Breast cancer has become a major concern, prompting the use of computer-aided design (CAD) system networks known as Convolutional Neural Networks (CNNs) to detect it early in histopathological images. Our study approach involves training a CNN to recognize subtle signs of cancerous and non-cancerous by analyzing numerous histopathological images. We also enhance the image quality through pre-processing and apply techniques to further optimize the model's performance. We have used a publicly available Breast histopathology Image (BHI) dataset for this study which comprises of 2 different classes, namely, Invasive Ductal Carcinoma (IDC) positive and Invasive Ductal Carcinoma (IDC) negative. Experimental results show that our studied approach achieves a patch-wise classification performance of 95%. The results are competitive compared to the results of other state-of-the-art methods. Our findings suggest that this study has the potential to assist doctors in diagnosing breast cancer earlier, ultimately saving lives.