[The value of artificial and human intelligence - the example of bone scintigraphy].

Journal: Magyar onkologia
Published Date:

Abstract

We present a possible method of Artificial Intelligence (AI) based applications that can effectively filter noise-sensitive bone scintigraphy images. The use of special AI, based on preliminary examinations, allows us to significantly reduce study time or activity administered to the patient, thus reducing the patient, assistant, and physician radiation. We present the features of the AI filtering application, its teaching process, which is important to understand, so that the physician can safely take the processed image of the AI as a "secondary reliable opinion" to help them make a more accurate diagnosis. We also examine the robustness of the algorithm, the specificities and challenges of complex clinical control.

Authors

  • Ákos Kovács
    Mediso Medical Imaging Systems Kft., Budapest, Hungary. tamas.bukki@mediso.com.
  • Gábor Légrádi
    Mediso Medical Imaging Systems Kft., Budapest, Hungary. tamas.bukki@mediso.com.
  • András Wirth
    Mediso Medical Imaging Systems Kft., Budapest, Hungary. tamas.bukki@mediso.com.
  • Ferenc Nagy
    Institute of Plant Biology, Biological Research Centre of the Hungarian Academy of Sciences, H-6726 Szeged, Hungary.
  • Attila Forgács
    ScanoMed, Orvosi, Diagnosztikai, Oktató és Kutató Kft., Debrecen, Hungary.
  • Sándor Barna
    ScanoMed, Orvosi, Diagnosztikai, Oktató és Kutató Kft., Debrecen, Hungary.
  • Ildikó Garai
    ScanoMed, Orvosi, Diagnosztikai, Oktató és Kutató Kft., Debrecen, Hungary.
  • Tamás Bükki
    Mediso Medical Imaging Systems Kft., Budapest, Hungary. tamas.bukki@mediso.com.