Your browser doesn't support javascript.
loading
Comparison of clinical geneticist and computer visual attention in assessing genetic conditions.
Duong, Dat; Johny, Anna Rose; Ledgister Hanchard, Suzanna; Fortney, Christopher; Flaharty, Kendall; Hellmann, Fabio; Hu, Ping; Javanmardi, Behnam; Moosa, Shahida; Patel, Tanviben; Persky, Susan; Sümer, Ömer; Tekendo-Ngongang, Cedrik; Lesmann, Hellen; Hsieh, Tzung-Chien; Waikel, Rebekah L; André, Elisabeth; Krawitz, Peter; Solomon, Benjamin D.
Afiliación
  • Duong D; Medical Genetics Branch, National Human Genome Research Institute, Bethesda, Maryland, United States of America.
  • Johny AR; Institute for Genomic Statistics and Bioinformatics, University Hospital Bonn, Rheinische Friedrich-Wilhelms-Universität Bonn, Bonn, Germany.
  • Ledgister Hanchard S; Medical Genetics Branch, National Human Genome Research Institute, Bethesda, Maryland, United States of America.
  • Fortney C; Social and Behavioral Research Branch, National Human Genome Research Institute, Bethesda, Maryland, United States of America.
  • Flaharty K; Medical Genetics Branch, National Human Genome Research Institute, Bethesda, Maryland, United States of America.
  • Hellmann F; Chair for Human-Centered Artificial Intelligence, University of Augsburg, Augsburg, Germany.
  • Hu P; Medical Genetics Branch, National Human Genome Research Institute, Bethesda, Maryland, United States of America.
  • Javanmardi B; Institute for Genomic Statistics and Bioinformatics, University Hospital Bonn, Rheinische Friedrich-Wilhelms-Universität Bonn, Bonn, Germany.
  • Moosa S; Division of Molecular Biology and Human Genetics, Stellenbosch University, Stellenbosch, South Africa.
  • Patel T; Department of Medical Genetics, Tygerberg Hospital, Tygerberg, South Africa.
  • Persky S; Medical Genetics Branch, National Human Genome Research Institute, Bethesda, Maryland, United States of America.
  • Sümer Ö; Social and Behavioral Research Branch, National Human Genome Research Institute, Bethesda, Maryland, United States of America.
  • Tekendo-Ngongang C; Chair for Human-Centered Artificial Intelligence, University of Augsburg, Augsburg, Germany.
  • Lesmann H; Medical Genetics Branch, National Human Genome Research Institute, Bethesda, Maryland, United States of America.
  • Hsieh TC; Institute for Genomic Statistics and Bioinformatics, University Hospital Bonn, Rheinische Friedrich-Wilhelms-Universität Bonn, Bonn, Germany.
  • Waikel RL; Institute for Genomic Statistics and Bioinformatics, University Hospital Bonn, Rheinische Friedrich-Wilhelms-Universität Bonn, Bonn, Germany.
  • André E; Medical Genetics Branch, National Human Genome Research Institute, Bethesda, Maryland, United States of America.
  • Krawitz P; Chair for Human-Centered Artificial Intelligence, University of Augsburg, Augsburg, Germany.
  • Solomon BD; Institute for Genomic Statistics and Bioinformatics, University Hospital Bonn, Rheinische Friedrich-Wilhelms-Universität Bonn, Bonn, Germany.
PLoS Genet ; 20(2): e1011168, 2024 Feb.
Article en En | MEDLINE | ID: mdl-38412177
ABSTRACT
Artificial intelligence (AI) for facial diagnostics is increasingly used in the genetics clinic to evaluate patients with potential genetic conditions. Current approaches focus on one type of AI called Deep Learning (DL). While DL- based facial diagnostic platforms have a high accuracy rate for many conditions, less is understood about how this technology assesses and classifies (categorizes) images, and how this compares to humans. To compare human and computer attention, we performed eye-tracking analyses of geneticist clinicians (n = 22) and non-clinicians (n = 22) who viewed images of people with 10 different genetic conditions, as well as images of unaffected individuals. We calculated the Intersection-over-Union (IoU) and Kullback-Leibler divergence (KL) to compare the visual attentions of the two participant groups, and then the clinician group against the saliency maps of our deep learning classifier. We found that human visual attention differs greatly from DL model's saliency results. Averaging over all the test images, IoU and KL metric for the successful (accurate) clinician visual attentions versus the saliency maps were 0.15 and 11.15, respectively. Individuals also tend to have a specific pattern of image inspection, and clinicians demonstrate different visual attention patterns than non-clinicians (IoU and KL of clinicians versus non-clinicians were 0.47 and 2.73, respectively). This study shows that humans (at different levels of expertise) and a computer vision model examine images differently. Understanding these differences can improve the design and use of AI tools, and lead to more meaningful interactions between clinicians and AI technologies.
Asunto(s)
Texto completo
Añadir a Mi BVS
Imprimir
XML
PubMed Links
Buscar en Google

Texto completo: 1 Colección: 01-internacional Base de datos: MEDLINE Asunto principal: Computadores / Inteligencia Artificial Límite: Humans Idioma: En Revista: PLoS Genet Asunto de la revista: GENETICA Año: 2024 Tipo del documento: Article País de afiliación: Estados Unidos Pais de publicación: Estados Unidos
Texto completo
Añadir a Mi BVS
Imprimir
XML
PubMed Links
Buscar en Google

Texto completo: 1 Colección: 01-internacional Base de datos: MEDLINE Asunto principal: Computadores / Inteligencia Artificial Límite: Humans Idioma: En Revista: PLoS Genet Asunto de la revista: GENETICA Año: 2024 Tipo del documento: Article País de afiliación: Estados Unidos Pais de publicación: Estados Unidos