Diagnosing acute lymphoblastic leukemia (ALL) presents a considerable obstacle, often requiring invasive and costly examinations that may have adverse effects on patients. Regrettably, the problem is worsened by the fact that access to these diagnostic tools is restricted in numerous regions. Traditionally, blood microscopic examination has been the main method for screening and diagnosing ALL. However, manual performance by laboratory staff and haematologists has its limitations. To tackle this challenge, the present study explored the use of artificial intelligence (AI) techniques in analyzing blood microscopy images. The study focused on employing deep Convolutional neural networks (CNNs) to identify ALL cases, differentiate them from hematogones, and further classify the ALL subtype. The study aimed to develop a well-optimized model specifically for mobile and web applications to enhance accessibility. The modelling process included several crucial steps. Initially, a segmentation method was used, applying color thresholding in the LAB color space. This method combined K-means clustering and a mask to separate important elements and reduce image noise. Next, three lightweight CNN architectures (MobileNetV2, EfficientNetB0, and NASNet Mobile) were evaluated. After a thorough comparison, the best model was selected and fine-tuned. It is important to note that this model achieved outstanding accuracy in classification, showcasing its strength and effectiveness. Consequently, mobile and web applications based on this cutting-edge model were developed. Additionally, the implementation of a feature for locating nearby hospitals and providing directions enhances the utility of the application. This feature enables users to swiftly access further assistance from medical professionals, ensuring timely intervention and care. In practical laboratory environments, this software functioned as a reliable screening tool, effectively differentiating ALL cases from other groups with high sensitivity and specificity. Ultimately, this groundbreaking method not only showcases the capabilities of AI in medical diagnostics but also emphasizes the significance of utilizing technology to improve healthcare results.