Dataset Information

Computer-assisted initial diagnosis of rare diseases.

ABSTRACT: Introduction. Most documented rare diseases have genetic origin. Because of their low individual frequency, an initial diagnosis based on phenotypic symptoms is not always easy, as practitioners might never have been exposed to patients suffering from the relevant disease. It is thus important to develop tools that facilitate symptom-based initial diagnosis of rare diseases by clinicians. In this work we aimed at developing a computational approach to aid in that initial diagnosis. We also aimed at implementing this approach in a user friendly web prototype. We call this tool Rare Disease Discovery. Finally, we also aimed at testing the performance of the prototype. Methods. Rare Disease Discovery uses the publicly available ORPHANET data set of association between rare diseases and their symptoms to automatically predict the most likely rare diseases based on a patient's symptoms. We apply the method to retrospectively diagnose a cohort of 187 rare disease patients with confirmed diagnosis. Subsequently we test the precision, sensitivity, and global performance of the system under different scenarios by running large scale Monte Carlo simulations. All settings account for situations where absent and/or unrelated symptoms are considered in the diagnosis. Results. We find that this expert system has high diagnostic precision (?80%) and sensitivity (?99%), and is robust to both absent and unrelated symptoms. Discussion. The Rare Disease Discovery prediction engine appears to provide a fast and robust method for initial assisted differential diagnosis of rare diseases. We coupled this engine with a user-friendly web interface and it can be freely accessed at http://disease-discovery.udl.cat/. The code and most current database for the whole project can be downloaded from https://github.com/Wrrzag/DiseaseDiscovery/tree/no_classifiers.

SUBMITTER: Alves R

PROVIDER: S-EPMC4963223 | biostudies-literature | 2016

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Computer-assisted initial diagnosis of rare diseases.

Alves Rui R Piñol Marc M Vilaplana Jordi J Teixidó Ivan I Cruz Joaquim J Comas Jorge J Vilaprinyo Ester E Sorribas Albert A Solsona Francesc F

PeerJ 20160721

Introduction. Most documented rare diseases have genetic origin. Because of their low individual frequency, an initial diagnosis based on phenotypic symptoms is not always easy, as practitioners might never have been exposed to patients suffering from the relevant disease. It is thus important to develop tools that facilitate symptom-based initial diagnosis of rare diseases by clinicians. In this work we aimed at developing a computational approach to aid in that initial diagnosis. We also aimed ...[more]

PMID: 27547534

Similar Datasets

Project description:BackgroundFor a number of persons with rare diseases (RDs) a definite diagnosis remains undiscovered with relevant physical, psychological and social consequences. Undiagnosed RDs (URDs) require other than specialised clinical centres, outstanding molecular investigations, common protocols and dedicated actions at national and international levels; thus, many "Undiagnosed RDs programs" have been gradually developed on the grounds of a well-structured multidisciplinary approach.MethodsThe Italian Undiagnosed Rare Diseases Network (IURDN) was established in 2016 to improve the level of diagnosis of persons with URD living in Italy. Six Italian Centres of Expertise represented the network. The National Centre for Rare Diseases at the Istituto Superiore di Sanità coordinates the whole project. The software PhenoTips was used to collect the information of the clinical cases.ResultsOne hundred and ten cases were analysed between March 2016 and June 2019. The age of onset of the diseases ranged from prenatal age to 51 years. Conditions were predominantly sporadic; almost all patients had multiple organs involvements. A total of 13/71 family cases were characterized by WES; in some families more than one individual was affected, so leading to 20/71 individuals investigated. Disease causing variants were identified in two cases and were associated to previously undescribed phenotypes. In 5 cases, new candidate genes were identified, although confirmatory tests are pending. In three families, investigations were not completed due to the scarce compliance of members and molecular investigations were temporary suspended. Finally, three cases (one familial) remain still unsolved. Twelve undiagnosed clinical cases were then selected to be shared at International level through PhenomeCentral in accordance to the UDNI statement.ConclusionsOur results showed a molecular diagnostic yield of 53,8%; this value is comparable to the diagnostic rates reported in other international studies. Cases collected were also pooled with those collected by UDNI International Network. This represents a unique example of global initiative aimed at sharing and validating knowledge and experience in this field. IURDN is a multidisciplinary and useful initiative linking National and International efforts aimed at making timely and appropriate diagnoses in RD patients who still do not have a confirmed diagnosis even after a long time.

Project description:BackgroundRare diseases may be defined as occurring in less than 1 in 2000 patients. Such conditions are, however, so numerous that up to 5.9% of the population is afflicted by a rare disease. The gambling industry attests that few people have native skill evaluating probabilities. We believe that both students and academics, under-estimate the likelihood of encountering rare diseases. This combines with pressure on curriculum time, to reduce both student interest in studying rare diseases, and academic content preparing students for clinical practice. Underestimation of rare diseases, may also contribute to unhelpful blindness to considering such conditions in the clinic.MethodsWe first developed a computer simulation, modelling the number of cases of increasingly rare conditions encountered by a cohort of clinicians. The simulation captured results for each year of practice, and for each clinician throughout the entirety of their careers. Four hundred sixty-two theoretical conditions were considered, with prevalence ranging from 1 per million people through to 64.1% of the population. We then delivered a class with two in-class on-line surveys evaluating student perception of the importance of learning about rare diseases, one before and the other after an in-class real-time computer simulation. Key simulation variables were drawn from the student group, to help students project themselves into the simulation.ResultsThe in-class computer simulation revealed that all graduating clinicians from that class would frequently encounter rare conditions. Comparison of results of the in-class survey conducted before and after the computer simulation, revealed a significant increase in the perceived importance of learning about rare diseases (p < 0.005).ConclusionsThe computer career simulation appeared to affect student perception. Because the computer simulation demonstrated clinicians frequently encounter patients with rare diseases, we further suggest this should be considered by academics during curriculum review and design.

Dataset Information

Computer-assisted initial diagnosis of rare diseases.

Publications

Computer-assisted initial diagnosis of rare diseases.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets