Veditha Yechuri

Led hospital bed utilization optimization using time-series forecasting and XGBoost, improving occupancy efficiency by 18% and reducing emergency wait times by 22%. Designed ML pipelines on Azure ML integrating admission, discharge, and transfer data, enabling hospital operations teams to make data-driven capacity planning decisions. Deployed predictive dashboards in Power BI for real-time patient-flow monitoring and built claims denial prediction models reducing denial rates by 25%. Automated pre-submission claim risk classification, decreasing manual rework by 30% and increasing RCM margins. Developed NLP-based clinical document classification pipelines using spaCy and scikit-learn, reducing indexing effort by 60%. Supported batch deployment of document classification services using Flask and Python for scalable digitization and audits. Built pharmacy demand forecasting models using ARIMA and Prophet, reducing stock-outs by 20% and lowering excess inventory costs by 15%. Created anal

Built machine learning models to classify diagnostic lab reports using OCR and TF-IDF features, reducing manual tagging effort by 55% and improving report retrieval efficiency. Processed scanned medical reports using Tesseract OCR and Python pipelines, enabling structured text extraction that supported downstream classification and compliance audits. Trained and evaluated Logistic Regression and SVM models on labeled healthcare datasets, achieving reliable multi-class accuracy. Analyzed historical OPD appointment data to identify no-show patterns, supporting predictive modeling that reduced missed appointments by up to 15% through targeted interventions. Generated analytical reports and model performance metrics to aid stakeholders in interpreting predictions and implementing reminder actions.

Led AI-driven prior authorization automation using ClinicalBERT and FHIR integration, reducing manual reviews by 40%, accelerating turnaround times by 30%, and saving $4.5M annually. Architected scalable NLP pipelines on AWS SageMaker with MLflow and SHAP explainability to ensure audit-ready clinical decisions while meeting payer compliance and regulatory requirements. Owned end-to-end deployment of healthcare ML models, collaborating with clinicians and SMEs to align predictions with medical policies, improving approval accuracy and provider satisfaction.