[Live from TRC 2025] DeepRad.AI in the Spotlight: AI-Powered LDCT Ushers in a New Era of Imaging Diagnostics

In an era of rapid advancements in medical AI, the integration of cutting-edge technologies into clinical workflows has become a central focus across the healthcare industry. From May 3 to 4, 2025, the Taiwan Radiological Congress (TRC) Annual Meeting took place at the Chang Yung-Fa Foundation. Among the event highlights was DeepRad.AI’s intelligent imaging platform, DeepLung-CAC 100, which drew significant attention from attendees. At the May 3 luncheon seminar, DeepRad.AI demonstrated how its AI solution can be effectively applied in early disease screening. The session welcomed nearly 200 participants and sparked active discussions, earning strong recognition and positive feedback from the medical community.

Three Radiology Experts Share Insights and Showcase the Latest Upgrades of the DeepLung-CAC 100 Platform

The luncheon seminar, titled “The Future of Clinical Practice: AI and Physicians in Collaborative Synergy,” featured three distinguished radiology experts: Dr. Kai-Hsiung Ko from Tri-Service General Hospital, Dr. Hsun Yueh from Taipei Municipal Wan Fang Hospital, and Dr. Yung-Chieh Chen from Taipei Medical University Hospital. Each speaker shared valuable clinical insights and real-world experiences with AI integration in radiology, while also highlighting the latest advancements in the DeepLung-CAC 100 platform.

Dr. Kai-Hsiung Ko presented on “One Image, Multiple Insights: AI-Driven Integrated Interpretation of LDCT for Dual Indications,” highlighting the latest upgrades to the DeepLung-CAC 100 platform. His talk demonstrated how AI can leverage low-dose CT (LDCT) imaging to enable multi-disease screening—including lung nodule detection, coronary artery calcium (CAC) risk assessment, and bone density classification—realizing the goal of “multi-disease detection in a single scan.” This integrated approach empowers clinicians to identify high-risk individuals earlier. Dr. Ko also noted that the platform has already helped identify patients with significant coronary calcification in clinical settings, enabling timely catheter-based interventions and reducing the risk of cardiovascular events.

Dr. Ko further shared hands-on experience using DeepLung-CAC 100. The latest version supports longitudinal image comparison and nodule progression trend charts, using the internationally recognized Lung-RADS v2022 criteria to assist physicians in tracking lesion development. The CAC risk assessment interface has also been comprehensively updated to include visualized risk listings, recommended follow-up intervals, and international risk classification references—significantly enhancing its clinical usability.

AI Proven to Significantly Enhance Efficiency — Average Reading Time Reduced by 53%

Dr. Yueh Hsun Lu delivered a presentation titled “Optimizing Radiology with AI: Evidence-Based Gains in Screening Speed and Accuracy,” sharing how AI tools can alleviate the reading burden on radiologists based on real clinical workflows. He noted that CT screening volume at his hospital has been growing by over 30% annually, leading to increasing pressure on radiology departments facing both efficiency and quality challenges. The DeepLung-CAC 100 platform enables rapid image interpretation and automatically generates structured reports. Clinical evaluations have shown that it reduces reading time by approximately 53%, significantly easing the workload for physicians.

Latest Version Integrates Lung-RADS v2022 and Enables Triple Risk Assessment in a Single Scan

Dr. Yueh Hsun Lu further emphasized the platform’s accuracy and reliability, noting that the AI tool achieved a sensitivity of 96.7% for lung nodule detection, with a false-positive rate of just 0.46%. It also features automated nodule tracking and classification capabilities. Dr. Lu and Dr. Kai-Hsiung Ko presented real-world cases demonstrating the platform’s ability to detect and monitor even small nodules in real time. Based on imaging trends, the system automatically highlights changes and assigns risk levels—supporting early detection of lung cancer.

A 2025 international study published in the Journal of the American Medical Association (JAMA) identified the Coronary Artery Calcium (CAC) Score as a key indicator for determining the need for interventional treatment. During his presentation, Dr. Yung-Chieh Chen emphasized that the risk stratification approach used by the DeepLung-CAC 100 platform is aligned with international research, categorizing individuals with a CAC score ≥100 as a group requiring follow-up. The platform also provides clinical recommendations and suggested screening intervals, helping physicians clearly communicate risk to patients and facilitating effective early intervention, monitoring, and referral workflows.

The DeepLung-CAC 100 platform also includes a bone density classification module, enabling simultaneous analysis of lung nodules, coronary artery calcification, and osteoporosis risk in a single scan. Its integration of workflow optimization, automated study prioritization, interactive AI-clinician interface, and one-click report generation has been well received by users, earning strong positive feedback from the clinical community.

Reducing Radiologists’ Workload While Enhancing Efficiency and Diagnostic Accuracy

Dr. Hung-Jen Chiou, President of the Taiwan Radiological Society, also delivered opening remarks at the event. He acknowledged that the medical community has historically been cautious about adopting AI. However, after witnessing DeepRad’s cardiopulmonary AI screening platform and its clinical applications, he expressed confidence that the technology has matured and now offers tangible benefits to clinical practice. The platform enables detection of lung nodules, coronary artery calcium (CAC) risk, and bone density risk from a single scan. It not only reduces radiologists’ workload and reading pressure but also improves diagnostic accuracy. Moreover, by consolidating multiple screenings into one examination, it may help lower patients’ overall radiation exposure—a significant advancement for public health. Dr. Chiou also expressed his appreciation to the development team and academic leaders whose collaboration has enabled practical AI implementation in Taiwan’s healthcare system.

At the exhibition area, DeepRad.AI showcased the full capabilities of the DeepLung-CAC 100 platform, including its structured reporting module and live interface demonstration. The booth attracted continuous foot traffic, with many physicians showing strong interest and willingness to collaborate—particularly in functions related to workflow efficiency, diagnostic accuracy, and multi-disease screening from a single scan.

DeepRad.AI also unveiled its roadmap for future development, including its upcoming DeepBrain-Cognito platform for 4D cognitive risk assessment in aging populations, reflecting its broader commitment to preventive healthcare. Beyond developing AI tools, DeepRad emphasizes co-creation with clinicians and user-centered design—underscoring the company’s ambition and Taiwan’s robust capability to bridge AI innovation with clinical integration in medical imaging.