Improved Machine Learning-Based Predictive Models for Breast Cancer Diagnosis.

The study presents a method for iterative parameter tuning of tree ensemble-based models using Bayesian hyperparameter tuning for states prediction, using breast cancer as an example. The proposed method utilizes three different datasets, including the Wisconsin Diagnostic Breast Cancer (WDBC) dataset, the Surveillance, Epidemiology, and End Results (SEER) breast cancer dataset, and the Breast Cancer Coimbra dataset (BCCD), and implements tree ensemble-based models, specifically AdaBoost, Gentle-Boost, LogitBoost, Bag, and RUSBoost, for breast cancer prediction. Bayesian optimization was used to tune the hyperparameters of the models iteratively, and the performance of the models was evaluated using several metrics, including accuracy, precision, recall, and f1-score. Our results show that the proposed method significantly improves the performance of tree ensemble-based models, resulting in higher accuracy, precision, recall, and f1-score. Compared to other state-of-the-art models, the proposed method is more efficient. It achieved perfect scores of 100% for Accuracy, Precision, Recall, and F1-Score on the WDBC dataset. On the SEER BC dataset, the method achieved an accuracy of 95.9%, a precision of 97.6%, a recall of 94.2%, and an F1-Score of 95.9%. For the BCCD dataset, the method achieved an accuracy of 94.7%, a precision of 90%, a recall of 100%, and an F1-Score of 94.7%. The outcomes of this study have important implications for medical professionals, as early detection of breast cancer can significantly increase the chances of survival. Overall, this study provides a valuable contribution to the field of breast cancer prediction using machine learning.

Finding breast cancer early and accurately is very important for helping patients live longer. We use the Wisconsin Diagnostic Breast Cancer (WDBC) dataset to look at two popular machine learning classifiers: Support Vector Machines (SVM) and k-nearest neighbors (KNN). Our work combines theory with practice by including things like data preprocessing exploratory feature evaluation, model configuration, hyperparameter, and validation through cross-validation techniques. We use common evaluation metrics like accuracy, precision, recall, F1-score, and Area Under the Curve (AUC) to judge performance. When we did experiments with both a 70:30 train-test split and 10-fold cross-validation, we found that SVM did a little better (~96.3% accuracy) than compared to KNN, which has an accuracy of about 96.5% and AUC values of about 0.96 and 0.968, respectively. SVM is better at dealing with high-dimensional data and making maximum-margin decision boundaries, but KNN is still appealing because it is simple and works well on datasets of moderate size. Even though both methods have their pros and cons, they also have some problems that are talked about, as well as ways they could be improved and how they could be used in the clinic. This paper evaluates and compares the theoretical and practical performance of SVM and KNN using the WDBC dataset. We include preprocessing, careful evaluation metrics, and hyperparameter optimization to provide an empirical comparison of both models. The study also discusses important trade-offs and real-world factors for using these algorithms in medical screening contexts.

Breast cancer is a disease that causes excessive fear in women around the world. The number of high death rates by breast cancer can be reduced by early detection. This can make breast cancer a disease that is easy to cure. A collection of datasets about breast cancer is used in the process of early detection. Early detection is carried out to analyze the state of the early stages of breast cancer patients. This research paper proposes machine learning methods, namely Generalized Linear Model, Logistic Regression, and Gradient Boosted Decision Tree to enhance the classification performance of Wisconsin Diagnostic Breast Cancer Data. The diagnosis results in two classes of cancer decisions which are malignant and benign by looking at evaluating the accuracy of the data classification test. The result shows that the Generalized Linear Model achieves the accuracy of 99.4%, which is higher than the accuracies of the previous studies for classifying the Wisconsin Diagnostic Breast Cancer dataset.

The purpose of the proposed study was to provide a solution to the Wisconsin Diagnostic Breast Cancer (WDBC) classification problem. The WDBC dataset, provided by the University of Wisconsin Hospital, was derived from a group of images using Fine Needle Aspiration (biopsies) of the breast. An ensemble of Support Vector Machines (SVM's) was employed in this study. Support vectors with linear, polynomial and RBF kernel functions were trained using a fraction of the WDBC dataset as a training set. The five top performing models were recruited into the ensemble. The classification was then carried out using the majority opinion of the ensemble. The SVM ensemble successfully classified more than 99 percent of the testing data and in the process yielded 100 percent benign tumor prediction accuracy. Keywords: Support Vector Machines; Breast Cancer; Machine Learning, Ensemble, Classifiers

According to the WHO, the cases of Breast Cancer are now a global issue as the cases are increasing day by day globally. Hence the early detection of malignant tumor is utmost important to prevent patient’s early demise due to patient’s ignorance. If patient comes to doctor and their application detects the tumor is non cancerous or it is just a benign tumor but actually it is malignant tumor then early detection alone can not play such an important role for cancer prevention. So, It is equally important to use best Machine Learning model, which surely detects malignant as malignant tumor and benign as benign tumor. Such accurate model is also highly required along with early detection of Breast Cancer disease. Our work is dedicated in this direction that our suggested model predicts malignant tumor as malignant and benign as benign tumor with high accuracy than existing model is providing. This paper suggests an architecture which produces high accuracy of the model when applied on Wisconsin Diagnostic Breast Cancer (WDBC) dataset obtained by digitized fine needle aspirate images. Our suggested model is the modified version of SVM. It is providing 63% accuracy when data points are not scaled, 98% when date is correctly scaled and 99% when the data points are correctly scaled with appropriate regularization and K fold validation techniques are applied on Support Vector Machine (SVM) instead of simply using default SVM with default parameters. Evidently, our suggested SVM provides 99% accuracy for detecting tumor as benign or malignant over WDBC dataset obtained by digitized fine needle aspirate images.

Currently, as governments across the world attempt to halt the spread of COVID-19, many of their efforts are being hampered by shortages of crucial testing kits. Even in many developed countries, the health system has come to the point of standstill due to the increasing demand for intensive care units. One study reveals that 40,000 women die in a year from breast cancer, which is one woman every 13 minutes dying from this disease every day. Breast cancer remains the number one form of cancer that women are diagnosed with around the world. Any hospital in the world will have a multidisciplinary team of experts including radiologists, oncologists, surgeons, and other respiratory physicians and consultants. This multidisciplinary team uses a wide variety of data taken from CT scans, x-rays, PET scans, biopsies, clinical examination, imaging mammography, ultrasound (sonography), pathology biopsy, and fine-needle aspiration to identify whether a patient has cancer. Individuals with cancer, particularly those who are receiving systemic anticancer treatments, have been postulated to be at increased risk of mortality from COVID-19. Therefore, it is the need of an hour to implement an early detection of cancer that will prevent patients from COVID-19 and due treatment will be provided. Breast cancer that is detected early is far easier to cure, but at the same time it is difficult to detect. An accurate breast cancer staging is an essential task performed by pathologists (experts who examine the laboratory samples) to inform clinical management for the appropriate treatment. For instance, evaluation of the extent of cancer spread by histopathologic analysis or microscopic examination is an important part of breast cancer staging, and the process involves the decision-making of human behavior with time-consuming and costly sample examination and a period of more than one month to have breast cancer type, stage, etc., to decide the appropriate therapy to patients by the time it can affect the results of therapeutic treatments and survival of patients. The application of artificial intelligence (AI) machine learning technology with deep learning algorithms to whole-slide pathology images can potentially improve the diagnostic accuracy of breast cancer at a very early stage. I have developed a simple machine-learning model using the Python programming language and neural network, inspired by the structure of the human brain, with 12 attributes, out of which 10 real-valued features from each cell nucleus obtained from the Wisconsin diagnostic breast cancer dataset available in public domain for the education capacity that explains about the stage of breast cancer M (Malignant) and B (Benign). This model predicts the presence of breast cancer with an accuracy of 98.8% at a very early stage. The continuous development of technology in the medical field will save countless lives and the overall quality of human life will continue to improve over time. Citation Format: Subash Kumar. Early detection of breast cancer during the Covid-19 pandemic using artificial intelligence neural networks [abstract]. In: Proceedings of the AACR Virtual Meeting: COVID-19 and Cancer; 2020 Jul 20-22. Philadelphia (PA): AACR; Clin Cancer Res 2020;26(18_Suppl):Abstract nr PO-003.

<p>Breast cancer is the most frequent cancer in women and the second-leading cause of cancer-related deaths globally. The main problems in managing breast cancer are high heterogeneity and the formation of therapeutic resistance. White blood cells, omics and large Wisconsin diagnostic breast cancer datasets present the three-decade genomic revolution and advance the understanding of cellular function. The precision of cancer diagnosis has also increased over the past decades. High throughput sequencing, screening, and artificial intelligence technologies have significantly improved and increased the methodologies used for diagnosis, prognosis, and therapy. This paper follows several phases of breast cancer, studies datasets and evaluate many algorithms of machine learning (ML) used for analysis and feature selection i.e. k-means, similarity correlation, genetic algorithm, and principal component analysis, have been used to recognize the subset of proteins with the highest significance for breast cancer prediction by using different biomarkers. The best correlation, as determined by Pearson correlation, between copy number and protein is 0.014, and the accuracy achieved by the genetic algorithm is 93.5% using multi-omics datasets.</p>

Breast cancer has been one of the leading causes of death among women in the world. Early detection of this disease can save patient’s lives and reduce mortality. Due to the large number of features involved in the diagnosis of this disease, the breast cancer diagnosis process can be time consuming. To reduce cost and time and improving accuracy of breast cancer diagnosis, this paper propose a feature selection algorithm based on particle swarm optimization (PSO) combined with machine learning methods for selection the most effective features for breast cancer diagnosis among all features. In order to evaluate the efficiency of the proposed feature selection method, it was tested on three most common breast cancer datasets available in the University of California, Irvine (UCI) repository named: Coimbra dataset (CD), Wisconsin Diagnostic Breast Cancer dataset (WDBC) and Wisconsin Prognostic Breast Cancer dataset (WPBC). In the Coimbra dataset with all its 9 features and without PSO feature selection algorithm the highest obtained accuracy was 87% by Support Vector Machine method, while with PSO feature selection algorithm the accuracy reached to 91% and the number of features was reduced from 9 to 4. In the WDBC dataset with all its 30 features and without PSO feature selection algorithm the highest obtained accuracy was 99% by Random Forest method, while with PSO feature selection algorithm the accuracy reached to 100% and the number of features was reduced from 30 to 19. In the WPBC dataset with all its 33 features and without PSO feature selection algorithm the highest obtained accuracy was 94% by Support Vector Machine method, while with PSO feature selection algorithm the accuracy reached to 96% and the number of features was reduced from 33 to 17. The results of this paper indicated that the proposed feature selection algorithm based on PSO algorithm can improve the accuracy of breast cancer diagnosis. While it has selected fewer and more effective features than the total number of features in the original datasets.

Does a young woman living with a smoker or taking a job working in a smoky bar have a greater chance of developing breast cancer? Some scientists believe that such situations can indeed raise a woman’s risk of developing breast cancer before the age of 50. Because epidemiological and toxicological studies show that women’s breast tissue may be especially sensitive to exposure to carcinogens prior to first pregnancy, these researchers contend that public education should be directed at alerting adolescents and young women to the potential risk. However, not everyone in the international public health community agrees that the evidence to date supports a link between passive smoking and breast cancer, and some say that women are being alarmed unnecessarily. This disagreement has sparked debate that is sometimes heated. The stakes are high because breast cancer is the most common cancer in women in industrialized countries, according to the WHO. It is the leading cancer killer of nonsmoking women, and second only to lung cancer deaths among women who smoke. Among the researchers interviewed for this article who disagree that there is enough evidence to link secondhand smoke (SHS) with breast cancer, the majority call the evidence to date “suggestive but not sufficient,” as the Surgeon General’s 2006 report, The Health Consequences of Involuntary Exposure to Tobacco Smoke, put it. That characterization is based largely on the fact that the research considered when the Surgeon General’s report was being amassed did not clearly link even active smoking to breast cancer. Researchers in this camp do, however, stress that ongoing campaigns to prohibit smoking in public will protect the whole of society against the wide variety of ills proven to be caused by SHS. These include lung cancer, cardiovascular disease, and sudden infant death syndrome, among others. A smaller group contends that the question of whether or not SHS causes breast cancer is a political issue with the potential to compromise the scientific process. “A premature decision about causality could jeopardize the credibility of the entire review process and all of the other, established effects of secondhand smoke,” says Michael Thun, national vice president of epidemiology and surveillance research for the American Cancer Society. Adds Valerie Beral, director of the University of Oxford Cancer Research UK Epidemiology Unit, “To prematurely come to conclusions about the causation when there is a big division in the scientific community . . . is bad science.” Thun debated the subject in a series of public forums held in conjunction with scientific meetings. Taking the opposing view was Kenneth C. Johnson, a research scientist with the Public Health Agency of Canada, who was one of the first scientists to discern a potential link. During the debates, Johnson pointed out there are about the same number of studies linking breast cancer to passive smoking as there were linking lung cancer to SHS in 1986, when the Surgeon General concluded that passive smoking caused lung cancer. Johnson also says that more of the breast cancer studies are statistically significant, and that the estimated risk for breast cancer is higher.

Cohn et al. (2008) suggested that birth cohort trends in breast cancer rates for women under 50 years of age are consistent with declining use of DDT (dichloro-diphenyltrichloroethane) after 1959. They cited Weiss (2007) in claiming that increased detection and treatment of in situ breast cancer must be considered when interpreting recent trends in breast cancer mortality rates in young women. The remarks of Weiss (2007) relate to women 40–49 years of age, and earlier detection and improved treatment of breast cancer has had a marked impact on breast cancer mortality rates in these women since 1990 (Berry et al. 2005; Chu et al. 1996). The birth cohort trends relevant to examining the possible impact of childhood DDT exposure on U.S. breast cancer rates, however, were firmly established well before 1990 in women < 40 years of age (Tarone 2007). Cohn et al. (2007) reported a large increase in breast cancer risk estimates for p,p′-DDT [1,1,1-trichloro-2,2-bis(p-chlorophenyl)ethane] exposure with successive birth cohorts after 1930. Their reported odds ratio estimates by period of birth for the highest tertile of p,p′-DDT exposure were 0.6 for women born in 1931 or earlier (i.e., ≥ 14 years of age in 1945), 3.9 for women born in 1932–1937 (i.e., 8–13 years of age in 1945), 9.6 for women born in 1938–1941 (i.e., 4–7 years of age in 1945), and 11.5 for women born in 1942 or later (i.e., 0.25), and there was a marginally significant decrease in rates for women 35–39 years of age (p = 0.04). Thus the trends in breast cancer mortality rates among women born in 1930–1945 are not consistent with the sharply increasing trend in odds ratios for childhood DDT exposure by birth period reported by Cohn et al. (2007). The most recent mortality rate contributing to the reported regression analyses (corresponding to women in the 35- to 39-year age group born in 1945) was for 1983, well before improvements in detection and treatment would have had any impact on breast cancer mortality rates. Women born after 1945 were exposed to DDT for each of the first 13 years of life (and all years thereafter). In addition, DDT exposure increased from 1945 through 1959, when DDT use peaked (with dietary exposure peaking in 1965) (Wolff et al. 2005). If DDT exposure early in life markedly increases breast cancer risk, then some evidence of the increasing DDT use after 1945 might be expected in breast cancer mortality rate trends for young women born from 1946 through 1959 (Tarone 2007). Breast cancer mortality rates decreased significantly among women 20–24 years of age (p = 0.009) and 25–29 years of age (p = 0.0002) born between 1946 and 1959 (SEER 2006; Tarone 2007). The most recent rate contributing to these regression analyses was for 1987 (corresponding to women in the 25-to 29-year age group born in 1959). Breast cancer mortality rates decreased even more markedly (p < 0.0001) for women in the 30- to 34-year and 35- to 39-year age groups born from 1946 through 1959; some of the recent rates in these latter age groups were almost certainly affected by improved breast cancer detection and treatment, although decreasing trends were apparent in both age groups for rates well before 1990 (Tarone 2007). Thus, U.S. breast cancer mortality rates in women between the ages of 20 and 39 who were born between 1930 and 1959 show no evidence of an increase in breast cancer risk associated with their marked increase in DDT exposure during childhood. The observed birth cohort trends in breast cancer rates do not refute a possible association between childhood DDT exposure and breast cancer risk, and contrary to the implication of Cohn et al. (2008), no such claim was made in my earlier letter (Tarone 2008). The regression analyses reported above suffer the weaknesses of all ecologic analyses, and in fact, the decreasing birth cohort risk of breast cancer in baby boomers has been observed in spite of trends in established risk factors (e.g., parity, age at first birth, and oral contraceptive use) that would predict increasing breast cancer rates among U.S. women born after 1945. If, as suggested by Cohn et al. (2007), the public health significance of DDT exposure early in life is large, then this would provide additional evidence that the factor or factors responsible for the paradoxical decrease in birth cohort risk of breast cancer observed among U.S. baby boomers must have a very powerful impact on breast cancer etiology, large enough to turn an expected increasing trend in breast cancer rates among baby boomers into a decreasing trend.

Due to increased number of breast cancer cases, there is rise in concern about the disease. So, in today’s era, machine learning classification can play a vital role in classifying the degree of malignancy of the cancer. This paper uses four different classifiers such as KNN, logistic regression, naive Bayes and decision tree to classify the cancer into two classes of benign and malignant. The proposed model is simulated over Wisconsin diagnostic breast cancer dataset and evaluated by accuracy metric. The model classifies the disease based on the most accurate classifier. The KNN model shows significantly better result than the other classifier.

Effects of radiotherapy and of differences in the extent of surgery for early breast cancer on local recurrence and 15-year survival: an overview of the randomised trials

Trends in breast cancer incidence, survival, and mortality

In the presented study, the authors have compared the performances of various machine learning and deep learning algorithms on two separate datasets: the Wisconsin Diagnostic Breast Cancer dataset and the Wisconsin Breast Cancer Dataset. We evaluated the extent of their ability to correctly classify the samples as “malignant” or “benign”. The respective performances of these algorithms were judged on the grounds of various evaluation metrics such as accuracy, precision, recall, specificity, and the AUC-ROC curve for both datasets. From the experimental results, we concluded that the deep learning approaches have provided better results on evaluation grounds than the machine learning ones. In quantitative terms, ANN performed most consistently among all the considered approaches for either dataset with an accuracy of 100% and 98% respectively.KeywordsConvolutional neural networksClassificationMachine learning algorithmsDecision boundaryDeep learningArtificial neural networks

Breast cancer is the most common cancer for females worldwide. The literature presents a comparison of four machine learning algorithms: Classification Tree, Pruned Classification Tree, Bagging, and Random Forests on the Wisconsin Diagnostic Breast Cancer dataset by measuring the test sensitivity, specificity, precision, negative predictive value, and misclassification error rate. The result shows that the performance of pruned classification tree is better than that of a normal classification tree; aggregated tree models (Bagging and Random Forests) are is better than only use one single decision tree to predict. The best tree model is Random Forests which has the lowest test misclassification error rate of 2.92%.