Automatic Detection and Classification of Mammograms Using Improved Extreme Learning Machine with Deep Learning

Background and objectiveBreast cancer, the most intrusive form of cancer affecting women globally. Next to lung cancer, breast cancer is the one that provides a greater number of cancer deaths among women. In recent times, several intelligent methodologies were come into existence for building an effective detection and classification of such noxious type of cancer. For further improving the rate of early diagnosis and for increasing the life span of victims, optimistic light of research is essential in breast cancer classification. Accordingly, a new customized method of integrating the concept of deep learning with the extreme learning machine (ELM), which is optimized using a simple crow-search algorithm (ICS-ELM). Thus, to enhance the state-of-the-art workings, an improved deep feature-based crow-search optimized extreme learning machine is proposed for addressing the health-care problem. The paper pours a light-of-research on detecting the input mammograms as either normal or abnormal. Subsequently, it focuses on further classifying the type of abnormal severities i.e., benign type or malignant.Materials and methodsThe digital mammograms for this work are taken from the Curated Breast Imaging Subset of DDSM (CBIS-DDSM), Mammographic Image Analysis Society (MIAS), and INbreast datasets. Herein, the work employs 570 digital mammograms (250 normal, 200 benign and 120 malignant cases) from CBIS-DDSM dataset, 322 digital mammograms (207 normal, 64 benign and 51 malignant cases) from MIAS database and 179 full-field digital mammograms (66 normal, 56 benign and 57 malignant cases) from INbreast dataset for its evaluation. The work utilizes ResNet-18 based deep extracted features with proposed Improved Crow-Search Optimized Extreme Learning Machine (ICS-ELM) algorithm.ResultsThe proposed work is finally compared with the existing Support Vector Machines (RBF kernel), ELM, particle swarm optimization (PSO) optimized ELM, and crow-search optimized ELM, where the maximum overall classification accuracy is obtained for the proposed method with 97.193% for DDSM, 98.137% for MIAS and 98.266% for INbreast datasets, respectively.ConclusionThe obtained results reveal that the proposed Computer-Aided-Diagnosis (CAD) tool is robust for the automatic detection and classification of breast cancer.     

钼靶和超声检查在乳腺癌临床诊断的准确性的比较分析

摘要 目的：比较与分析钼靶和超声检查在乳腺癌临床诊断的准确性。方法：2018年8月到2021年1月选择在本院进行诊治的乳腺肿瘤患者110例作为研究对象,所有患者都给予钼靶和超声检查,记录影像学特征并判断诊断价值。结果：在110例患者中,病理诊断为乳腺良性肿瘤76例、乳腺癌34例。恶性组钼靶的分叶征、钙化、大角征、毛刺征等比例高于良性组,病灶大小也高于良性组(P<0.05)。恶性组超声的形态不规则、边缘不光整、高回声晕、回声衰减、微钙化等比例高于良性组(P<0.05)。钼靶乳腺影像报告及数据系统(Breast imaging report and data system,BI-RADS)判断为乳腺良性肿瘤72例,乳腺癌38例;超声BI-RADS判断为乳腺良性肿瘤75例,乳腺癌35例,钼靶鉴别诊断乳腺癌的敏感性为93.4%,特异性为97.1%,准确性为94.5%;超声鉴别诊断乳腺癌的敏感性为98.7%,特异性为100.0%,准确性为99.1%。多因素logistic回归分析显示病灶大小、分叶征、回声衰减、毛刺征为导致误诊的重要因素(P<0.05)。结论：乳腺癌在钼靶和超声检查中都有明显的征象特征,超声诊断的准确性更高,病灶大小、分叶征、回声衰减、毛刺征为影响诊断效果的很重要因素。     

Staging practices and breast cancer stage among population-based registries in the MENA region

BackgroundAvailability of stage information by population-based cancer registries (PBCR) remains scarce for diverse reasons. Nevertheless, stage is critical cancer control information particularly for cancers amenable to early detection. In the framework of the Global Initiative for Cancer Registry Development (GICR), we present the status of stage data collection and dissemination among registries in the Middle East and Northern Africa (MENA) region as well as the stage distribution of breast cancer patients.MethodsA web-based survey exploring staging practices and breast cancer stage was developed and sent to 30 PBCR in 18 countries of the MENA region.ResultsAmong 23 respondent PBCR, 21 collected stage data, the majority (80%) for all cancers. Fourteen registries used a single classification (9 TNM and 5 SEER), 7 used both staging systems in parallel. Out of 12,888 breast cancer patients (seven registries) 27.7% had unknown TNM stage (11.1% in Oman, 46% in Annaba). When considering only cases with known stage, 65.3% were early cancers (TNM I+II), ranging from 57.9% in Oman to 83.3% in Batna (Algeria), and 9.9% were stage IV cancers. Among the nine registries providing SEER Summary stage for breast cancer cases, stage was unknown in 19% of the cases, (0 in Bahrain, 39% in Kuwait). Stage data were largely absent from the published registry reports.ConclusionDespite wide stage data collection by cancer registries, missing information and low dissemination clearly limit informing efforts on early detection. The use of two classification systems in parallel implies additional workload and might undermine completeness. The favourable results of early cancer (TNM I+II) in two thirds of breast cancer patients needs to be interpreted with caution and followed up in time. Although efforts to improve quality of stage data are needed, our findings are particularly relevant to the WHO Global Breast Cancer Initiative.     

A Hybrid CNN-SVM Threshold Segmentation Approach for Tumor Detection and Classification of MRI Brain Images

ObjectiveIn this research paper, the brain MRI images are going to classify by considering the excellence of CNN on a public dataset to classify Benign and Malignant tumors.Materials and MethodsDeep learning (DL) methods due to good performance in the last few years have become more popular for Image classification. Convolution Neural Network (CNN), with several methods, can extract features without using handcrafted models, and eventually, show better accuracy of classification. The proposed hybrid model combined CNN and support vector machine (SVM) in terms of classification and with threshold-based segmentation in terms of detection.ResultThe findings of previous studies are based on different models with their accuracy as Rough Extreme Learning Machine (RELM)-94.233%, Deep CNN (DCNN)-95%, Deep Neural Network (DNN) and Discrete Wavelet Autoencoder (DWA)-96%, k-nearest neighbors (kNN)-96.6%, CNN-97.5%. The overall accuracy of the hybrid CNN-SVM is obtained as 98.4959%.ConclusionIn today's world, brain cancer is one of the most dangerous diseases with the highest death rate, detection and classification of brain tumors due to abnormal growth of cells, shapes, orientation, and the location is a challengeable task in medical imaging. Magnetic resonance imaging (MRI) is a typical method of medical imaging for brain tumor analysis. Conventional machine learning (ML) techniques categorize brain cancer based on some handicraft property with the radiologist specialist choice. That can lead to failure in the execution and also decrease the effectiveness of an Algorithm. With a brief look came to know that the proposed hybrid model provides more effective and improvement techniques for classification.     

Ability to Generate Patient-Derived Breast Cancer Xenografts Is Enhanced in Chemoresistant Disease and Predicts Poor Patient Outcomes

BackgroundBreast cancer patients who are resistant to neoadjuvant chemotherapy (NeoCT) have a poor prognosis. There is a pressing need to develop in vivo models of chemo resistant tumors to test novel therapeutics. We hypothesized that patient-derived breast cancer xenografts (BCXs) from chemo- naïve and chemotherapy-exposed tumors can provide high fidelity in vivo models for chemoresistant breast cancers.MethodsPatient tumors and BCXs were characterized with short tandem repeat DNA fingerprinting, reverse phase protein arrays, molecular inversion probe arrays, and next generation sequencing.ResultsForty-eight breast cancers (24 post-chemotherapy, 24 chemo-naïve) were implanted and 13 BCXs were established (27%). BCX engraftment was higher in TNBC compared to hormone-receptor positive cancer (53.8% vs. 15.6%, p = 0.02), in tumors from patients who received NeoCT (41.7% vs. 8.3%, p = 0.02), and in patients who had progressive disease on NeoCT (85.7% vs. 29.4%, p = 0.02). Twelve patients developed metastases after surgery; in five, BCXs developed before distant relapse. Patients whose tumors developed BCXs had a lower recurrence-free survival (p = 0.015) and overall survival (p<0.001). Genomic losses and gains could be detected in the BCX, and three models demonstrated a transformation to induce mouse tumors. However, overall, somatic mutation profiles including potential drivers were maintained upon implantation and serial passaging. One BCX model was cultured in vitro and re-implanted, maintaining its genomic profile.ConclusionsBCXs can be established from clinically aggressive breast cancers, especially in TNBC patients with poor response to NeoCT. Future studies will determine the potential of in vivo models for identification of genotype-phenotype correlations and individualization of treatment.     

Accuracy and completeness of the New Zealand Cancer Registry for staging of invasive breast cancer

PurposePopulation based cancer registries are an invaluable resource for monitoring incidence and mortality for many types of cancer. Research and healthcare decisions based on cancer registry data rely on the case completeness and accuracy of recorded data. This study was aimed at assessing completeness and accuracy of breast cancer staging data in the New Zealand Cancer Registry (NZCR) against a regional breast cancer register.MethodologyData from 2562 women diagnosed with invasive primary breast cancer between 1999 and 2011 included in the Waikato Breast Cancer Register (WBCR) were used to audit data held on the same individuals by the NZCR. WBCR data were treated as the benchmark.ResultsOf 2562 cancers, 315(12.3%) were unstaged in the NZCR. For cancers with a known stage in the NZCR, staging accuracy was 94.4%. Lower staging accuracies of 74% and 84% were noted for metastatic and locally invasive (involving skin or chest wall) cancers, respectively, compared with localized (97%) and lymph node positive (94%) cancers. Older age (>80 years), not undergoing therapeutic surgery and higher comorbidity score were significantly (p < 0.01) associated with unstaged cancer. The high proportion of unstaged cancer in the NZCR was noted to have led to an underestimation of the true incidence of metastatic breast cancer by 21%. Underestimation of metastatic cancer was greater for Māori (29.5%) than for NZ European (20.6%) women. Overall 5-year survival rate for unstaged cancer (NZCR) was 55.9%, which was worse than the 5-year survival rate for regional (77.3%), but better than metastatic (12.9%) disease.ConclusionsUnstaged cancer and accuracy of cancer staging in the NZCR are major sources of bias for the NZCR based research. Improving completeness and accuracy of staging data and increasing the rate of TNM cancer stage recording are identified as priorities for strengthening the usefulness of the NZCR.     

Assessment of surveillance versus etiologic factors in the reciprocal association between papillary thyroid cancer and breast cancer

BackgroundMutually increased risks for thyroid and breast cancer have been reported, but the contribution of etiologic factors versus increased medical surveillance to these associations is unknown.MethodsLeveraging large-scale US population-based cancer registry data, we used standardized incidence ratios (SIRs) to investigate the reciprocal risks of thyroid and breast cancers among adult females diagnosed with a first primary invasive, non-metastatic breast cancer (N = 652,627) or papillary thyroid cancer (PTC) (N = 92,318) during 2000–2017 who survived ≥1-year.ResultsPTC risk was increased 1.3-fold [N = 1434; SIR = 1.32; 95 % confidence interval (CI) = 1.25–1.39] after breast cancer compared to the general population. PTC risk declined significantly with time since breast cancer (Poisson regression = P_trend <0.001) and was evident only for tumors ≤2 cm in size. The SIRs for PTC were higher after hormone-receptor (HR)+ (versus HR-) and stage II or III (versus stage 0-I) breast tumors. Breast cancer risk was increased 1.2-fold (N = 2038; SIR = 1.21; CI = 1.16–1.26) after PTC and was constant over time since PTC but was only increased for stage 0-II and HR + breast cancers.ConclusionAlthough some of the patterns by latency, stage and size are consistent with heightened surveillance contributing to the breast-thyroid association, we cannot exclude a role of shared etiology or treatment effects.     

17号染色体不同倍体与乳腺癌临床病理特征的相关性

摘要 目的：探讨17号染色体不同倍体与乳腺癌临床病理特征的相关性。方法：选取2018年1月至2019年12月确诊为乳腺癌的患者78例(乳腺癌组)与乳腺良性肿瘤患者78例(良性组),采用原位荧光杂交检测所有患者的病灶组织染色体不同倍体情况,分析患者的临床病理特征并进行相关性分析。结果：乳腺癌组17号染色体的多倍体率为85.9 %,显著高于良性组(3.8 %,P<0.05)。不同年龄、性别、发病位置、病理类型乳腺癌患者的17号染色体多倍体率对比差异无统计学意义(P>0.05),不同淋巴结转移、组织学分化、临床分期、ER阳性、PR阳性患者的17号染色体多倍体率对比差异具有统计学意义(P<0.05)。Pearson分析显示17号染色体多倍体率与乳腺癌患者的淋巴结转移、组织学分化、临床分期、ER阳性、PR阳性存在显著相关性(P<0.05);多因素Logistic回归分析显示淋巴结转移、组织学分化、临床分期、ER阳性、PR阳性都为17号染色体多倍体的主要危险影响因素(P<0.05)。结论：乳腺癌患者多伴随有17号染色体多倍体,与其患者的淋巴结转移、组织学分化、临床分期、ER阳性、PR阳性等临床病理特征显著相关。     

乳腺良性肿块与乳腺癌患者的超声弹性成像对比

目的:对比乳腺良性肿块与乳腺癌患者的超声弹性成像,明确超声弹性成像的应用价值。方法:选取2014年5月-2016年1月我院乳腺肿块患者128人次共146例肿块,根据病理结果分为乳腺良性肿块和乳腺癌,比较超声弹性成像与病理结果。结果:128个患者共计肿块146例,99例结节为良性肿块,其中32例为乳腺纤维腺瘤,29例为乳腺增生结节,20例为乳腺脂肪瘤,6例为乳腺血管脂肪瘤,4例为乳腺导管腺瘤,8例为乳腺导管内乳头状瘤;47例肿块为恶性,其中37例肿块为浸润性导管癌,9例肿块为粘液腺癌,1例肿块为硬癌。乳腺良性肿块患者81人次共99例,其中1分43例(43.43%),2分34例(34.34%),3分18例(18.18%),4分4例(4.04%);乳腺癌患者47例,其中3分9例(19.15%),4分20例(42.55%),5分18例(38.30%)。超声弹性成像鉴别乳腺良性肿块与乳腺癌的灵敏度为95.96%,特异性为80.85%,准确度为91.10%,阴性预测值为90.48%,阳性预测值为91.35%。结论:超声弹性成像鉴别乳腺良性肿块与乳腺癌的灵敏度高达95.96%,具有较高准确度,可辅助诊断乳腺疾病。     

Discovery of pathway biomarkers from coupled proteomics and systems biology methods

Background

Breast cancer is worldwide the second most common type of cancer after lung cancer. Plasma proteome profiling may have a higher chance to identify protein changes between plasma samples such as normal and breast cancer tissues. Breast cancer cell lines have long been used by researches as model system for identifying protein biomarkers. A comparison of the set of proteins which change in plasma with previously published findings from proteomic analysis of human breast cancer cell lines may identify with a higher confidence a subset of candidate protein biomarker.

Results

In this study, we analyzed a liquid chromatography (LC) coupled tandem mass spectrometry (MS/MS) proteomics dataset from plasma samples of 40 healthy women and 40 women diagnosed with breast cancer. Using a two-sample t-statistics and permutation procedure, we identified 254 statistically significant, differentially expressed proteins, among which 208 are over-expressed and 46 are under-expressed in breast cancer plasma. We validated this result against previously published proteomic results of human breast cancer cell lines and signaling pathways to derive 25 candidate protein biomarkers in a panel. Using the pathway analysis, we observed that the 25 “activated” plasma proteins were present in several cancer pathways, including ‘Complement and coagulation cascades’, ‘Regulation of actin cytoskeleton’, and ‘Focal adhesion’, and match well with previously reported studies. Additional gene ontology analysis of the 25 proteins also showed that cellular metabolic process and response to external stimulus (especially proteolysis and acute inflammatory response) were enriched functional annotations of the proteins identified in the breast cancer plasma samples. By cross-validation using two additional proteomics studies, we obtained 86% and 83% similarities in pathway-protein matrix between the first study and the two testing studies, which is much better than the similarity we measured with proteins.

Conclusions

We presented a ‘systems biology’ method to identify, characterize, analyze and validate panel biomarkers in breast cancer proteomics data, which includes 1) t statistics and permutation process, 2) network, pathway and function annotation analysis, and 3) cross-validation of multiple studies. Our results showed that the systems biology approach is essential to the understanding molecular mechanisms of panel protein biomarkers.

     

Murine model of hepatic breast cancer

Background and aimsBreast cancer is the most common cancer in women and the second leading cause of cancer-related deaths in this population. Breast cancer related deaths have declined due to screening and adjuvant therapies, yet a driving clinical need exists to better understand the cause of the deadliest aspect of breast cancer, metastatic disease. Breast cancer metastasizes to several distant organs, the liver being the third most common site. To date, very few murine models of hepatic breast cancer exist.MethodsIn this study, a novel murine model of liver breast cancer using the MDA-MB-231 cell line is introduced as an experimental (preclinical) model.ResultsHistological typing revealed consistent hepatic breast cancer tumor foci. Common features of the murine model were vascular invasion, lung metastasis and peritoneal seeding.ConclusionsThe novel murine model of hepatic breast cancer established in this study provides a tool to be used to investigate mechanisms of hepatic metastasis and to test potential therapeutic interventions.     

Deep Learning-Based Metaheuristic Weighted K-Nearest Neighbor Algorithm for the Severity Classification of Breast Cancer

ObjectiveThe most widespread and intrusive cancer type among women is breast cancer. Globally, this type of cancer causes more mortality among women, next to lung cancer. This made the researchers to focus more on developing effective Computer-Aided Detection (CAD) methodologies for the classification of such deadly cancer types. In order to improve the rate of survival and earlier diagnosis, an optimistic research methodology is required in the classification of breast cancer. Consequently, an improved methodology that integrates the principle of deep learning with metaheuristic and classification algorithms is proposed for the severity classification of breast cancer. Hence to enhance the recent findings, an improved CAD methodology is proposed for redressing the healthcare problem.Material and MethodsThe work intends to cast a light-of-research towards classifying the severities present in digital mammogram images. For evaluating the work, the publicly available MIAS, INbreast, and WDBC databases are utilized. The proposed work employs transfer learning for extricating the features. The novelty of the work lies in improving the classification performance of the weighted k-nearest neighbor (wKNN) algorithm using particle swarm optimization (PSO), dragon-fly optimization algorithm (DFOA), and crow-search optimization algorithm (CSOA) as a transformation technique i.e., transforming non-linear input features into minimal linear separable feature vectors.ResultsThe results obtained for the proposed work are compared then with the Gaussian Naïve Bayes and linear Support Vector Machine algorithms, where the highest accuracy for classification is attained for the proposed work (CSOA-wKNN) with 84.35% for MIAS, 83.19% for INbreast, and 97.36% for WDBC datasets respectively.ConclusionThe obtained results reveal that the proposed Computer-Aided-Diagnosis (CAD) tool is robust for the severity classification of breast cancer.     

Small molecules baicalein and cinnamaldehyde are potentiators of measles virus-induced breast cancer oncolysis

BackgroundAlthough the breast cancer mortality has slowed down from 2008 to 2017, breast cancer incidence rate continues to rise and thus, new and/or improved treatments are highly needed. Among them, oncolytic virotherapy which has the ability of facilitating the antitumor adaptive immunity, appears as a promising anticancer therapy. Oncolytic measles virus (MV) is particularly suitable for targeting breast cancer due to the upregulation of MV's receptor nectin-4. Nonetheless, with limited clinical success currently, ways of boosting MV-induced breast cancer oncolysis are therefore necessary. Oncolytic virotherapy alone and combined with chemotherapeutic drugs are two strategic areas with intensive development for the search of anticancer drugs. Considering that baicalein (BAI) and cinnamaldehyde (CIN) have demonstrated antitumor properties against multiple cancers including breast cancer, they could be good partners for MV-based oncolytic virotherapy.PurposeTo assess the in vitro effect of BAI and CIN with MV and assess their combination effects.MethodsWe examined the combinatorial cytotoxic effect of oncolytic MV and BAI or CIN on MCF-7 breast cancer cells. Potential anti-MV activities of the phytochemicals were first investigated in vitro to determine the optimal combination model. Synergism of MV and BAI or CIN was then evaluated in vitro by calculating the combination indices. Finally, cell cycle analysis and apoptosis assays were performed to confirm the mechanism of synergism.ResultsOverall, the viral sensitization combination modality using oncolytic MV to first infect MCF-7 breast cancer cells followed by drug treatment with BAI or CIN was found to produce significantly enhanced tumor killing. Further mechanistic studies showed that the combinations ‘MV-BAI’ and ‘MV-CIN’ display synergistic anti-breast cancer effect, mediated by elevated apoptosis.ConclusionWe demonstrated, for the first time, effective combination of oncolytic MV with BAI or CIN that could be further explored and potentially developed into novel therapeutic strategies targeting nectin-4-marked breast cancer cells.     

Breast cancer incidence by country of birth among immigrant women in British Columbia,Canada

IntroductionBreast cancer rates vary internationally and between immigrant and non-immigrant populations. We describe breast cancer incidence by birth region and country in British Columbia, Canada.MethodsWe linked population-based health and immigration databases for a population with >1.29 million immigrants to assess breast cancer incidence among immigrant and non-immigrant women. We report age-standardized incidence ratios (SIRs) by birth region and country using non-immigrant women as the standard.ResultsSIRs varied widely by both birth country and region. Low rates were found for South (SIR = 0.52, 95% CI: 0.47,0.59) and East Asian (SIR = 0.75, 95% CI: 0.72,0.79) women and a higher rate for Western Europeans (SIR = 1.15, 95% CI: 1.01,1.30).ConclusionThere is considerable variation in SIRs across some of British Columbia’s largest immigrant populations and several demonstrate significantly different risk profiles compared to non-immigrants. These findings provide unique data to support breast cancer prevention and control.     

Asporin Is a Fibroblast-Derived TGF-β1 Inhibitor and a Tumor Suppressor Associated with Good Prognosis in Breast Cancer

BackgroundBreast cancer is a leading malignancy affecting the female population worldwide. Most morbidity is caused by metastases that remain incurable to date. TGF-β1 has been identified as a key driving force behind metastatic breast cancer, with promising therapeutic implications.ConclusionsOur data show that asporin is a stroma-derived inhibitor of TGF-β1 and a tumor suppressor in breast cancer. High asporin expression is significantly associated with less aggressive tumors, stratifying patients according to the clinical outcome. Future pre-clinical studies should consider options for increasing asporin expression in TNBC as a promising strategy for targeted therapy.     

Generation,analysis and functional annotation of expressed sequence tags from the sheepshead minnow (Cyprinodon variegatus)

Background

Genomic islands (GIs) are clusters of alien genes in some bacterial genomes, but not be seen in the genomes of other strains within the same genus. The detection of GIs is extremely important to the medical and environmental communities. Despite the discovery of the GI associated features, accurate detection of GIs is still far from satisfactory.

Results

In this paper, we combined multiple GI-associated features, and applied and compared various machine learning approaches to evaluate the classification accuracy of GIs datasets on three genera: Salmonella, Staphylococcus, Streptococcus, and their mixed dataset of all three genera. The experimental results have shown that, in general, the decision tree approach outperformed better than other machine learning methods according to five performance evaluation metrics. Using J48 decision trees as base classifiers, we further applied four ensemble algorithms, including adaBoost, bagging, multiboost and random forest, on the same datasets. We found that, overall, these ensemble classifiers could improve classification accuracy.

Conclusions

We conclude that decision trees based ensemble algorithms could accurately classify GIs and non-GIs, and recommend the use of these methods for the future GI data analysis. The software package for detecting GIs can be accessed at http://www.esu.edu/cpsc/che_lab/software/GIDetector/.

     

Type 2 diabetes,obesity, and breast cancer risk among Japanese women of the atomic bomb survivor cohort

BackgroundMuch less is known about diabetes than obesity as a predictor of breast cancer incidence and most previous studies have been conducted in white populations. Therefore, this project within the Radiation Effects Research Foundation’s cohort of Japanese atomic bomb survivors aimed to determine the independent contributions of obesity and diabetes to develop breast cancer.MethodsAfter excluding women with unknown A-bomb radiation dose, a radiation dose of ≥100 mGy, a pre-existing history of breast cancer, and missing body mass index (BMI), the analysis included 29,818 women. Breast cancer status and deaths until 2009 were identified from cancer registries and vital records. Cox regression with age as the time metric was applied to estimate hazard ratios (HR) and 95% confidence intervals (CI) for BMI and diabetes status as time-varying exposures alone and in combination while adjusting for known confounders.ResultsDiabetes prevalence increased from 2.6% to 5.3% and 7.5% from the first to the second and third data collection. During 27.6 ± 12.2 years of follow-up, 703 women had developed breast cancer (mean age of 66.0 ± 12.9 years) and 31 (4.4%) had been diagnosed with diabetes. A diagnosis of diabetes was not significantly associated with breast cancer incidence without (HR 1.12, 95% CI 0.77–1.64) and with BMI (HR 1.01, 95% CI 0.69–1.49) as a covariate. The respective HRs for overweight and obesity were 1.61 (95% CI 1.34–1.93) and 2.04 (95% CI 1.40–2.97).ConclusionsAmong a long-time Japanese cohort, excess body weight but not a diabetes diagnosis was significantly associated with breast cancer risk.