Cancer remains a major global health challenge, with significant disparities in access to advanced diagnostic and prognostic technologies, especially in resource-constrained settings. Existing medical treatments and devices for cancer diagnosis are often prohibitively expensive, limiting their reach and impact. Pathologists' scarcity exacerbates cancer diagnosis accuracy, elevating mortality risks. To address these critical issues, this study presents OVision - a low cost, deep learning-powered framework developed to assist in histopathological diagnosis. The key objective is to leverage the portable, low-power computing Raspberry Pi. By designing standalone devices that eliminate the need for internet connectivity and high-end infrastructure, we can dramatically reduce costs while maintaining accuracy. As a proof of concept, the study demonstrated the viability of this framework through a compact, self-contained device capable of accurately detecting ovarian cancer subtypes with 95% accuracy, on par with traditional methods, while costing a small fraction of the price. This portable, off-grid solution has immense potential to improve access to precision cancer diagnostics, especially in underserved regions of the world that lack the resources to deploy expensive, infrastructure-heavy medical technologies. In addition, by classifying each tile, the tool can provide percentages of each histologic subtype detected within the slide. This capability enhances the diagnostic precision, offering a detailed overview of the heterogeneity within each tissue sample, helps in understanding the complexity of histologic subtypes and tailoring personalized treatment plans. In conclusion, this work proposes a transformative model for developing affordable, accessible medical devices that can bring advanced healthcare benefits to all, laying the foundation for a more equitable, inclusive future of precision medicine.