Your browser doesn't support javascript.
loading
A 178-clinical-center experiment of integrating AI solutions for lung pathology diagnosis.
Ibragimov, Bulat; Arzamasov, Kirill; Maksudov, Bulat; Kiselev, Semen; Mongolin, Alexander; Mustafaev, Tamerlan; Ibragimova, Dilyara; Evteeva, Ksenia; Andreychenko, Anna; Morozov, Sergey.
Afiliação
  • Ibragimov B; Department of Computer Science, University of Copenhagen, Copenhagen, Denmark. bulat@di.ku.dk.
  • Arzamasov K; Research and Practical Clinical Center for Diagnostics and Telemedicine Technologies of the Moscow Healthcare Department, Moscow, Russia.
  • Maksudov B; School of Electronic Engineering, Dublin City University, Dublin, Ireland.
  • Kiselev S; Innopolis University, Innopolis, Russia.
  • Mongolin A; Innopolis University, Innopolis, Russia.
  • Mustafaev T; Nova Information Management School, Universidade Nova de Lisboa, Lisbon, Portugal.
  • Ibragimova D; Innopolis University, Innopolis, Russia.
  • Evteeva K; University Clinic Kazan State University, Kazan, Russia.
  • Andreychenko A; University Clinic Kazan State University, Kazan, Russia.
  • Morozov S; Research and Practical Clinical Center for Diagnostics and Telemedicine Technologies of the Moscow Healthcare Department, Moscow, Russia.
Sci Rep ; 13(1): 1135, 2023 01 20.
Article em En | MEDLINE | ID: mdl-36670118
ABSTRACT
In 2020, an experiment testing AI solutions for lung X-ray analysis on a multi-hospital network was conducted. The multi-hospital network linked 178 Moscow state healthcare centers, where all chest X-rays from the network were redirected to a research facility, analyzed with AI, and returned to the centers. The experiment was formulated as a public competition with monetary awards for participating industrial and research teams. The task was to perform the binary detection of abnormalities from chest X-rays. For the objective real-life evaluation, no training X-rays were provided to the participants. This paper presents one of the top-performing AI frameworks from this experiment. First, the framework used two EfficientNets, histograms of gradients, Haar feature ensembles, and local binary patterns to recognize whether an input image represents an acceptable lung X-ray sample, meaning the X-ray is not grayscale inverted, is a frontal chest X-ray, and completely captures both lung fields. Second, the framework extracted the region with lung fields and then passed them to a multi-head DenseNet, where the heads recognized the patient's gender, age and the potential presence of abnormalities, and generated the heatmap with the abnormality regions highlighted. During one month of the experiment from 11.23.2020 to 12.25.2020, 17,888 cases have been analyzed by the framework with 11,902 cases having radiological reports with the reference diagnoses that were unequivocally parsed by the experiment organizers. The performance measured in terms of the area under receiving operator curve (AUC) was 0.77. The AUC for individual diseases ranged from 0.55 for herniation to 0.90 for pneumothorax.
Assuntos

Texto completo: 1 Base de dados: MEDLINE Assunto principal: Pneumotórax / Radiografia Torácica Tipo de estudo: Diagnostic_studies Limite: Humans Idioma: En Ano de publicação: 2023 Tipo de documento: Article

Texto completo: 1 Base de dados: MEDLINE Assunto principal: Pneumotórax / Radiografia Torácica Tipo de estudo: Diagnostic_studies Limite: Humans Idioma: En Ano de publicação: 2023 Tipo de documento: Article