With the advancement of technology, research focus in cancer diagnosis has shifted towards digitizing histopathology images. Particularly in the assessment process for breast cancer diagnosis, challenges such as color inconsistencies, nuclei segmentation, mitosis detection, identification of tubular formations, and delineation of cancerous regions have emerged. As a solution within this project, a new methodology has been proposed for automated breast cancer diagnosis and grading. The suggested methodology comprises seven stages: 1) Color Normalization; 2) Nuclei Segmentation; 3) Mitosis Enumeration and Detection; 4) Nuclei Atypia Evaluation; 5) Cancerous Region Detection; 6) Tubular Segmentation; and 7) Datasets. These steps aim to offer an alternative to pathologists' subjective assessments, encouraging their diagnoses to be supported by quantitative values. The objective of this study is to develop an original methodology that could assist in the creation of an automatic or semi-automatic breast cancer detection and grading system. Contributions made to the proposed seven-stage methodology are outlined as follows: 1) Application of a color normalization technique to minimize staining inconsistencies in histopathology images. 2) Proposal of two unique deep learning architectures (Lightweight U-Net and ComSegNet) for nuclei segmentation. 3) Development of an original methodology incorporating deep learning and fuzzy logic for mitosis enumeration and detection. 4) Implementation of a nuclear atypia evaluation approach summarizing morphological tissue heterogeneity in histopathological images. 5) Utilization of deep learning methods from the literature for cancerous region detection. 6) Proposal of a unique attention-based U-Net model for segmentation of tubular formations. 7) Creation of two original HE stained digital histopathology datasets containing nuclei and mitosis data. This interdisciplinary project, conducted in collaboration with Ankara University's Department of Pathology, aimed to generate HE stained digital histopathology datasets on breast tissue. Its goal is to contribute to clinical applications by presenting a more accurate, objective, and reliable method for breast cancer diagnosis.