No ise in ultra so nic im a g ing B. BREYER, T. VICULIN and B. VOJNOVIC The feasibility of noise measurement in medical ultrasonic imaging has been studied and a theoretical evaluation of stochastic noise influence upon the perceived image has been undertaken. The influence of noise on a perceived image has been calculated by considering the light intensity perception characteristic of the human eye and it has been found that the additive noise in the video signal pushes the small signals in the image towards the invisible (black) region. Impulse noise measurements of amplitude and time jitter have been set up using a standard water/Ccl, interface and pulse amplitude analysis to take into account the random nature of noise by distribution measurements. Introduction Noise in grey-scale images Noise is the ultimate limitation of any measurement. Medical ultrasonic imaging measurements consist of time measure- ments for depth assessment, transducer angle and position measurements for echo position assessment, and amplitude measurements for reflector quality evaluation and grey-scale visualization. All these measurements can include random errors, and we have set up methods for amplitude and echo arrival time distribution measurements to obtain quantitative data on system performance. Knowledge of complete distribu- tion, and in certain cases of mean and standard deviation, gives quantitative figures for describing the random noise. Furthermore the system can in this way be specified. The methods were adapted from nuclear electronics but a stand- ard and reproducible reflection interface had to be included. The water/Ccl4 interface is easily reproduced and in principle easy to make. However, practical considerations, cleanliness and temperature stability problems make it less acceptable as a standard, so for further measurements special plastic materials have to be chosen.’ In grey-scale echography one attempts to include as many sound intensities as possible in the image. Since the useful range of echoes can be of the order 50-60 dB,2-4 compres- sion characteristics are used to fit this span into the dyna- mics of available monitors.435 This is not however the final grey-scale characteristic perceived since there is the eye-brain light intensity transfer function that modifies and compresses the displayed image intensitites in a physiological way.6-g The general, average, form of the lightness perception charac- teristic is quasi-logarithmic, starting at ‘black’ for zero light intensities and going up to some non-damaging ‘extremely bright’ perceived lightness. One measures such characteristics by setting up a statistically significant poll of observers, since the perceived lightness is a subjective feeling. The question of the influence of random noise on a grey-scale image has been studied by including an essential part of the system - the eye perception characteristics. Since data on the perception of light intensity are purely empirical and average, numerical rather than analytical calculation was undertaken to evaluate the quality of noise influence upon the perceived image. In principle low intensity parts of the image will be pushed into the darker region by noise disturb- ances that diminish that particular signal and will be made brighter when the noise makes the signal larger. Since the perception characteristic is not linear, this will result in a shift of the perceived mean value. The effect will be different in cases when maximum pulse signal is being stored, that is, all signals will on average, after a sufficient number of over- writings, be pushed towards the brighter region. I I I I 40 60 00 100 Relative lightwss, X B. Brayer, is at the Medical Physics Department, Gynaecological Cancer Center, Petrova 13.41000 Zagreb, Yugoslavia; T. Viculin is at the Institute for Tumours and Allied Diseases, llica 197, 41000 Zagreb, Yugoslavia; B. VojnoviC is at the Institute ‘Ruder Bos’koviC’, Bijenidka 54,410OO Zagreb, Yugoslavia. Paper received 22 June 1979. Revised 5 November 1979. / Noise disturb line spectrum Fig. 1 Relative light intensity perception characteristic. Noise dis- tribution at the primary (physical) side is transformed to a different one at the secondary (physiological) side zyxwvutsrqponmlkjihgfedcbaZYXW ULTRASONICS. MARCH 1980 0041-624X/80/020081 -04/$02.00 0 1980 IPC Business Press 81