<p>Breast cancer is a prevalent disease among women that begins in the breast tissue, typically originating in the ducts or lobules. Identifying breast cancer at an early stage was difficult because distinguishing between non-cancerous and cancerous tissues is a challenging task. The prevailing techniques have difficulty in accurately differentiating malignant tissues from benign tissues due to high visual similarity and complex patterns present in the histopathological images. Most of the methods often failed to capture the complex structure and subtle differences necessary for accurate detection. To overcome these limitations, a Parallel Convolutional Neural Network with Jaya One-to-One-Based Optimizer (PCNN_JOOBO) is designed for detecting breast cancer using histopathological images. Here, the JOOBO integrates Jaya Optimization (Jaya) and One-to-One-Based Optimizer (OOBO). Initially, breast histopathological images are obtained and filtered using a median filter. Subsequently, the blood cell regions are segmented using the Parallel Reverse Attention Network (PraNet), which is optimized using the newly developed JOOBO. Next, shape features and Convolutional Neural Network (CNN) features are derived from the segmented area. At last, the detection of breast cancer is performed by a Parallel Convolutional Neural Network (PCNN), which is trained using the JOOBO. A new PCNN_JOOBO obtained an accuracy of 95.877%, a True Positive Rate (TPR) of 96.556%, and a True Negative Rate (TNR) of 95.168% at a 90% learning set. The developed PCNN_JOOBO is capable of identifying changes in tissue architecture and recognizing fine-grained patterns, which helps to accurately detect breast cancer at an early stage. The early detection of breast cancer helps to prevent the disease from progressing and spreading to other organs and helps to reduce the mortality rate of patients. Also, this method can generalize well across varied patient data.</p>