PSI - Issue 33

306 Dionysios Linardatos et al. / Procedia Structural Integrity 33 (2021) 304–311 Linardatos / Structural Integrity Procedia 00 (2021) 000 – 000 3 inverse square law at the CMOS detector plane for the RQA-5 beam quality. In order to minimize pile-up distortions, a dedicated collimation system was used; a 1 mm thick lead (Pb) foil with a hole of 400 μm diameter. In addition, the measured X-ray spectra were corrected for the efficiency of the CdTe detector. The entrance surface air Kerma value (ESAK or K a ) at the detector’s surface was calculated according to (Michail et al., 2011b): K a =0.00869× ∑ (1.83⋅10 -6 ⋅Φ 0 (E)⋅E⋅(μ en (E)/ϱ) air ) E 0 E min (1) Where Φ 0 (E) is the measured X-ray photon fluence (photons/mm 2 ) at energy E and (μ en (E)/ϱ) air is the X-ray mass energy absorption coefficient of air at energy E, obtained from the literature (Greening, 2017). The X-ray exposure at the entrance surface of the CMOS photodiode array detector was measured for a range of tube current time products (mAs) (Linardatos et al., 2021). 2.3.1. Noise Equivalent Quanta (NEQ) The ratio of the signal response and the amplitude variance, properly normalized, provides information about the maximum available SNR as a function of frequency and, squared, can also be expressed as the NEQ (Dobbins III, 1995, 2000; Michail et al., 2014). This concept relates the noise equivalent quanta with the detective quantum efficiency (DQE; which is the ratio of the detector output to the input SNR; DQE = SNR 2 out / SNR 2 in ) (Kandarakis et al., 2005; Kandarakis, 2016; Seferis et al., 2018) and the input signal. It can be given as (Karpetas et al., 2017; Michail et al., 2016a, 2018a): NEQ(u) = DQE(u)⋅Φ 0 (2) The latter provides an index of the signal to noise ratio associated with the diagnostic value of the medical image (Bosmans et al., 2005; Dobbins III, 2000, 1995), assuming that the large area signal Φ 0 , measured with the portable X-ray spectrometer, is linearly related to the input signal (Michail et al., 2014). 2.3.2. Information Capacity (IC) The concept of information capacity (IC) has been introduced within the context of Shannon's information theory, in order to assess image information content (Jones, 1960, 1962; Kanamori and Murashima, 1970; Kandarakis et al., 2001; Maiorchuk et al., 1974; Shannon, 1948; Wagner et al., 1979). In digital imaging, the continuous spatial distribution of an optically generated image is sampled by the discrete sensitive pixels on a photodiode array, whose outputs are converted into digitized signals and stored in an image processing system for numerical evaluation. According to Shannon’s information theory, the image IC per unit of image area, may be defined as follows (Shannon, 1948): IC = lim T→∞ ( log 2 N S /T ) = n P log 2 N S (3) Where N S are the different signal intensity levels, T is the duration of the signal and n p the number of pixels of the imaging system. As it has been shown from previous studies, information capacity can also be expressed in terms of the detective quantum efficiency of the digital imaging system: IC = π ∫ log 2 (1+(DQE(u)⋅Φ 0 ))udu 0 ∞ (4) Where 0 ( )   is the measured X-ray photon fluence (photons/mm 2 ). Consequently: 2.3. Image quality