Researchers from UdL and IRBLleida build a web tool for the assisted diagnosis of rare diseases

| Published by Universitat de Lleida / IRBLleida

'Rare Disease Discovery’ is a prototype that automatically ranks the rare diseases that are most likely associated to a list of patient symptoms. The prototype shows a precision of moret han 80% on the benchmark tests

Researchers from UdL and IRBLleida build a web tool for the assisted diagnosis of rare diseases

Researchers from Universitat de Lleida (UdL) and Institut de Recerca Biomèdica de Lleida Fundació Dr. Pifarré (IRBLleida) developed the prototype for a free web tool that assists the diagnosis of more than 4.000 rare diseases, such as Beta thalassemia; Turner syndrome, or Canavan's disease, among others.

Rare Disease Discovery was built using publicly available datasets that associate rare diseases to their known symptoms. The tool uses the list of symptoms from the patient to provide the medical professionals with a list of rare diseases that can be associated to that patient, ranked from the most to the least likely disease.

The tool was benchmarked retrospectively in a group of 187 patients that had a confirmed rare disease diagnostic, showing a precision of around 80%, as described in the paper that was recently published in Peer J. Further benchmarking with larger and more diverse sets of patients is the logical next step to confirm the utility of the tool.

Most known rare diseases have genetic origins. They are hard to diagnose by family doctors due to their low frequency and a final diagnosis almost always requires genetic testing. Because of this it is important to provide family doctors with tools that assist in the initial diagnosis of those rare diseases. Developing such a tool was the goal of researchers from the Departments of Ciències Mèdiques Bàsiques i Informàtica i Enginyeria Industrial from UdL and IRBLleida. This free web prototype can be accessed at http://disease-discovery.udl.cat/

The tool was developed by Rui Alves, Joaquim Cruz, Ester Vilaprinyó i Albert Sorribas (Department of Ciències Mèdiques Bàsiques, UdL and IRBLleida), Jorge Comas (UdL-Institut de Tecnologia Química i Biològica António Xavier, Portugal), Marc Piñol (Department of Informàtica i Enginyeria Industrial, UdL), Francesc Solsona, Jordi Vilaplana i Ivan Teixidó (Department of Informàtica i Enginyeria Industrial, UdL and INSPIRES).

 

Reference article: Alves et al. (2016), Computer-assisted initial diagnosis of rare diseases. Peer J 4:e2211; DOI10.7717/peerj.2211

Related Document: Download

Rare Disease Discovery