Validation of a new fully automated software for 2D digital mammographic breast density evaluation in predicting breast cancer risk.

We compared accuracy for breast cancer (BC) risk stratification of a new fully automated system (DenSeeMammo-DSM) for breast density (BD) assessment to a non-inferiority threshold based on radiologists' visual assessment. Pooled analysis was performed on 14,267 2D mammograms collected from women aged 48-55 years who underwent BC screening within three studies: RETomo, Florence study and PROCAS. BD was expressed through clinical Breast Imaging Reporting and Data System (BI-RADS) density classification. Women in BI-RADS D category had a 2.6 (95% CI 1.5-4.4) and a 3.6 (95% CI 1.4-9.3) times higher risk of incident and interval cancer, respectively, than women in the two lowest BD categories. The ability of DSM to predict risk of incident cancer was non-inferior to radiologists' visual assessment as both point estimate and lower bound of 95% CI (AUC 0.589; 95% CI 0.580-0.597) were above the predefined visual assessment threshold (AUC 0.571). AUC for interval (AUC 0.631; 95% CI 0.623-0.639) cancers was even higher. BD assessed with new fully automated method is positively associated with BC risk and is not inferior to radiologists' visual assessment. It is an even stronger marker of interval cancer, confirming an appreciable masking effect of BD that reduces mammography sensitivity.

AI for reading screening mammograms: the need for circumspection.

KEY POINTS: • The studies on AI reading of screening mammograms have methodological limitations that undermine the conclusion that AI could do better than radiologists. • These studies do not informon numbers of extra breast cancers found by AI that could represent overdiagnosis. • The ability of AI to detect breast cancers is overestimated because there are no result on biopsy procedures that should be performed when mammograms are positive at AI reading but not at radiology reading.

Computer-aided detection of breast cancers using Haar-like features in automated 3D breast ultrasound.

PURPOSE: Automated 3D breast ultrasound (ABUS) has gained interest in breast imaging. Especially for screening women with dense breasts, ABUS appears to be beneficial. However, since the amount of data generated is large, the risk of oversight errors is substantial. Computer aided detection (CADe) may be used as a second reader to prevent oversight errors. When CADe is used in this fashion, it is essential that small cancers are detected, while the number of false positive findings should remain acceptable. In this work, the authors improve their previously developed CADe system in the initial candidate detection stage. METHODS: The authors use a large number of 2D Haar-like features to differentiate lesion structures from false positives. Using a cascade of GentleBoost classifiers that combines these features, a likelihood score, highly specific for small cancers, can be efficiently computed. The likelihood scores are added to the previously developed voxel features to improve detection. RESULTS: The method was tested in a dataset of 414 ABUS volumes with 211 cancers. Cancers had a mean size of 14.72 mm. Free-response receiver operating characteristic analysis was performed to evaluate the performance of the algorithm with and without using the aforementioned Haar-like feature likelihood scores. After the initial detection stage, the number of missed cancer was reduced by 18.8% after adding Haar-like feature likelihood scores. CONCLUSIONS: The proposed technique significantly improves our previously developed CADe system in the initial candidate detection stage.

Evaluation of the effect of computer-aided classification of benign and malignant lesions on reader performance in automated three-dimensional breast ultrasound.

RATIONALE AND OBJECTIVES: To investigate the effect of a newly developed computer-aided diagnosis (CAD) system on reader interpretation of breast lesions in automated three-dimensional (3D) breast ultrasound. MATERIALS AND METHODS: A CAD system was developed to differentiate malignant lesions from benign lesions including automated lesion segmentation in three dimensions; extraction of lesion features such as spiculation, margin contrast, and posterior acoustic behavior; and a classification stage. Eighty-eight patients with breast lesions were included for an observer study: 47 lesions were malignant and 41 were benign. Eleven readers (seven radiologists and four residents) read the cases with and without CAD. We compared the performance of readers with and without CAD using receiver operating characteristic (ROC) analysis. RESULTS: The CAD system had an area under the ROC curve (AUC) of 0.92 for discriminating benign and malignant lesions, whereas the unaided reader AUC ranged from 0.77 to 0.92. Mean performance of inexperienced readers improved when CAD was used (AUC = 0.85 versus 0.90; P = .007), whereas mean performance of experienced readers did not change with CAD (AUC = 0.89). CONCLUSIONS: By using the CAD system for classification of lesions in automated 3D breast ultrasound, which on its own performed as good as the best readers, the performance of inexperienced readers improved while that of experienced readers remained unaffected.

Causes of death among Belgian professional military radar operators: a 37-year retrospective cohort study.

Concerns have been raised about the safety of the electromagnetic (microwave) emissions of radars, as well as about the ionizing radiation that is emitted on short distances (<2 m) by devices in the radars producing the microwaves. We retrospectively compared the cause specific mortality of 4,417 Belgian professional male military personnel who served in anti-aircraft radar units in Germany between 1963 and 1994 to the cause specific mortality of 2,932 Belgian military personnel who served at the same time in the same place in battalions not equipped with radars. Mean follow-up was 26 years in the 2 groups. Age-adjusted rate ratio (RR) and 95% confidence intervals were derived from a Poisson regression model. Four hundred twenty-four deaths in the battalions groups and 271 deaths in the control battalions occurred (RR = 1.04 (0.96-1.14)). For specific causes of deaths, RRs were 1.22 (1.03-1.47) for neoplasms and 3.51 (1.19-10.3) for symptoms, signs and ill defined conditions. RRs for other causes of death were not significantly different from 1.00. Among deaths from neoplasms, RR for hemolymphatic cancer was 7.22 (1.09-47.9). RRs for other causes of cancer deaths were not significantly different from 1.00. The results for specific tumor types were all based on very small numbers. The RR for cancer increased with decreasing age and suggested that the RR for cancer increased with the duration of stay in radar battalions. In conclusion, exposure of professional military personnel to anti-aircraft radars that existed in Western Europe from the 1960s until the 1990s may have resulted in an increase in the incidence of hemolymphatic cancers. It remains to be established whether this increase is due to microwaves generated by radars or ionizing radiation produced by electronic devices producing the microwaves.

All-cause mortality among Belgian military radar operators: a 40-year controlled longitudinal study.

BACKGROUND: It has been suggested that exposure to radiofrequency/microwaves radiations could be associated with greater health hazards and higher mortality. METHODS: The all-cause mortality of 27,671 Belgian militaries who served from 1963 until 1994 in battalions equipped with radars for anti-aircraft defence was studied over the period 1968-2003. End of the seventies, technical modifications brought to the shielding of the micro-wave generators resulted in a reduction in irradiations. A control group was formed by 16,128 militaries who served during the same period in the same military area but who were never exposed to radars. Administrative procedures for identifying militaries and their vital status were equivalent in the radar and the control groups. RESULTS: The age-standardized mortality ratio (SMR) in the radar battalions was 1.05 (95% CI: 0.95-1.16) in professional militaries, and 0.80 (95% CI: 0.75-0.85) in conscripts. In professional militaries no difference in mortality was found according to duration (less than, or five years or more) or to period of service (before 1978 or after 1977). CONCLUSIONS: During a 40-year period of observation, we found no increase in all-cause mortality in Belgian militaries who were in close contact with radar equipments of anti-aircraft defence battalions.

A multicentre epidemiological study on sunbed use and cutaneous melanoma in Europe.

A large European case-control study investigated the association between sunbed use and cutaneous melanoma in an adult population aged between 18 and 49 years. Between 1999 and 2001 sun and sunbed exposure was recorded in 597 newly diagnosed melanoma cases and 622 controls in Belgium, France, The Netherlands, Sweden and the UK. Fifty three percent of cases and 57% of controls ever used sunbeds. The overall adjusted odds ratio (OR) associated with ever sunbed use was 0.90 (95% CI: 0.71-1.14). There was a South-to-North gradient with high prevalence of sunbed exposure in Northern Europe and lower prevalence in the South (prevalence of use in France 20%, OR: 1.19 (0.68-2.07) compared to Sweden, prevalence 83%, relative risk 0.62 (0.26-1.46)). Dose and lag-time between first exposure to sunbeds and time of study were not associated with melanoma risk, neither were sunbathing and sunburns (adjusted OR for mean number of weeks spent in sunny climates >14 years: 1.12 (0.88-1.43); adjusted OR for any sunburn >14 years: 1.16 (0.9-1.45)). Host factors such as numbers of naevi and skin type were the strongest risk indicators for melanoma. Public health campaigns have improved knowledge regarding risk of UV-radiation for skin cancers and this may have led to recall and selection biases in both cases and controls in this study. Sunbed exposure has become increasingly prevalent over the last 20 years, especially in Northern Europe but the full impact of this exposure on skin cancers may not become apparent for many years.

Sex differences in numbers of nevi on body sites of young European children: implications for the etiology of cutaneous melanoma.

BACKGROUND: Since 1950, the greatest increase in cutaneous melanoma incidence in fair-skinned males took place on the trunk and on the head and neck, whereas in females, it took place on the limbs, mainly on the lower limbs. We examined the influence of sex on numbers and size of nevi on different body sites in white European schoolchildren. METHODS: Information about each holiday period since birth to interview was recorded from parents of six hundred twenty-eight 6- to 7-year-old children in four European cities (Brussels (Belgium), Bochum (Germany), Lyons (France), and Rome (Italy)). Number and anatomic location of small (2-4.9 mm) and large (>/=5 mm) nevi and individual susceptibility to sunlight were independently assessed. RESULTS: After adjustment for host characteristics, sun exposure, and sun protection habits, males had 7% [95% confidence interval (95% CI), -7 to 19] more small nevi than females. However, compared to females, numbers of small nevi were increased by 17% (95% CI, 1-31) on the head and neck and by 16% (95% CI, 2-27) on the trunk and shoulders. In contrast, in males, the number of small nevi on upper limbs was decreased by -5% (95% CI, -26 to 13), and on lower limbs by -8% (95% CI, -34 to 13). The number of large nevi was 6% higher in males than in females (95% CI, -26 to 30). CONCLUSIONS: The sex differences in small nevus distribution in schoolchildren reflect the sex differences in the anatomic distribution of melanoma in adults. Sex differences in sun exposure behaviors, dressing, and clothing would just add their effects to the sex-dependent inherited propensity to develop nevi on a given body site. These results reinforce the hypothesis by which childhood would be a decisive period for the occurrence of sun-induced biological events implicated in the genesis of cutaneous melanoma.

The impact of reimbursement criteria on the appropriateness of 'statin' prescribing.

BACKGROUND: In Belgium, regulations restrict the reimbursement of statins to patients with total serum cholesterol above 250 mg/dl (6.41 mmol/l) after a three-month lipid-lowering diet. We investigated the possible impact of these regulations on characteristics of Belgian patients receiving a lipid-lowering drug. DESIGN: From 1998 to 2000, standard questionnaires on coronary risk factors and treatments with lipid-lowering drugs were completed by 301 GPs sampled at random in the 11 Belgian Provinces. Questionnaires had to be completed for 18 consecutive patients 35 years old or more attending GPs' practices, irrespective of the underlying motive for attendance. RESULTS: Of the 5511 patients included in the study, 1519 (28%) had established coronary disease or diabetes mellitus, or > or = 2 non-cholesterol coronary risk factors. Most (70%) of these patients were not treated with a lipid-lowering drug. Only 22% of patients with established coronary disease, 10% of patients with diabetes mellitus and 9% of patients with > or = 2 coronary risk factors were treated with a statin. Fifty-nine percent of fibrate users and 50% of statin users had a pre-treatment cholesterol level above 250 mg/dl, but had no or only one non-cholesterol coronary risk factor. CONCLUSIONS: In Belgium, the majority of patients at higher risk of coronary event do not benefit from lipid-lowering drugs, particularly the statins. In contrast, one of two statin users and three of five fibrate users should probably not receive the drug prescribed. Regulation based on blood cholesterol level encourages the overlooking of other risk factors relevant for selecting patients having the greatest chance to benefit from statin treatment.

Number and size of nevi are influenced by different sun exposure components: implications for the etiology of cutaneous melanoma (Belgium, Germany, France, Italy).

OBJECTIVE: The solar wavelength involved in melanoma occurrence remains unknown. Sunburns and latitude gradient are indicators of greater exposure to the ultraviolet B radiation. We examined the associations between the components of holiday sun exposure with numbers of small (2-4.9 mm) and large nevi (> or = 5 mm) in young children. METHODS: Information about each holiday period from birth to interview was recorded from parents of 628 6-7-year-old children in four European countries. Sun exposure was characterized using four different components: cumulative duration of holidays, number of holiday periods, sunburn episodes and latitude gradient between habitual living place and holiday places. RESULTS: Individual susceptibility to sunlight, cumulative duration and number of holiday periods were moderately associated with increasing number of small but not of large nevi. The number of small nevi together with sunburn history and latitude gradient were strong predictors of large nevi number. In contrast, sunburn history and latitude gradient were not associated with small nevi. CONCLUSIONS: Exposure to high doses of ultraviolet B radiation would be implicated in the formation of large nevi, while solar radiation other than the ultraviolet B could be implicated in the development of small nevi in children. Given that numbers of large nevi are strong predictors of melanoma in children and adolescents, these results agree with the hypothesis that high ultraviolet B doses are needed for melanoma occurrence, but probably not for nevi formation.

Technical quality control practices in mammography screening programs in 22 countries.

OBJECTIVE: To assess current technical quality control (QC) practices within breast cancer screening or surveillance programs internationally. MATERIALS AND METHODS: The International Breast Cancer Screening Network (IBSN) conducted an extensive survey of quality assurance (QA) activities in developed countries known to have population-based breast cancer screening or surveillance programs in place. Twenty-three countries were sent questionnaires that included items about QA and QC requirements at screening sites, the minimum frequencies of QC test performance, and the personnel responsible for performing QC tests. RESULTS: All 23 countries in the IBSN completed general information on their QA practices. Twenty-two countries responded with complete details on their technical QC practices. The responses indicated a pattern of consistently high-quality control practices among population-based breast cancer screening and surveillance programs. Most programs performed the great majority of QC tests. Variations were observed in the performance frequencies of QC tests and in the personnel responsible for performing QC tests. CONCLUSION: QC practices among population-based breast cancer screening and surveillance programs are highly evolved, with the great majority of responding countries following prescribed QC protocols. Further research is needed on appropriate performance frequencies for mammography QC tests.