Description du projet
Une application de diagnostic sur smartphone pour les professionnels médicaux
Les erreurs médicales constituent un problème mondial, responsable de milliers de cas de handicaps permanents et de décès chaque année. Le projet Kahun, financé par l’UE, finance la société Kahun Medical (Israël) qui développe une application pour le web et pour smartphone dédiée au diagnostic différentiel (DD) et destinée aux médecins et aux étudiants en médecine. L’application, dont la base de données de définitions, de références et de concepts médicaux est régulièrement mise à jour à partir de la littérature médicale, permet aux utilisateurs de saisir des symptômes et des résultats de tests en laboratoire, en leur présentant un DD ainsi que des voies possibles, allant de simples variables jusqu’au diagnostic potentiel, en passant par une série de suggestions pour la prise en charge du patient. En outre, ses algorithmes bayésiens innovants identifient la probabilité de relations croisées entre les éléments de la base de données. Cet outil novateur est susceptible de présenter un intérêt majeur par les temps qui courent, dans un contexte où l’endiguement de la pandémie de COVID-19 représente une priorité absolue.
Objectif
Medical errors are recognized as a world-wide problem - this includes mistakes in performing Differential Diagnosis resulting in sending patients to unnecessary tests, missing vital tests, and additional mistakes along the patient management process. In E.U countries alone, medical errors amount to 750,000 Million errors a year, resulting 3.2 million more hospitalizations, 260,000 permanent disabilities and 95,000 deaths each year.
The product is a smartphone application (and a web-based version), designed for doctors and medical students. Inspired by crowd-based platforms, its database is regularly updated using a dedicated editor for inputting medical definitions, concepts, and references from the professional medical literature, and for annotating the knowledge and defining relations between elements. So far, Kahun has already established an exclusive database for Internal Medicine's ontology, comprising of more than 50 thousand elements and relations.
Kahun's innovative Bayesian algorithms, identify probabilistic relations between the elements in the database - within the tens of thousands of elements scattered across many layers of depth, it creates a massive knowledge graph of probabilistic relationships within the medical ontology. By now, Kahun generates more than 3 Million insights for relations between medical concepts.
Kahun allows users to enter variables such as symptoms and lab test results, and presents them a Differential Diagnosis, along with a clear Path to Evidence – track leading from variables to possible result, presented in a graphical correlations map between all elements and with specific references from the literature, and it provides a series of suggestions for patient management such as which lab tests or imaging are the best next step.
The company plans to expand the knowledge graph into additional medical fields and provide even greater value. The company expects to reach total revenues of €207 Million by the end of 2026.
Champ scientifique
- natural sciencescomputer and information sciencesknowledge engineeringontology
- natural sciencescomputer and information sciencesdatabases
- medical and health sciencesclinical medicineinternal medicine
- medical and health scienceshealth sciencesinfectious diseasesRNA virusescoronaviruses
- engineering and technologyelectrical engineering, electronic engineering, information engineeringinformation engineeringtelecommunicationsmobile phones
Mots‑clés
Programme(s)
Régime de financement
SME-2 - SME instrument phase 2Coordinateur
6744320 Tel Aviv
Israël
L’entreprise s’est définie comme une PME (petite et moyenne entreprise) au moment de la signature de la convention de subvention.