Artificial Intelligence-Based Ultrasound Imaging Classification for Infant Neurological Impairment Disorders: A Review

Artificial intelligence (AI) is now conventionally used in medicine for supporting decision making processes in diagnostic procedures. Ultrasound, as one of the main diagnostic tools in medicine has been a subject to a substantial amount of research in the field of automatization of diagnostics made by ultrasound. As one of the most widespread and virtually irreplaceable mode of monitoring fetal maturation and neural development during the prenatal stage, ultrasound imaging has to be as reproducible as possible and become devoid of interoperator variability. As a means of achieving that, artificial intelligence can be employed to evaluate ultrasound recordings in real time. KANET test was developed and is accepted as a golden standard for fetal neurodevelopmental disorder risk assessment. Even though the test itself is based on a very strict scale, and gynecologists can be trained to perform it with extensive confidence, its automation with AI would bring upon to increase in its presence in diagnostics.