pipeline aspects, namely, type of input data, sampling strategy, feature selection method and machine learning algorithm. On a cohort of 281 lesions from 183 patients, it was found that (1) radiomic features extracted from the lesion volume of interest were less stable to segmentation than the equivalent extraction from the whole gland volume of interest; and (2) radiomic features extracted from the whole gland volume of interest produced higher performance and less overfitted classifiers than radiomic features extracted from the lesions volumes of interest. This result suggests that the areas surrounding the tumour lesions offer relevant information regarding the Gleason Score that is ultimately attributed to that lesion.

Text Mining for Building Biomedical Networks Using Cancer as a Case Study.

In the assembly of biological networks it is important to provide reliable interactions in an effort to have the most possible accurate representation of real-life systems. Commonly, the data used to build a network comes from diverse high-throughput essays, however most of the interaction data is available through scientific literature. This has become a challenge with the notable increase in scientific literature being published, as it is hard for human curators to track all recent discoveries without using efficient tools to help them identify these interactions in an automatic way. This can be surpassed by using text mining approaches which are capable of extracting knowledge from scientific documents. One of the most important tasks in text mining for biological network building is relation extraction, which identifies relations between the entities of interest. Many interaction databases already use text mining systems, and the development of these tools will lead to more reliable networks, as well as the possibility to personalize the networks by selecting the desired relations. This review will focus on different approaches of automatic information extraction from biomedical text that can be used to enhance existing networks or create new ones, such as deep learning state-of-the-art approaches, focusing on cancer disease as a case-study.

COVID-19 recommender system based on an annotated multilingual corpus.

Tracking the most recent advances in Coronavirus disease 2019 (COVID-19)-related research is essential, given the disease's novelty and its impact on society. However, with the publication pace speeding up, researchers and clinicians require automatic approaches to keep up with the incoming information regarding this disease. A solution to this problem requires the development of text mining pipelines; the efficiency of which strongly depends on the availability of curated corpora. However, there is a lack of COVID-19-related corpora, even more, if considering other languages besides English. This project's main contribution was the annotation of a multilingual parallel corpus and the generation of a recommendation dataset (EN-PT and EN-ES) regarding relevant entities, their relations, and recommendation, providing this resource to the community to improve the text mining research on COVID-19-related literature. This work was developed during the 7th Biomedical Linked Annotation Hackathon (BLAH7).

Information Retrieval Using Machine Learning for Biomarker Curation in the Exposome-Explorer.

Objective: In 2016, the International Agency for Research on Cancer, part of the World Health Organization, released the Exposome-Explorer, the first database dedicated to biomarkers of exposure for environmental risk factors for diseases. The database contents resulted from a manual literature search that yielded over 8,500 citations, but only a small fraction of these publications were used in the final database. Manually curating a database is time-consuming and requires domain expertise to gather relevant data scattered throughout millions of articles. This work proposes a supervised machine learning pipeline to assist the manual literature retrieval process. Methods: The manually retrieved corpus of scientific publications used in the Exposome-Explorer was used as training and testing sets for the machine learning models (classifiers). Several parameters and algorithms were evaluated to predict an article's relevance based on different datasets made of titles, abstracts and metadata. Results: The top performance classifier was built with the Logistic Regression algorithm using the title and abstract set, achieving an F2-score of 70.1%. Furthermore, we extracted 1,143 entities from these articles with a classifier trained for biomarker entity recognition. Of these, we manually validated 45 new candidate entries to the database. Conclusion: Our methodology reduced the number of articles to be manually screened by the database curators by nearly 90%, while only misclassifying 22.1% of the relevant articles. We expect that this methodology can also be applied to similar biomarkers datasets or be adapted to assist the manual curation process of similar chemical or disease databases.

Hybrid semantic recommender system for chemical compounds in large-scale datasets.

The large, and increasing, number of chemical compounds poses challenges to the exploration of such datasets. In this work, we propose the usage of recommender systems to identify compounds of interest to scientific researchers. Our approach consists of a hybrid recommender model suitable for implicit feedback datasets and focused on retrieving a ranked list according to the relevance of the items. The model integrates collaborative-filtering algorithms for implicit feedback (Alternating Least Squares and Bayesian Personalized Ranking) and a new content-based algorithm, using the semantic similarity between the chemical compounds in the ChEBI ontology. The algorithms were assessed on an implicit dataset of chemical compounds, CheRM-20, with more than 16.000 items (chemical compounds). The hybrid model was able to improve the results of the collaborative-filtering algorithms, by more than ten percentage points in most of the assessed evaluation metrics.

Basophil Activation Test Reduces Oral Food Challenges to Nuts and Sesame.

BACKGROUND: Nut allergic patients are often IgE sensitized to other nuts/seeds and need multiple oral food challenges (OFCs) before the safe nuts can be introduced in the diet. However, OFCs are time-consuming and risky procedures. OBJECTIVE: To assess the utility of the basophil activation test (BAT) to predict the allergic status and reduce the need for an OFC in children with 1 or more nut or seed allergies. METHODS: Participants in the Pronuts study recruited at the Geneva and the London centers were tested on the BAT to hazelnut, cashew nut, sesame, almond, and peanut, Ara h 1, Ara h 2, Ara h 6, using FlowCAST, a commercially available BAT kit, and flow cytometry. RESULTS: The BAT to hazelnut, cashew nut, sesame, almond, and peanut discriminated between allergic and nonallergic children, to the respective nut or seed. The optimal allergen concentration and their optimal, positive, and negative cutoffs were identified for the BAT and the other tests, for each nut and seed. Using the BAT as a second step in the diagnostic process, after equivocal skin prick test and IgE to extracts and components, reduced the number of total OFCs by 5% to 15% and positive OFCs by 33% to 75% (except for hazelnut) with 0% false-negatives and a diagnostic accuracy of 96% to 100%. CONCLUSION: The BAT proved to be a useful diagnostic tool, used in a stepwise approach, to predict the allergic status and reduce the number of OFCs in the Pronuts study participants with at least 1 nut allergy willing to consume selected nuts.

Using Neural Networks for Relation Extraction from Biomedical Literature.

Using different sources of information to support automated extracting of relations between biomedical concepts contributes to the development of our understanding of biological systems. The primary comprehensive source of these relations is biomedical literature. Several relation extraction approaches have been proposed to identify relations between concepts in biomedical literature, namely, using neural networks algorithms. The use of multichannel architectures composed of multiple data representations, as in deep neural networks, is leading to state-of-the-art results. The right combination of data representations can eventually lead us to even higher evaluation scores in relation extraction tasks. Thus, biomedical ontologies play a fundamental role by providing semantic and ancestry information about an entity. The incorporation of biomedical ontologies has already been proved to enhance previous state-of-the-art results.

A hybrid approach toward biomedical relation extraction training corpora: combining distant supervision with crowdsourcing.

Biomedical relation extraction (RE) datasets are vital in the construction of knowledge bases and to potentiate the discovery of new interactions. There are several ways to create biomedical RE datasets, some more reliable than others, such as resorting to domain expert annotations. However, the emerging use of crowdsourcing platforms, such as Amazon Mechanical Turk (MTurk), can potentially reduce the cost of RE dataset construction, even if the same level of quality cannot be guaranteed. There is a lack of power of the researcher to control who, how and in what context workers engage in crowdsourcing platforms. Hence, allying distant supervision with crowdsourcing can be a more reliable alternative. The crowdsourcing workers would be asked only to rectify or discard already existing annotations, which would make the process less dependent on their ability to interpret complex biomedical sentences. In this work, we use a previously created distantly supervised human phenotype-gene relations (PGR) dataset to perform crowdsourcing validation. We divided the original dataset into two annotation tasks: Task 1, 70% of the dataset annotated by one worker, and Task 2, 30% of the dataset annotated by seven workers. Also, for Task 2, we added an extra rater on-site and a domain expert to further assess the crowdsourcing validation quality. Here, we describe a detailed pipeline for RE crowdsourcing validation, creating a new release of the PGR dataset with partial domain expert revision, and assess the quality of the MTurk platform. We applied the new dataset to two state-of-the-art deep learning systems (BiOnt and BioBERT) and compared its performance with the original PGR dataset, as well as combinations between the two, achieving a 0.3494 increase in average F-measure. The code supporting our work and the new release of the PGR dataset is available at https://github.com/lasigeBioTM/PGR-crowd.

Robotic Versus Laparoscopic Approach to Hiatal Hernia Repair: Results After 7 Years of Robotic Experience.

INTRODUCTION: Robotic hiatal hernia repair offers potential advantages over traditional laparoscopy, most notably enhanced visualization, improved ergonomics, and articulating instruments. The clinical outcomes, however, have not been adequately evaluated. We report outcomes of laparoscopic and robotic hiatal hernia repairs. METHODS: A retrospective observational cohort study was performed of all hiatal hernia repairs performed from 2006 through 2019. Operative, demographic, and outcomes data were compared between laparoscopic and robotic groups. Discrete variables were analyzed with Chi-square of Fisher's exact test. Continuous variables were analyzed with Student's t test (mean) or Wilcoxon rank sum (medians). All analyses were performed using R statistical software. RESULTS: Laparoscopic repair was performed in 278 patients and robotic repair in 114. More recurrent hernias were repaired robotically (24.5% vs 12.9%, P = .08). Operative times were no different between groups (175 vs 179 minutes; P = .681). Robotic repair resulted in significantly shorter length of stay (LOS; 2.3 vs 3.3 days; P = .003). Rate of readmission was no different, and there were no differences in acute complications. For patients with at least 1 year of follow-up, recurrence rates were lower after robotic repair (13.3% vs 32.8%; P = .008); however, mean follow-up is significantly longer after laparoscopic repair (23.7 ± 28.4 vs 15.1 ± 14.9 months; P < .001). DISCUSSION: Robotic hiatal hernia repair offers technical advantages over laparoscopic repair with similar clinical outcomes.

Improving accessibility and distinction between negative results in biomedical relation extraction.

Accessible negative results are relevant for researchers and clinicians not only to limit their search space but also to prevent the costly re-exploration of research hypotheses. However, most biomedical relation extraction datasets do not seek to distinguish between a false and a negative relation among two biomedical entities. Furthermore, datasets created using distant supervision techniques also have some false negative relations that constitute undocumented/unknown relations (missing from a knowledge base). We propose to improve the distinction between these concepts, by revising a subset of the relations marked as false on the phenotype-gene relations corpus and give the first steps to automatically distinguish between the false (F), negative (N), and unknown (U) results. Our work resulted in a sample of 127 manually annotated FNU relations and a weighted-F1 of 0.5609 for their automatic distinction. This work was developed during the 6th Biomedical Linked Annotation Hackathon (BLAH6).

Biomarkers of severity and threshold of allergic reactions during oral peanut challenges.

BACKGROUND: Oral food challenge (OFC) is the criterion standard to assess peanut allergy (PA), but it involves a risk of allergic reactions of unpredictable severity. OBJECTIVE: Our aim was to identify biomarkers for risk of severe reactions or low dose threshold during OFC to peanut. METHODS: We assessed Learning Early about Peanut Allergy study, Persistance of Oral Tolerance to Peanut study, and Peanut Allergy Sensitization study participants by administering the basophil activation test (BAT) and the skin prick test (SPT) and measuring the levels of peanut-specific IgE, Arachis hypogaea 2-specific IgE, and peanut-specific IgG4, and we analyzed the utility of the different biomarkers in relation to PA status, severity, and threshold dose of allergic reactions to peanut during OFC. RESULTS: When a previously defined optimal cutoff was used, the BAT diagnosed PA with 98% specificity and 75% sensitivity. The BAT identified severe reactions with 97% specificity and 100% sensitivity. The SPT, level of Arachis hypogaea 2-specific IgE, level of peanut-specific IgE, and IgG4/IgE ratio also had 100% sensitivity but slightly lower specificity (92%, 93%, 90%, and 88%, respectively) to predict severity. Participants with lower thresholds of reactivity had higher basophil activation to peanut in vitro. The SPT and the BAT were the best individual predictors of threshold. Multivariate models were superior to individual biomarkers and were used to generate nomograms to calculate the probability of serious adverse events during OFC for individual patients. CONCLUSIONS: The BAT diagnosed PA with high specificity and identified severe reactors and low threshold with high specificity and high sensitivity. The BAT was the best biomarker for severity, surpassed only by the SPT in predicting threshold. Nomograms can help estimate the likelihood of severe reactions and reactions to a low dose of allergen in individual patients with PA.

Identification of biological mechanisms underlying a multidimensional ASD phenotype using machine learning.

The complex genetic architecture of Autism Spectrum Disorder (ASD) and its heterogeneous phenotype makes molecular diagnosis and patient prognosis challenging tasks. To establish more precise genotype-phenotype correlations in ASD, we developed a novel machine-learning integrative approach, which seeks to delineate associations between patients' clinical profiles and disrupted biological processes, inferred from their copy number variants (CNVs) that span brain genes. Clustering analysis of the relevant clinical measures from 2446 ASD cases in the Autism Genome Project identified two distinct phenotypic subgroups. Patients in these clusters differed significantly in ADOS-defined severity, adaptive behavior profiles, intellectual ability, and verbal status, the latter contributing the most for cluster stability and cohesion. Functional enrichment analysis of brain genes disrupted by CNVs in these ASD cases identified 15 statistically significant biological processes, including cell adhesion, neural development, cognition, and polyubiquitination, in line with previous ASD findings. A Naive Bayes classifier, generated to predict the ASD phenotypic clusters from disrupted biological processes, achieved predictions with a high precision (0.82) but low recall (0.39), for a subset of patients with higher biological Information Content scores. This study shows that milder and more severe clinical presentations can have distinct underlying biological mechanisms. It further highlights how machine-learning approaches can reduce clinical heterogeneity by using multidimensional clinical measures, and establishes genotype-phenotype correlations in ASD. However, predictions are strongly dependent on patient's information content. Findings are therefore a first step toward the translation of genetic information into clinically useful applications, and emphasize the need for larger datasets with very complete clinical and biological information.

Factors affecting salvage rate of infected prosthetic mesh.

BACKGROUND: Prosthetic mesh infection (PMI) is a challenging complication of ventral hernia repair (VHR). The sparsity of data leaves only experience and judgment to guide surgical decision-making. METHODS: Retrospective review of patients diagnosed with PMI. Subsequent abdominal operation (SAO) constitutes any intraabdominal operation occurring after the index hernia repair prior to PMI presentation. Any mesh removal was considered salvage failure. Analysis was performed using Chi-square test, Fishers Exact, or Mann-Whitney U test. Analyses completed using R Version 3.0.2. RESULTS: We identified 213 instances of PMI. Most cases (58.7%) involved intraperitoneal mesh. Thirty-seven percent of patients had an SAO, only 25.3% of which were clean cases. Enteroprosthetic fistula occurred in 38 patients (17.8%). Mean time to presentation was 19.9 mos after index hernia repair or SAO for infection alone, and 48.1 mos when a fistula was present (p < 0.001). Percutaneous drainage was used to treat 29 cases, successfully in 10 (34.5%), 8 of which were macroporous polypropylene and 2 biologic mesh. Negative pressure wound therapy (NPWT) was used in 46 patients, but successful in only 16 (34.8%), all of which were macroporous polypropylene. Local wound care alone successfully salvaged only 16 of 85 meshes (18.8%), 13 of which were macroporous polypropylene. Macroporous polypropylene mesh was salvaged in 65% of cases overall, and 72.2% when in an extraperitoneal position. Mesh salvage was not possible in any case involving composite or PTFE mesh, and rarely for microporous polypropylene (7.7%) multifilament polyester (4.2%), or intraperitoneal mesh (2.4%). Closure of the defect after mesh removal significantly lowers recurrence rate (p < 0.001). CONCLUSION: PMI involving composite, PTFE, multifilament polyester, or microporous polypropylene mesh requires explantation in nearly all cases. Infected macroporous polypropylene mesh in an extraperitoneal position is salvageable in most cases. Furthermore, the risk of secondary mesh infection after SAO, particularly with intraperitoneal mesh, should be considered during index VHR.

Linking chemical and disease entities to ontologies by integrating PageRank with extracted relations from literature.

BACKGROUND: Named Entity Linking systems are a powerful aid to the manual curation of digital libraries, which is getting increasingly costly and inefficient due to the information overload. Models based on the Personalized PageRank (PPR) algorithm are one of the state-of-the-art approaches, but these have low performance when the disambiguation graphs are sparse. FINDINGS: This work proposes a Named Entity Linking framework designated by Relation Extraction for Entity Linking (REEL) that uses automatically extracted relations to overcome this limitation. Our method builds a disambiguation graph, where the nodes are the ontology candidates for the entities and the edges are added according to the relations established in the text, which the method extracts automatically. The PPR algorithm and the information content of each ontology are then applied to choose the candidate for each entity that maximises the coherence of the disambiguation graph. We evaluated the method on three gold standards: the subset of the CRAFT corpus with ChEBI annotations (CRAFT-ChEBI), the subset of the BC5CDR corpus with disease annotations from the MEDIC vocabulary (BC5CDR-Diseases) and the subset with chemical annotations from the CTD-Chemical vocabulary (BC5CDR-Chemicals). The F1-Score achieved by REEL was 85.8%, 80.9% and 90.3% in these gold standards, respectively, outperforming baseline approaches. CONCLUSIONS: We demonstrated that RE tools can improve Named Entity Linking by capturing semantic information expressed in text missing in Knowledge Bases and use it to improve the disambiguation graph of Named Entity Linking models. REEL can be adapted to any text mining pipeline and potentially to any domain, as long as there is an ontology or other knowledge Base available.