Loading...

Tracking of Human Sperm Cell using a Dynamic Bayesian Network Based Framework

Arasteh, Abdollah | 2018

665 Viewed
  1. Type of Document: Ph.D. Dissertation
  2. Language: Farsi
  3. Document No: 51320 (05)
  4. University: Sharif University of Technology
  5. Department: Electrical Engineering
  6. Advisor(s): Vosoughi Vahdat, Bijan; Salman Yazdi, Reza
  7. Abstract:
  8. Infertility is an important problem to deal in medicine. In every four couples, on average, one couple is affected by infertility in developing countries. In the majority of cases, the infertility of men has a relationship with spermatozoa and semen, and can be measured by semen and spermatozoa analysis for more advanced diagnosis and treatments. Analysis of the movement patterns of spermatozoa were performed by expert screeners earlier, but nowadays, many of these analyzes are performed using computer-based systems called computer assisted sperm analysis (CASA). The benefits of using CASA instead of expert screeners are achieving system-independency and numerical results at the end of the analysis. However, CASA systems have issues in image processing and automatic tracking of spermatozoa, e.g. illumination and recording conitions and occulusion of cells by each other. Multi-target tracking in an image sequence is a widespread problem for which many solutions have been provided. Recently, Dynamic Bayesian Networks based frameworks are used and discussed for tracking of the objects in an image sequence, which have showed to be a robust and precise method and gives better insight of the moving targets model. One of the methods used in the current research was based on Hybrid Dynamic Bayesian Network (HDBN) which showed better performance in comparison with the other well-known methods in this field. With achieving an F1 measure of 77%, it performed better than NNF (62%) and MHT (54%) in tracking cells. The next implemented method was a heuristic method based on Markov chain Monte Carlo (MCMC) and Metropolis-Hastings sampling, which succeeded to show robustness against cell miss-detections and false alarms. In this heuristic method when the probability of detection was 90% and the probability of false alarm was 50%, F1 measure achieved was 80% which was better than NNF (20%), MHT (65%), and LKF (45%)
  9. Keywords:
  10. Infertility ; Dynamic Bayesian Network ; Cells Tracking ; Computer Aided Sperm Analysis

 Digital Object List

 Bookmark

No TOC