This work addresses a variety of problems in quantitative measurement of Computed Tomography (CT) and Magnetic Resonance Imaging (MRI) images. Our CT work involves studies comparing image resolution using Iterative Reconstruction (IR) with resolution using Filtered Back Projection (FBP), for low-contrast and high-contrast objects in phantom images across vendors and exposure levels. MRI work involves the use of the NIST breast MRI phantom (see Figure 1), to quantify the effects of a wide range of imaging parameters.
Computed Tomography (CT) is one of the most highly utilized medical imaging modalities. 70 million CT examinations are performed annually in the United States. There is great concern about the risks related to radiation-induced cancer. A recent study found an increase, from 0.4 % (1996) to 1.5-2.0 % (current), in the estimated cancer risk from CT radiation of all cancers in the United States. CT radiation has been associated with the development of malignancies in children such as leukemia and brain cancer. Therefore, it is imperative for CT radiation exposure to be as low as reasonably achievable. The use of Iterative Reconstruction (IR) has been heavily promoted by vendors to reduce radiation exposure in CT imaging acquisition. IR is an iterative algorithm used to reconstruct 2D and 3D images from projections of an object. The main advantages of IR over Filtered Back Projection (FBP) are the ability to incorporate attenuation corrections and reduction in image noise.
We used low-contrast images of Module 2 of the American College of Radiology (ACR) phantom, which was designed to test a wide range of scanner parameters. We have found that Iterative Reconstruction (IR) yielded superior low-contrast detectability compared to Filtered Back Projection (FBP) across three vendors and three exposure levels. Furthermore, we concluded that IR has the potential ability to reduce exposure by up to 60 % (12 mGy vs 7.2 mGy) without substantial loss of low-contrast detectability compared to FBP. In addition, there is substantial variability among the major vendors not only in baseline image quality using FBP but also in the relative benefit of the use of IR. We are currently analyzing a similar study using the high-contrast Module 4 of the ACR phantom (see Figure 2).
Quantitative magnetic resonance imaging (MRI) is increasingly being used for diagnosis and monitoring of breast cancer, so it is important that MRI measurements are standardized and quality control protocols are implemented. The NIST's Magnetics group is involved in the creation and measurement of a breast phantom, designed with a distribution mix of fats and fibroglandular tissue, for testing and evaluation of MRI measurements. Our group is assisting with this process by providing automated segmentation and measurement of individual components within the phantom from images under a wide range of imaging parameters.
Group 3A of the Quantitative Imaging Biomarkers Alliance (QIBA) of the Radiological Society of North America (RSNA), studies the variability of lesion volume measurements of CT images due to algorithm use. An initial challenge was designed to test both variability and bias in measurements of phantom lung lesions. A second challenge using clinical test-retest sets was also carried out, to measure both reproducibility in test-retest measurements for individual algorithms and to measure reproducibilty of these measurement across different algorithms. The goal of these studies is to understand the error due to the use of different volume measurement algorithms to promote the use of volume measurement as a biomarker for change of the state of a disease.
Paper: Ganesh Saiprasad, Olav Christianson, Joseph Chen, Zhitong Yang, Alden Dima, James Filliben, Adele Peskin, Eliot Siegel, and Ehsan Samei, Evaluation of Low-Contrast Detectability of Iterative Reconstruction across Multiple Institutions, Manufacturers and Exposure Levels, submitted to Radiology(download pdf)
Paper: Maria Athelogou, Hyun J Kim, Alden Dima, Ganesh Saiprasad, Adele Peskin, Hubert Beaumont, et.al., Algorithm variability in the estimation of lung nodule volume from phantom CT scans: results of the QIBA 3A public challenge, submitted to Investigative Radiology.(download pdf)