RESUMO
OBJECTIVES: To compare the location of AI markings on screening mammograms with cancer location on diagnostic mammograms, and to classify interval cancers with high AI score as false negative, minimal sign, or true negative. METHODS: In a retrospective study from 2022, we compared the performance of an AI system with independent double reading according to cancer detection. We found 93% (880/949) of the screen-detected cancers, and 40% (122/305) of the interval cancers to have the highest AI risk score (AI score of 10). In this study, four breast radiologists reviewed mammograms from 126 randomly selected screen-detected cancers and all 120 interval cancers with an AI score of 10. The location of the AI marking was stated as correct/not correct in craniocaudal and mediolateral oblique view. Interval cancers with an AI score of 10 were classified as false negative, minimal sign significant/non-specific, or true negative. RESULTS: All screen-detected cancers and 78% (93/120) of the interval cancers with an AI score of 10 were correctly located by the AI system. The AI markings matched in both views for 79% (100/126) of the screen-detected cancers and 22% (26/120) of the interval cancers. For interval cancers with an AI score of 10, 11% (13/120) were correctly located and classified as false negative, 10% (12/120) as minimal sign significant, 26% (31/120) as minimal sign non-specific, and 31% (37/120) as true negative. CONCLUSION: AI markings corresponded to cancer location for all screen-detected cancers and 78% of the interval cancers with high AI score, indicating a potential for reducing the number of interval cancers. However, it is uncertain whether interval cancers with subtle findings in only one view are actionable for recall in a true screening setting. CLINICAL RELEVANCE STATEMENT: In this study, AI markings corresponded to the location of the cancer in a high percentage of cases, indicating that the AI system accurately identifies the cancer location in mammograms with a high AI score. KEY POINTS: ⢠All screen-detected and 78% of the interval cancers with high AI risk score (AI score of 10) had AI markings in one or two views corresponding to the location of the cancer on diagnostic images. ⢠Among all 120 interval cancers with an AI score of 10, 21% (25/120) were classified as a false negative or minimal sign significant and had AI markings matching the cancer location, suggesting they may be visible on prior screening. ⢠Most of the correctly located interval cancers matched only in one view, and the majority were classified as either true negative or minimal sign non-specific, indicating low potential for being detected earlier in a real screening setting.
Assuntos
Neoplasias da Mama , Detecção Precoce de Câncer , Mamografia , Humanos , Neoplasias da Mama/diagnóstico por imagem , Mamografia/métodos , Feminino , Estudos Retrospectivos , Noruega/epidemiologia , Pessoa de Meia-Idade , Detecção Precoce de Câncer/métodos , Idoso , Inteligência Artificial , Interpretação de Imagem Radiográfica Assistida por Computador/métodosRESUMO
Background: Volumetric mammographic density (VMD) measures can be obtained automatically, but it is not clear how these relate to breast cancer risk factors.Methods: The cohort consisted of 46,428 women (ages 49-71 years) who participated in BreastScreen Norway between 2007 and 2014 and had information on VMD and breast cancer risk factors. We estimated means of percent and absolute VMD associated with age, menopausal status, body mass index (BMI), and other factors.Results: The associations between VMD and most breast cancer risk factors were modest, although highly significant. BMI was positively associated with absolute VMD, whereas inversely associated with percent VMD. Percent VMD was inversely associated with a 5-year older age at screening in premenopausal and postmenopausal women (-0.18% vs. -0.08% for percent VMD and -0.11 cm3 vs. -0.03 cm3 for absolute VMD). This difference was largest among postmenopausal women with BMI < 25 kg/m2 (P for interaction with percent VMD < 0.0001), never users of postmenopausal hormone therapy (P for interaction < 0.0001), and premenopausal women with a family history of breast cancer (P for interaction with absolute VMD = 0.054).Conclusions: VMD is associated with several breast cancer risk factors, the strongest being BMI, where the direction of the association differs for percent and absolute VMD. The inverse association with age appears modified by menopausal status and other breast cancer risk factors.Impact: Because VMD methods are becoming widely available in screening and clinical settings, the association between VMD measures and breast cancer risk factors should be investigated further in longitudinal studies. Cancer Epidemiol Biomarkers Prev; 27(9); 1065-74. ©2018 AACR.
Assuntos
Densidade da Mama , Neoplasias da Mama/patologia , Detecção Precoce de Câncer , Mamografia/métodos , Medição de Risco/métodos , Fatores Etários , Idoso , Neoplasias da Mama/epidemiologia , Estudos de Coortes , Estudos Transversais , Feminino , Seguimentos , Humanos , Incidência , Pessoa de Meia-Idade , Noruega/epidemiologia , Pós-Menopausa , Pré-Menopausa , Prognóstico , Fatores de RiscoRESUMO
Most cervical high-grade squamous intraepithelial lesions (HSILs) persist, but approximately one third regress (ie, no HSIL in follow-up biopsies). To identify factors related to histologic proven persistence or regression. Twenty-eight small histologic (marker) biopsies with adequate follow-up were analyzed for human papillomavirus (HPV) genotypes and different immunoquantitative proliferation, cell cycle regulation, and differentiation markers. All cases had a biopsy-interval between the marker and first follow-up biopsy of at least 100 days (median, 8.2 months; range, 3.4-22.5 months). Follow-up was classified as regression or persistence. All lesions were high-risk (hr) HPV and p16 positive, 63% for HPV-16 or HPV-16 mixed with other hr genotypes, while 37% had other hrHPV types. The marker biopsies of the persistent HSILs had lower p53 and retinoblastoma protein (pRb) detected in the deep half of the epithelium (P = 0.001 and 0.02, respectively) than nonpersistent HSILs. The degree of positivity of p16, Ki-67, cyclin D1, lesion extent, positivity of the resection margins, and patient age were all unrelated to persistence or regression. Lesions with HPV-16 or mixed-16 genotypes had a significantly lower percentage of pRb (P = 0.02), p53 (P = 0.02), and cyclin D (P = 0.04) positive nuclei in the deep epithelial layers. In agreement with this, type-16 positive HSILs had a lower regression percentage than those with other HPV types, but the difference was not significant. HSILs with combined negativity/low positivity for p53 and pRb protein in small histologic biopsies are highly likely to persist, contrasting those in which one of these cell cycle regulators is strongly positive (p53 > 15%; pRb > 40%).