The emergence of data analytics has significantly expanded the research landscape within the field of information systems. Machine learning (ML) techniques have demonstrably yielded applications across diverse domains, including healthcare. Notably, the predictive capabilities of ML models have shown promise in detecting asymptomatic diseases characterized by ambiguous symptomatology, such as Dry Eye Disease (DED). Motivated by the need for further exploration in this area of diagnosis, this study seeks to identify latent patterns within inflammatory blood test results and correlate them with the presence of DED in patients. Utilizing data from 453 subjects, the study evaluated the effectiveness of logistic regression, deep learning, and Support Vector Machine (SVM) algorithms. The findings revealed SVM's potential to detect DED with an accuracy of 70% and a positive predictive value (PPV) of approximately 92%, demonstrating its effectiveness in identifying the majority of positive cases even with a limited dataset of inflammatory blood test results. This research paves the way for future investigations by opening doors to further analysis using routine inflammatory blood tests or other relevant factors, potentially leading to significant advancements in the field of non-invasive DED diagnosis.