International Journal of
Medicine and Medical Sciences

  • Abbreviation: Int. J. Med. Med. Sci.
  • Language: English
  • ISSN: 2006-9723
  • DOI: 10.5897/IJMMS
  • Start Year: 2009
  • Published Articles: 535

Full Length Research Paper

Ultrasound image acquisition by a personal computer- Application of artificial neural network

Mitra Ahmad Soltani-MD, MS
  • Mitra Ahmad Soltani-MD, MS
  • Lutalo Medical Research Corporation 15- Shariaty all- Kord St- Ressalat HW- Tehran –Iran
  • Google Scholar


  •  Accepted: 20 January 2012
  •  Published: 26 January 2012

Abstract

 

Traditional  probes  consist  of  40  to  60  crystals  each attached  to a pin attached  to a specialized cable of minimum 2  to 3 m. Each crystal has piezoelectric properties.   To capture  images by personal computers to  improve  accessibility  and  reduce  the  cost  of  having  ultrasonic  image  with  special  reference  to obstetrics  and  gynecology  emergency  settings,  a probe was designed consisting of three Doppler transducers  (each  with  4  pins  hence  generating  4  signals,  altogether  12  signal)  that  by  an  analog switching with micro (1000 HZ per second) can generate 12000 signals per second changing the scan line  form  linear  into  a  plane.    The  signals  are  translated  into  WAV  sound  format  file  that  can  be displayed by a Windows-based program of a personal computer. The pattern produced is created by the sound of blood flow in an organ. This vascular pattern was matched with traditional sonography of the organ. By training the network, the resolution of  images can be improved further based on the formula: 

output = 0.77*target + 38.   The probe can capture  images 1/6  resolution of a  traditional probe, deeper penetration  (19  cm  depth),  1/34  price  and  weight  of  a  traditional  ultrasound  equipment.

 

Key words: Doppler transducers, scan line, artificial neural network, ultrasound, obstetrics and gynecology