Breast cancer is primary cancer affecting women and ranks second as the leading cause of female mortality. The crucial aspect is identifying the presence of breast cancer and pinpointing the affected area. Medical imaging consistently advances, and early detection of cancer is vital in lowering cancer death rates. The enhancement procedure for mammograms involves filtering and discrete wavelet transforms. Contrast stretching is utilized to boost image contrast. Improved Breast cancer early detection and diagnosis are achieved by segmenting mammogram images. From the segmented breast region, features are retrieved. The proposed system identifies the cancer region and classifies patients as either normal or cancerous. The input mammography image is subjected to pre-processing techniques, and undesirable parts of the image are cropped off. Using morphological techniques, the tumor location is separated from the surrounding tissue and marked on the original mammography image. If the mammogram image is normal, the patient is deemed normal; otherwise, the patient is diagnosed with cancer. The Decision Tree Algorithm is utilized for categorization in this study for reasons of comparison.
Breast tissue cells can become cancerous and grow into breast cancer. It is the highly prevalent cancer in women around the world. When cells begin to multiply uncontrollably and form a tumour, breast cancer occurs. These tumors may be malignant (cancerous) or benign (non-cancerous). Malignant tumours have the potential to spread to other parts of the body and represent a threat to life if they are not identified and treated in a timely way. Breast cancer detection is the process of determining whether a person has the disease. Early breast cancer detection is crucial because it enables quick treatment and increases the likelihood of a favorable outcome. A number of screening procedures, such as mammography, ultrasound, MRI, and clinical breast exams, can find breast cancer.
Mammography is a quite popular technique to diagnose breast cancer, which involves capturing X-ray images of breast tissue. While MRI creates detailed images of the breast with a magnetic field and radio waves, ultrasound employs high-frequency sound to create breast tissue images. It is important to keep in mind that not all breast lumps or anomalies are malignant. In reality, benign breast masses predominate. But if a lump or other alteration is found, it should be examined by a medical expert to see if more testing or treatment is required.
In order to analyze the dataset, support vector machine algorithms are used to ascertain the correctness of mammography pictures. The decision tree approach is additionally employed for classification, enabling a comparison of the effectiveness of the two algorithms.
The dataset has been split into numerous classes by the support vector machine technique according to particular attributes, giving information based on the accuracy of the mammography image. The decision tree technique, contrasted with, iteratively breaks the data down into smaller groups depending on the most useful qualities, ultimately arriving at a choice at each node of the tree.
Researchers can assess the accuracy of classification findings and determine whether the method is more successful at spotting breast cancer in mammography pictures by using both the SVM and decision tree algorithms. To assess each algorithm's performance, various metrics can be assessed, including sensitivity, specificity, and positive predictive value.
The suggested system seeks to address two issues in particular. A Gaussian mixture model is employed to address the first issue, which is locating the breast cancer-affected area. The Support Vector Machine (SVM) algorithm and the Decision Tree algorithm are is employed to address second challenge, which is to categorize patients as normal or malignant.
Objectives of the proposed system are as follows:
Acquiring input mammogram images from the dataset.
Pre-process the images using the median filter.
Extracting features such as mean, variance, and Gabor features.
Segmenting the pre-processed mammogram image.
Classifying the input images using both the Decision Tree algorithm and the Support Vector Machine (SVM) algorithm.
Analyzing performance metrics, such as accuracy and precision.
Comparing the output matrix obtained using the Decision Tree algorithm and the Support Vector Machine algorithm.
II. RELATED WORKS
According to a study by Vaishnavi Patil, Shravani Burud, Goutami Pawar, Tanaya Rayajadhav, and Sunil B. Hebbale, undesirable elements and fluctuations in the scanned images, such as noise and brightness variations, can be eliminated. Through the use of an image preparation technique, they have eliminated these undesirable elements.
Vishal Deshwal and Mukta Sharma  created a grid search approach for identifying breast cancer in this journal. Without a grid search, the support vector machine model is initially assessed. Grid search is then used to test the support vector machine model. Finally, a comparison study was conducted, and a new model was developed in light of the results. Before fitting it for prediction, the new model is based on a data grid search, which enhances the results.
An algorithm for tumor detection has been proposed by Y. Irenaeus Anna Rejani et al .
A breast CAD technique based on feature fusion with convolutional neural network (CNN) deep features was studied by Zhiqiong Wang, Mo Li, Huaxia Wang, Hanyu Jiang, Yudong Yao, Hao Zhang, and Junchang Xin . First, a mass identification technique based on unsupervised extreme learning machine (ELM) clustering and CNN deep features were suggested. Furthermore, a feature set was created by combining deep features, morphological characteristics, texture features, and density features. The ELM classifier is then created to differentiate between benign and malignant breast tumors.
An efficient AdaBoost algorithm for early breast cancer diagnosis and detection was proposed by Jing Zheng, Denan Lin, Zhongjun Gao, Shuang Wang, Mingjie He, and Jipeng Fan .
A method with two primary parts has been proposed by M. R. Al-Hadidi, A. Alarabeyyat, and M. Alhanahnah . In the first phase, image processing techniques are used to prepare the mammography pictures for feature and pattern extraction. The collected features are used as input for the Back Propagation Neural Network (BPNN) model and the Logistic Regression (LR) model, two different supervised learning models, in the following part.
By analyzing hostile ductal carcinoma tissue zones in whole-slide images, Saad Awadh Alanazi, M. M. Kamruzzaman, Md Nazirul Islam Sarker, Madallah Alruwaili, Yousef Alhwaiti, Nasser Alshammari, and Muhammad Hameed Siddiqi  proposed a technique to improve the automatic identification of breast cancer.
III. THE PROPOSEDMODEL
The method of detecting breast cancer involves several key steps. First, an input image is collected and pre-processed to prepare it for image processing and segmentation. Next, the image is segmented to isolate the region of interest Features are taken from the image following segmentation, which are then used to train a classifier model. Once the model is trained, It can be utilized to foresee the status of new images. Overall, the breast cancer detection process involves an assortment of imaging methods, feature extraction, and machine learning algorithms to accurately identify breast cancer and inform treatment decisions.
IV. CLASSIFICATION OF CANCER
Support vector machine algorithm and decision tree algorithm are two methods by which the classification of cancer can be accomplished.
A. Support Vector Machine Algorithm
The support vector machine (SVM) algorithm is a powerful machine-learning technique that can be used to aid in the detection and classification of breast cancer.
In the context of breast cancer detection, SVMs can be trained on large datasets of breast cancer cases, along with various clinical and demographic variables such as patient age, tumor size, and histological features. The algorithm learns to classify cases as either benign or malignant based on patterns in these variables.
Once the SVM is trained, it can be used to analyze new breast cancer cases and make predictions about the likelihood of malignancy. The SVM can also be used to identify important features that contribute to the classification, which can aid in the development of new diagnostic tools and treatments.
B. Decision Tree Algorithm
The decision tree algorithm is another machine-learning technique that can be used for breast cancer detection. The algorithm is trained on a dataset of breast cancer cases and clinical variables and learns to classify cases as either benign or malignant based on patterns in these variables. Decision tree algorithms can also identify important features that contribute to the classification and can be easily interpreted by healthcare professionals, aiding in decision-making, and improving
The proposed system utilizes two classification algorithms, a decision tree and a support vector machine (SVM), for mammogram images analyzation. The system has an accuracy of 68.2% and 82.02% and a precision of 71.69% and 81.26% for dataset1 when utilizing the decision tree and SVM algorithms, respectively. Similarly, for dataset 2, the system has an accuracy of 66.66% and 91.66% and a precision of 68.61% and 93.33% when using the decision tree and SVM algorithms, respectively.
According to these findings, the SVM algorithm outperforms the decision tree algorithm in terms of accuracy and precision.
 Vaishnavi Patil, Shravani Burud, Goutami Pawar,Tanaya Rayajadhav, & Sunil B. Hebbale. (2020). Breast Cancer Detection using MATLAB Functions. Advancement in Image Processing and Pattern Recognition, 3(2), 1-6.
 Vishal Deshwal and Mukta Sharma. Breast Cancer Detection using SVM Classifier with Grid Search Technique. International Journal of Computer Applications 178(31):18-23, July 2019.
 Y.Ireaneus Anna Rejani et al /International Journal on Computer Science and Engineering Vol.1(3), 2009, 127-130.
 Zhiqiong Wang, Mo Li, Huaxia Wang, Hanyu Jiang, Yudong Yao, Hao Zhang, And Junchang Xin “Breast Cancer Detection using Extreme learning Machine-Based on Feature Fusion with CNN Deep Features”, IEEE Access, August 14, 2019.
 Jing Zheng, Denan Lin, Zhongjun Gao, Shuang Wang, Mingjie He, And Jipeng Fan “Deep Learning Assisted Efficient AdaBoost Algorithm for Breast Cancer Detection and Early Diagnosis”, IEEE Access, June 4, 2020.
 M. R. Al-Hadidi, A. Alarabeyyat and M. Alhanahnah, \"Breast Cancer Detection Using K-Nearest Neighbor Machine Learning Algorithm,\" 2016 9th International Conference on Developments in Systems Engineering (DeSE), Liverpool, 2016, P. 35-39.
 Saad Awadh Alanazi, M. M. Kamruzzaman , Md Nazirul Islam Sarker , Madallah Alruwaili , Yousef Alhwaiti , Nasser Alshammari , and Muhammad Hameed Siddiqi “Boosting Breast Cancer Detection Using Convolutional Neural Network,” Journal of Healthcare Engineering Volume 2021, 5 April 2021.