8. Áp dụng công cụ javacript để phát triển giải pháp tự động hoá phân tích kết quả siêu âm doppler xuyên sọ

Abstract

The objective of this study was to develop an automated process for interpreting the results of transcranial doppler ultrasound. Javascript programming language was used to create functions for interpreting flow velocity in the arteries of the brain. In the trial period, 100 subjects were randomly assigned to undergo ultrasound and have their arterial flow velocity calculated using manual calculation or automated Javacript solution. For the application period, the Javascript solution was applied to ultrasound results of 43,134 subjects. Data is normalized and encoded as numbers and strings. The application interface is a digitized cerebral blood flow velocity measurement result sheet in the form of an HTML file. During the trial period, the reporting speed of the Javascript solution group was significantly faster (23.6 ± 3.1s) than the manual group (605.7 ± 6.2s) (p < 0.001). The overall accuracy of automated responses (100%) was significantly higher than manual responses (75%) (p < 0.05). The duration of automated results interpretation is shortened 25.6 times which saves 6,973 working hours, while also achieves 100% overall accuracy and stores 9.7 million data points in comparison to the manual process. This study has developed a viable automated solution to support the process of interpreting result in quick, reliable and efficient way and also creating a standard data set of cerebral blood flow velocity.