Performance of a high fill factor, indirect detection prototype flat-panel imager for mammography

Youcef El-Mohri, Larry E. Antonuk, Qihua Zhao, Yi Wang, Yixin Li, Hong Du, Amit Sawant

Research output: Contribution to journalArticlepeer-review

43 Scopus citations


Empirical and theoretical investigations of the performance of a small-area, high-spatial-resolution, active matrix flat-panel imager, operated under mammographic conditions, is reported. The imager is based on an indirect detection array incorporating a continuous photodiode design, as opposed to the discrete photodiode design employed in conventional flat-panel imagers. Continuous photodiodes offer the prospect of higher fill factors, particularly for arrays with pixel pitches below ∼100 μm. The array has a pixel-to-pixel pitch of 75 μm and a pixel format of 512x512, resulting in an active area of ∼3.8×3.8 cm2. The array was coupled to two commercially available, structured CsI:Tl scintillators of ∼150 μm thickness: one optimized for high light output (FOS-HL) and the other for high spatial resolution (FOS-HR), resulting in a pair of imager configurations. Measurements of sensitivity, modulation transfer function (MTF), noise power spectra (NPS), and detective quantum efficiency (DQE) were performed with a 26 kVp mammography beam at exposures ranging from ∼0.5 to ∼19 mR. MTF results from both CsI:Tl scintillators show that the array demonstrates good spatial resolution, indicating effective isolation between adjacent pixels. The effect of additive noise of the system on DQE was observed to be significantly higher for the FOS-HR scintillator compared to the FOS-HL scintillator due to lower sensitivity of the former. For the FOS-HL scintillator, DQE performance was generally high at high exposures, limited by the x-ray quantum efficiency, Swank factor and the MTF of the scintillators. For both scintillators, the DQE performance degrades at lower exposures due to the relatively large contribution of additive noise. Theoretical calculations based on a cascaded systems model were found to be in general agreement with the empirically determined NPS and DQE values. Finally, such calculations were used to predict potential DQE performance for hypothetical 50 μm pixel pitch imagers, employing similar continuous photodiode design and realistic inputs derived from the empirical measurements.

Original languageEnglish (US)
Pages (from-to)315-327
Number of pages13
JournalMedical physics
Issue number1
StatePublished - Jan 2007


  • Active matrix flat-panel imager
  • Cascaded systems model
  • Continuous photodiode
  • Detective quantum efficiency
  • Digital mammography

ASJC Scopus subject areas

  • Biophysics
  • Radiology Nuclear Medicine and imaging


Dive into the research topics of 'Performance of a high fill factor, indirect detection prototype flat-panel imager for mammography'. Together they form a unique fingerprint.

Cite this