As a member of the Carnegie Mellon DIG Lab, I participated in the PHORA study to explore the effects of explainable AI on clinical patients. I helped design and build a patient-facing dashboard that communicates individualized pulmonary hypertension risk using interpretable machine-learning models. My primary contributions were on the backend, where I developed the data architecture, engineered longitudinal features, built preprocessing and ETL pipelines, integrated Firebase-based storage, and implemented the XGBoost survival model with SHAP-value explainability. I also collaborated with clinicians and patients in iterative user-centered design sessions. Through this study, I examined how different explanation formats influence patient trust, comprehension, and clinical decision-making.