Roger Hunt

Towards optimization in digital chest radiography using Monte Carlo modelling

A Monte Carlo based computer model of the x-ray imaging system was used to investigate how various image quality parameters of interest in chest PA radiography and the effective dose E vary with tube voltage (90-150 kV), additional copper filtration (0-0.5 mm), anti-scatter method (grid ratios 8-16 and air gap lengths 20-40 cm) and patient thickness (20-28 cm) in a computed radiography (CR) system. Calculated quantities were normalized to a fixed value of air kerma (5.0 microGy) at the automatic exposure control chambers. Soft-tissue nodules were positioned at different locations in the anatomy and calcifications in the apical region. The signal-to-noise ratio, SNR, of the nodules and the nodule contrast relative to the contrast of bone (C/C(B)) as well as relative to the dynamic range in the image (C(rel)) were used as image quality measures. In all anatomical regions, except in the densest regions in the thickest patients, the air gap technique provides higher SNR and contrast ratios than the grid technique and at a lower effective dose E. Choice of tube voltage depends on whether quantum noise (SNR) or the contrast ratios are most relevant for the diagnostic task. SNR increases with decreasing tube voltage while C/C(B) increases with increasing tube voltage.

Testing a wavelet based noise reduction method using computer-simulated mammograms

A wavelet based method of noise reduction has been tested for mammography using computer-simulated images for which the truth is known exactly. This method is based on comparing two images at different scales, using a cross-correlation-function as a measure of similarity to define the image modifications in the wavelet domain. The computer-simulated images were calculated for noise-free primary radiation using a quasi-realistic voxel phantom. Two images corresponding to slightly different geometry were produced. Gaussian noise was added with a mean value of zero and a standard deviation equal to 0.25% to 10% of the actual pixel value to simulate quantum noise with a certain level. The added noise could be reduced by more than 70% using the proposed method without any noticeable corruption of the structures for 4% added noise. The results indicate that it is possible to save 50% dose in mammography by producing two images (each 25% of the dose for a standard mammogram). Additionally, a reduction or even a removal of the anatomical noise might be possible and therefore better detection rates of breast cancer in mammography might be achievable.

How does lesion location affect detection performance in digital mammography

We investigated how the thickness of a mass lesion at the observer detection threshold varied with lesion location in the breast. A digital mammography system was used to acquire radiographs of an anthropomorphic breast phantom. Mammograms were acquired with and without mass lesions, thereby permitting a difference image to be generated corresponding to the lesion alone. This isolated lesion was added at a reduced intensity to a non-lesion digital mammogram during a 4-Alternate Forced-Choice (4-AFC) experiment. The lesion intensity that corresponded to a 92% correct performance level in the 4-AFC experiments was determined (I92%). Values of I92% were obtained at different locations in the anthropomorphic phantom, thereby permitting the importance of breast thickness and structured background on lesion detection to be investigated. Lesion detection (I92%) was found to be best in high signal intensity regions (black) and ~25% lower in the low signal regions (white). Lesion detection also appeared to depend on the characteristics of the structured background. The experimental results showed a good correlation with a computation that used a convolution of the lesion and the local background region in the mammogram.

Comparison of Experimental and Theoretical Assessments of Detail Visibility in Digital Mammography

An experimental and modelling study is being made of the influence of tube voltage, target material and exposure on the performance of digital mammography systems. Digital images of the ACR accreditation phantom at 80 mAs, 25–32 kV and at 28 kV, 5–500 mAs were read by eight observers, and the numbers of fibres, specks and masses visible determined. The computer model simulates photon transport through phantom, anti-scatter grid and image receptor. It calculates image pixels and the signal-to-noise ratio per pixel (SNR) for the phantom details. For exposures below 100 mAs, the numbers of fibres and masses visualised were found to be consistent with a constant SNR threshold for detection. For the visualisation of specks, the product of SNR and speck area was approximately constant. At higher mAs, the number of objects visualised was little influenced by exposure, due to the limited dynamic range of the phantom. The results validate the use of computational models to predict performance for simple detection tasks against a uniform background.