Two-view screening and extending the age range: the balance of benefit and risk

Radiation Dose Reduction in Digital Mammography by Deep-Learning Algorithm Image Reconstruction: A Preliminary Study

Purpose

To develop a denoising convolutional neural network-based image processing technique and investigate its efficacy in diagnosing breast cancer using low-dose mammography imaging.

Materials and Methods

A total of 6 breast radiologists were included in this prospective study. All radiologists independently evaluated low-dose images for lesion detection and rated them for diagnostic quality using a qualitative scale. After application of the denoising network, the same radiologists evaluated lesion detectability and image quality. For clinical application, a consensus on lesion type and localization on preoperative mammographic examinations of breast cancer patients was reached after discussion. Thereafter, coded low-dose, reconstructed full-dose, and full-dose images were presented and assessed in a random order.

Results

Lesions on 40% reconstructed full-dose images were better perceived when compared with low-dose images of mastectomy specimens as a reference. In clinical application, as compared to 40% reconstructed images, higher values were given on full-dose images for resolution (p < 0.001); diagnostic quality for calcifications (p < 0.001); and for masses, asymmetry, or architectural distortion (p = 0.037). The 40% reconstructed images showed comparable values to 100% full-dose images for overall quality (p = 0.547), lesion visibility (p = 0.120), and contrast (p = 0.083), without significant differences.

Conclusion

Effective denoising and image reconstruction processing techniques can enable breast cancer diagnosis with substantial radiation dose reduction.

Relative biological effectiveness of mammography X-rays at the level of DNA and chromosomes in lymphocytes

PURPOSE

In many countries, breast cancer screening programs based on periodic mammography exist, giving a large group of women regularly a small dose of ionizing radiation. In order to assess the benefit/risk ratio of those programs the relative biological effectiveness (RBE) of mammography X-rays needs to be determined.

MATERIALS AND METHODS

Blood of five healthy donors was irradiated in vitro with 30 kV X-rays and (60)Co γ-rays with doses between 5 and 2000 mGy. The phosphorylated histone subtype H2A isoform X-foci (γH2AX-foci) technique was used to quantify the number of DNA double-strand breaks (DSB) after irradiation. Chromosomal damage resulting from non- or misrepaired DNA DSB was quantified with the micronucleus (MN)-assay and the sensitivity was improved by counting only centromere negative micronuclei (MNCM-).

RESULTS

The threshold detection dose obtained with the γH2AX-foci test was 10 mGy for mammography X-rays compared to 50 mGy for γ-rays. With the MN-assay respectively MN-centromere-assay threshold detection doses of 100, respectively, 50 mGy were obtained for mammography X-rays compared to 200 respectively 100 mGy for γ-rays. An RBE of 1.4 was obtained with the γH2AX-foci assay. With the MN-assays low-dose RBE values between 3 and 4 were determined.

CONCLUSION

Our results indicate that exposure to mammography X-rays resulted in a modest increase in the induction of DSB compared to γ-rays. However, due to the higher linear energy transfer (LET) of mammography X-rays more clustered DNA damage is produced that is more difficult to repair and results in a more pronounced increase in micronucleus formation.

Computer-aided detection of breast masses: four-view strategy for screening mammography

PURPOSE

To improve the performance of a computer-aided detection (CAD) system for mass detection by using four-view information in screening mammography.

METHODS

The authors developed a four-view CAD system that emulates radiologists' reading by using the craniocaudal and mediolateral oblique views of the ipsilateral breast to reduce false positives (FPs) and the corresponding views of the contralateral breast to detect asymmetry. The CAD system consists of four major components: (1) Initial detection of breast masses on individual views, (2) information fusion of the ipsilateral views of the breast (referred to as two-view analysis), (3) information fusion of the corresponding views of the contralateral breast (referred to as bilateral analysis), and (4) fusion of the four-view information with a decision tree. The authors collected two data sets for training and testing of the CAD system: A mass set containing 389 patients with 389 biopsy-proven masses and a normal set containing 200 normal subjects. All cases had four-view mammograms. The true locations of the masses on the mammograms were identified by an experienced MQSA radiologist. The authors randomly divided the mass set into two independent sets for cross validation training and testing. The overall test performance was assessed by averaging the free response receiver operating characteristic (FROC) curves of the two test subsets. The FP rates during the FROC analysis were estimated by using the normal set only. The jackknife free-response ROC (JAFROC) method was used to estimate the statistical significance of the difference between the test FROC curves obtained with the single-view and the four-view CAD systems.

RESULTS

Using the single-view CAD system, the breast-based test sensitivities were 58% and 77% at the FP rates of 0.5 and 1.0 per image, respectively. With the four-view CAD system, the breast-based test sensitivities were improved to 76% and 87% at the corresponding FP rates, respectively. The improvement was found to be statistically significant (p < 0.0001) by JAFROC analysis.

CONCLUSIONS

The four-view information fusion approach that emulates radiologists' reading strategy significantly improves the performance of breast mass detection of the CAD system in comparison with the single-view approach.

Ethical concerns arising from screening procedures such as mammography and self-referral

Shortly after the discovery of X-rays, their harmful effects became apparent. While the harmful effects of ionising radiation were recognised at an early stage, it took some time before a systematic approach to radiation protection evolved. This began with the development of simple rules to reduce the risk of stochastic effects (such as cancer induction) and the occurrence of deterministic injuries (e.g. cataract formation). These rules have evolved into the basic principles of radiation protection, one of which is the concept of justification, which in its simplest form may be taken as doing more benefit than harm. The concept of justification has particular ethical concerns when applied to population screening procedures. These problems are described in this paper with a particular emphasis on the cases of screening and self-referral procedures.

Risk factors for induction of breast cancer by X-rays and their implications for breast screening

In order to discuss the balance of benefit and radiation risk in a breast screening programme, it is necessary to have numerical values for the probability of breast cancer induction by X-rays, stratified by age. Various sets of such values have been used hitherto, mainly in relation to breast screening in the UK, both within the NHS Screening Programme and more generally for younger age groups. Further sets have recently been reported. These different sets of values are described and discussed, together with the effects of using additive or relative risk models, and the effect of using a dose and dose rate modifying factor (DDREF). Possible new radiation risk factors for breast cancer induction by X-rays, drawn from these sets, are identified. These are used to calculate fresh values of cancer detection/induction ratios, as an index of benefit/risk, for screening age women and for younger women with and without a family history of breast cancer.

Digital mammography: effects of reduced radiation dose on diagnostic performance

PURPOSE

To experimentally determine the relationship between radiation dose and observer accuracy in the detection and discrimination of simulated lesions for digital mammography.

MATERIALS AND METHODS

This HIPAA-compliant study received institutional review board approval; the informed consent requirement was waived. Three hundred normal craniocaudal images were selected from an existing database of digital mammograms. Simulated mammographic lesions that mimicked benign and malignant masses and clusters of microcalcifications (3.3-7.4 cm in diameter) were then superimposed on images. Images were rendered without and with added radiographic noise to simulate effects of reducing the radiation dose to one half and one quarter of the clinical dose. Images were read by five experienced breast imaging radiologists. Results were analyzed to determine effects of reduced dose on overall interpretation accuracy, detection of microcalcifications and masses, discrimination between benign and malignant masses, and interpretation time.

RESULTS

Overall accuracy decreased from 0.83 with full dose to 0.78 and 0.62 with half and quarter doses, respectively. The decrease associated with transition from full dose to quarter dose was significant (P < .01), primarily because of an effect on detection of microcalcifications (P < .01) and discrimination of masses (P < .05). The level of dose reduction did not significantly affect detection of malignant masses (P > .5). However, reduced dose resulted in an increased mean interpretation time per image by 28% (P < .0001).

CONCLUSION

These findings suggest that dose reduction in digital mammography has a measurable but modest effect on diagnostic accuracy. The small magnitude of the effect in response to the drastic reduction of dose suggests potential for modest dose reductions in digital mammography.

The development of mammography

This review traces the development of mammography physics over the last 50 years, concentrating mainly on technological changes and their inter-relations. It has been written for physicists with no specific mammography experience but a general interest in radiology, as much as for those with recent involvement in mammography. Topics covered include industrial film, xerography, intensifying screens, x-ray tube developments, image quality test objects, patient dose and performance checks. Some of these developments were necessary before population screening of healthy women could be considered, while others have resulted from increased opportunities for equipment manufactures which screening programmes created. The standpoint of this review is that of a physicist with long experience in a UK centre where mammography was performed on dedicated equipment well over 40 years ago and where screening has been performed continuously for 30 years.

Enhanced biological effectiveness of low energy X-rays and implications for the UK breast screening programme

Recent radiobiological studies have provided compelling evidence that the low energy X-rays as used in mammography are approximately four times--but possibly as much as six times--more effective in causing mutational damage than higher energy X-rays. Since current radiation risk estimates are based on the effects of high energy gamma radiation, this implies that the risks of radiation-induced breast cancers for mammography X-rays are underestimated by the same factor. The balance of risk and benefit for breast screening have been re-analysed for relative biological effectiveness (RBE) values between 1 and 6 for mammography X-rays. Also considered in the analysis is a change in the dose and dose-rate effectiveness factor (DDREF) from 2 to 1, women with larger than average breasts and implications for women with a family history of breast cancer. A potential increase in RBE to 6 and the adoption of a DDREF of unity does not have any impact on the breast screening programme for women aged 50-70 years screened on a 3 yearly basis. Situations for which breast screening is not justified due to the potential cancers induced relative to those detected (the detection-to-induction ratio (DIR)) are given for a range of RBE and DDREF values. It is concluded that great caution is needed if a programme of early regular screening with X-rays is to be used for women with a family history of breast cancer since DIR values are below 10 (the lowest value considered acceptable for women below 40 years) even for modest increases in the RBE for mammography X-rays.

Breast dose from magnification films in mammography

In mammography, much attention has been given to estimating breast dose from exposures made in conventional "contact" geometry, but much less attention has been given to doses from magnification geometry. Estimation based on contact film dose is difficult because the grid is removed, the geometry is different, and calculation of scatter reaching the film is complex. In this paper, exposures of Perspex blocks of standard thickness to give the same film density in both geometries are compared on 20 X-ray sets of various designs with a nominal magnification 1.8. The ratio derived between doses in each geometry is 2.2+/-0.15, giving an average magnification film dose of 5.0+/-0.3 mGy.

Mammographic breast cancer screening for women previously treated with high breast doses for diseases such as Hodgkin's

In screening of a general population for breast cancer, benefit/risk ratios are of the order of 100/1. For the very small subgroup of women treated by radiotherapy for Hodgkin's disease below age 35, calculations of this type require different considerations, an overview of which is given in this text. It is concluded that although such previous exposures will increase their radiation risk, their increased risk of carrying an undetected breast cancer means that the potential benefit for them of screening is increased even more. In the United Kingdom, the Department of Health has recommended annual screening for these women.