A Scoping Review of Artificial Intelligence Applications for Diagnostic Accuracy in Kazakhstan's Healthcare System

A

BSTRACT

There are different challenges that persist with the use of traditional diagnostic methods in Kazakhstan. With the rise of new technology such as artificial intelligence (AI), medical diagnosis can become more accurate and efficient, especially in complicated cases. This study reviewed 11 publications which were published between the years 2020 and 2025. They examined the use of AI in medical diagnostics in Kazakhstan. Sources where these publications were gotten from include PubMed, Google Scholar and Semantic Scholar. Afterwards, the data were synthesized using the World Health Organization digital health framework. The accuracy of diagnosis, especially with the use of AI technologies, was between the ranges of 50% and 96%. As a descriptive summary statistic from our narrative synthesis, it had a median accuracy of diagnosis of about 91.0%. The sensitivity of breast cancer became better by 20% with the use of a convolutional neural network. The use of deep learning was similar to using ophthalmologists in detecting retinal pathology. Detection of stroke improved with the use of AI-enhanced computed tomography imaging in hospitals in urban areas. There are still some variations with the adoption of AI, especially in rural areas. This is majorly because of poor infrastructure and unskilled workers. Even with the high potential of AI to improve medical diagnosis accuracy, its use in Kazakhstan remains limited. This assessment demands for an all-inclusive national strategy for AI implementation in medical diagnostics. These contain effective administration initiatives in ignored regional zones, implementation of an unchanging data exchange arrangement and the expansion of instruction curricula for healthcare experts on the usage of AI.