New Taipei Jen-Kang Hospital Responsible AI Center

About Us - Welcome to Responsible AI at Jen-Kang

Leading the Future of Smart Healthcare, Committed to a People-Centric Approach.
Established in 2024 with support from the Ministry of Health and Welfare’s (MOHW) “Next-Generation Digital Healthcare Platform” program, the Responsible AI Center at New Taipei Jen-Kang Hospital is dedicated to promoting the safe, effective, and ethical application of Artificial Intelligence (AI) in healthcare. Our mission is to continuously enhance medical quality and patient safety through responsible AI innovation, ensuring that technological development always remains people-centric.
Core Objectives:

  • Ensure information security and patient privacy.
  •  Build trust in AI applications among physicians and patients.
  • Ensure continuous monitoring of AI products to guarantee the reliability of diagnosis and decision support.

Our Guiding Principles

We adhere to international trends and the spirit of Taiwan’s “Artificial Intelligence Basic Law” (Draft) to build a trustworthy AI framework.
Seven Core AI Values:

  • Human Autonomy: Respecting human decision-making and values.
  • Accountability: Clearly defining responsibilities.
  • Transparency & Explainability: Ensuring AI decision-making processes are understandable.
  • Safety: Ensuring the robustness and security of systems.
  • Fairness & Non-discrimination: Avoiding bias and discrimination.
  • Sustainability & Well-being: Promoting social equity and environmental sustainability.
  • Privacy Protection & Data Governance: Protecting personal data and promoting proper data use.


Nine Transparency Principles:
We integrate the “Nine Transparency Principles” with “Explainability Analysis” to ensure AI decision-making processes are transparent, understandable, and trustworthy for medical teams. This covers aspects such as AI purpose, development details, performance measurement, and ongoing maintenance.

Figure 1: The Seven Core Principles of AI Development

Our Responsible AI Framework - Our Approach

We implement the philosophy of Responsible AI in our daily operations through three core strategies:

  • Information Security Maintenance & Privacy Protection:
    Strict adherence to regulations such as the “Medical Institution Information Security Management Guidelines,” “Cybersecurity Management Act,” and “Personal Data Protection Act,” with regular maintenance and audits to protect patient data and system security.
  • Standardized Management Processes for AI Application Safety:
    Establishment of standardized processes covering AI product application, review, and continuous monitoring, providing a reliable safeguard for smart healthcare applications.
  • AI Lifecycle Management:
    Full lifecycle monitoring for introduced AI technology, including:
  • Monitoring Standards: Defining quantifiable metrics based on transparency principles (e.g., accuracy, bias rate).
  • Anomaly Response: Activating a suspension mechanism if metrics fall below thresholds, with anomalies reported to the committee for review.
  • Correction & Re-validation: Requiring the applicant unit to re-validate and re-calibrate the model after suspension.
    Smart Healthcare Committee:
    Composed of the hospital Superintendent, Vice Superintendent of IT, department directors, and external consultants, this committee is responsible for strategic review of AI technology introduction, implementation oversight, standards establishment, and data security, ensuring AI applications align with the hospital’s development direction and clinical needs.

AI in Action: Smart Ward Technology

Our hospital has adopted Smart Ward AI equipment provided by Wistron Medical Technology Co., Ltd., such as BestShape VS (Physiological Sensor) and BestShape Care (Motion Sensor). These devices utilize Machine Learning and Deep Learning algorithms and hold multiple international and Taiwan TFDA certifications (e.g., TFDA Permit No. 衛部醫器製字第007955號), demonstrating their safety and effectiveness in enhancing care quality and ward management efficiency.

BestShape VS (Physiological Sensor)

Contact Us

Advancing Smart Healthcare Together.
For inquiries about our center or if you are interested in collaborating on AI medical devices/software, please contact:

For Vendors

Vendors interested in collaboration are requested to complete the “Smart Healthcare Medical Device Landing Application Form.” This form covers basic information, safety evaluation, performance evaluation, the Nine Transparency Principles, personal data privacy, and validation assessment. Applications will be processed by our center, reviewed by the IT department and relevant units, and then submitted to the hospital-wide “Smart Healthcare Committee” for a final decision.
(Key points of management measures – for vendor reference):

  • SaMD Definition: Following TFDA guidelines.
  • Risk Assessment: Based on TFDA/FDA risk classification. Low-risk SaMD requires observational validation; high-risk (requiring interventional validation) is not currently adopted by our hospital.
  • TFDA Certification: Generally required. For non-certified products, low to medium-risk items may be trialed after committee evaluation and Superintendent approval.
  • IT & Medical Evaluation: To ensure information security, privacy, system compatibility, and clinical safety and effectiveness.
  • Continuous Monitoring: Deployed AI solutions will undergo periodic monitoring and evaluation.
Vendor Application Form