The COVID-19 pandemic has profoundly affected lives around the world, highlighting the urgent need for faster and more accurate ways to diagnose the virus. Up to the time of writing this article, The impact of this pandemic is still felt in our lives. However, traditional methods such as symptom analysis and PCR tests are valuable. In addition, they can often be slow and complicated. This paper presents a fresh approach to detecting COVID-19 and other lung diseases using chest X-rays, by leveraging convolutional neural networks (CNNs) and enhancing image quality with histogram equalization, the aim of the proposed approach is to improve diagnostic accuracy using AI and Machine learning tools. This research utilizes a dataset of 1,823 chest X-ray images, which will be categorized into three groups: COVID-19-positive, regular, and other lung viruses. Moreover, after splitting the data into training and validation sets, the CNN model was evaluated and achieved an impressive accuracy rate of 98.45%. Consequently, these results are encouraging and suggest that our method could play a vital role in speeding up COVID-19 diagnostics, ultimately easing the burden on healthcare professionals. Looking ahead, The dataset is set to be expanded, and advanced techniques will be explored to further strengthen the model, aiming for applications across other lung diseases. Additionally, the integration of the Top-k algorithm is planned to enhance decision-making by highlighting the most accurate diagnostic outputs. This will help healthcare providers focus on the most relevant cases, improving the effectiveness of diagnostics and supporting more informed clinical decisions.