Software Engineer | AcuSense BioMedical Technology | Sep 2021 - Sep 2022

• Intelligence Systems for Predicting the Risk of Intradialytic Hypotension
  
  • Developed Software as a Medical Device (SaMD) product that uses Machine Learning and Deep Learning algorithms to predict the risk of hemodialysis complications. This product reduced the probability of complications by 30% in 6 months for 70+ patients. Trained classification models with three years of clinical data for clinical trials, achieving ROC-AUC and sensitivity of 93% and 90%, respectively. Deployed the service on Ubuntu using Flask, MSSQL, and Docker, with Web APIs for a single page application (SPA) and API documentation written in Markdown.
    
• Telehealth Platform for Heart Failure
  • Designed and developed a telehealth platform to remotely assist medical staff in tracking the health status of heart failure patients. Conducted clinical user interviews and created mockups using Figma to ensure the system's specifications aligned with user needs. Built Web APP components and services with connecting APIs via Angular 14, and deployed the system using NGINX and CloudFlare to serve a district hospital.
• Medical Quality Monitor Dashboard
  • Developed a dashboard that visualizes clinical dynamics, statistics, and dialysis information to aid medical quality control for nurses and the doctor director. Created the dashboard using Prometheus, Grafana, and MSSQL, with the ability to statistically analyze time-series data for vital signs, medication records, and dialysis information for 200+ hemodialysis patients.
• IOT Multiple Beds Vital Sign Monitor System 
  • Validated a BLE device and RPI IoT connectivity system to monitor vital signs for 60+ patients across 6 wards during a clinical trial. Analyzed physiological data (heart rate, blood oxygen) from 400+ hemodialysis treatments to assist in clinical trial studies.

Technical Project Manager  | AcuSense BioMedical Technology | Sep 2020 - Sep 2022

• Interviewed dialysis ward staff and medical professionals to define AIoT ward monitoring system requirements. Wrote feasibility specifications and ensured they aligned with user needs.
• Planned and designed medical AI products, including conducting clinical user interviews, developing minimum viable products, and writing conferences and journal papers. It ensured alignment with regulatory requirements, such as FDA guidelines.
• Traced a medical quality improvement project for a district hospital by analyzing retrospective data from hemodialysis wards. We utilized SPSS and Excel to perform statistical analysis and identify areas for improvement in medical quality. Presented findings and recommendations to medical staff.
• Conducted clinical validation trials for TFDA class II medical devices, including writing software validation test reports, risk assessment reports, usability analysis reports, and user manuals.