Ultrasound Entropy Imaging for Evaluating the Stage of Fatty Liver Disease: Fundamental Studies, Clinical Validations, and Techniques Implementation

Project: National Science and Technology CouncilNational Science and Technology Council Academic Grants

Project Details

Abstract

Principal investigator has made efforts on the developments of ultrasound parametric imaging based on statistical distributions of backscattering for many years. Compared with the conventional statistical models, information-theory entropy emerged recently can be used to describe the behavior of backscattering and have some strengths, including (i) entropy is not a model-based parameter. For this reason, data used for entropy estimation do not need to follow a specific distribution, providing flexibility in applications, (ii) entropy estimation is performed from the analysis of radiofrequency data (demodulation for envelope detection is not necessary), making it better to reflect information associated with the signal waveform, (iii) parametric imaging can be implemented in ultrasound systems that cannot output raw data. This 3-years research project is aimed to develop a functional ultrasound parametric imaging based on information theory. During the first year, the foundation of the study will be established, during which the scattering and physical meanings of Shannon entropy in information theory will be examined, and entropy imaging combined with some advanced signal/image processing algorithms will be developed. During the second year, clinical verification will be conducted, in which fatty livers will be used in clinical performance verifications of Shannon entropy imaging. The verification results will be used for developing the foundation of a fatty liver radar-chart diagnostic model. During the third year, the results and experiences of the previous 2 years will be expanded and further investigated. Real-time entropy imaging system and a computer-aided diagnosis software will be built in clinical ultrasound systems. The results obtained from this study may fulfill the purposes of fostering research talent, cultivating novel techniques, and adding value to Taiwan’s domestic ultrasound products. In conclusion, the results will provide the domestic ultrasound industry with a compact and competitive value-added solution.

Project IDs

Project ID:PB10501-2693
External Project ID:MOST103-2221-E182-001-MY3
StatusFinished
Effective start/end date01/08/1631/07/17

Keywords

  • Information theory
  • Shannon entropy
  • fatty liver
  • computer-aided diagnosis

Fingerprint

Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.