The first paper in this issue by Nields reviews the current state-of-the art in screening mammography including digital mammography. The paper also touches upon recent developments in integration of digital mammography with ultrasound for detecting small breast abnormalities, which could be used in screening programs. The paper reviews the results of pilot clinical studies in minimally invasive breast tumor ablation techniques. The paper also describes a new x-ray based intraoperative thermal monitoring method that will enable more accurate minimally invasive treatment of breast tumors.

PET (positron emission tomography) is being increasingly proposed as a new modality for earlier detection and diagnosis of breast cancer, as evidenced by two papers on this topic in the February special issue. In PET, image reconstruction is typically done using standard CT (computed tomography) algorithms on data received concidentally at detectors separated by 180 degrees. However, this approach is limited by the noise sensitivity of standard CT algorithms because of the small numbers of events in PET as well as the use of a cylindrical arrays of detectors. In the next paper, Pawlak and Gordon discuss a new approach using a statistical CT reconstruction where each coincidence event is treated individually by estimating the location of an annihilation event that triggered the coincidence event using previously assigned locations of events processed earlier. The image is finally formed using these estimated annihilation locations. A probability distribution is constructed along each coincidence line and is generated from previous annihilation points using density estimation. The paper reviews different nonparametric methods of density estimation and presents simulation results.

The recent years have seen significant work on the use of breast CT for earlier breast cancer detection. Ideally, breast CT combines the desirable x-ray properties of mammography with the ability to image the breast in 3D. With breast CT, the challenge is to produce an image with sufficient image quality while minimizing radiation dose to the patient. Pan et al. discuss a new breast CT image reconstruction algorithm in the next paper. They derive the sufficiency conditions for exact image reconstruction of a 3D region of interest (ROI) from projections acquired with a reduced helical scan over a small angular range using a recently developed filtered-backprojection algorithm. Preliminary numerical results are also presented.

Vibro-acoustography has gained considerable interest as a new breast imaging modality. This method is based on low-frequency vibrations induced in the object by the radiation force of ultrasound. The paper by Alizad et al. reviews the use of vibro-acoustography in breast imaging. They also discuss detection of microcalcifications and arterial calcifications, and present their recent results using this technique.

Next, Wirth characterizes issues and shortcomings in the performance evaluation of image processing algorithms for computer-aided detection of breast cancer. His paper presents a framework with specific recommendations for establishing a performance evaluation process for breast image processing algorithms using standardized criteria.

We finally conclude with a technique that will help evaluate strategies and treatments for preventing breast cancer. Breast density is a well-known risk factor for breast cancer. The efficacy of new preventative breast cancer treatments can be more quantitatively assessed by measuring changes in breast density. Shepherd et al. present a novel automated method for measuring breast compositional density by comparing the opacity on the mammogram to two reference standards in the form of a small phantom, imaged with each breast, that is added in an unused corner of the mammography field. The paper presents the mathematical derivation of this technique. Phantom experiments using this technique have shown a long-term repeatability of better than 2% for a full range of breast composition density.

In conclusion, the modalities and techniques discussed in these two special issues represent leading-edge methods for detecting and diagnosing breast cancer. It is hoped that these papers will generate sufficient dialogue amongst the members of the breast cancer research community, and will foster a closer working-relationship and collaboration between these researchers eventually helping to push each technique to its limit. Ultimately, these advances will lead to improved and earlier breast cancer detection, reduction of unnecessary biopsies, and improvements in early treatment protocols.

Ecommerce Solutions
Online Payment Solutions