SGH's iScreen AI Tool Cuts CT Scan Reporting Time to Six Minutes

Singapore General Hospital (SGH) has launched a groundbreaking AI tool named iScreen, designed to detect intracranial bleeding from CT scans within approximately six minutes. This development significantly reduces the typical reporting time, which can range from 30 minutes to an hour. The rapid detection capability of iScreen is critical, as timely intervention can be the difference between life and death for patients experiencing brain hemorrhages.

The iScreen tool, developed in collaboration with local startup Iota MedTech, leverages advanced algorithms to analyze CT scans, identifying white spots indicative of bleeding. Upon detection, the system promptly alerts neuroradiologists and attending physicians via SMS, enabling quicker review and prioritization of urgent cases. This innovation is expected to enhance patient outcomes by minimizing delays in diagnosis and treatment, particularly in emergency situations where acute cases may require immediate surgical intervention.

SGH's data indicates that the sensitivity of iScreen is 93.1%, with a noted decrease in the proportion of bleeding cases taking over an hour to report—from 14.6% to 5%. This performance metric highlights the tool's reliability, although it also indicates a 6.9% rate of false positives. Senior consultants at SGH emphasize that the AI is intended to support, rather than replace, the expertise of human radiologists. Currently, SGH employs around 100 radiologists, including 21 specialized in neuroradiology, who handle thousands of scans daily.

The implementation of iScreen is part of a broader strategy to integrate AI into the Singapore public healthcare system. The tool is deployed on AimSG, a national AI-enabled medical imaging platform, which facilitates its adoption across various hospitals without the need for specific hardware or software. This positions iScreen as the first AI model for head CT scans on the AimSG platform, with potential for wider usage as discussions with other public hospitals are underway.

The collaboration between SGH and Iota MedTech began in 2019, driven by a shared goal to optimize resources in brain surgery and improve patient care. The development of iScreen reflects a growing trend in the healthcare sector, where AI technologies are increasingly being utilized to enhance diagnostic accuracy and operational efficiency. As healthcare providers become more comfortable with AI integration, the potential for scaling such technologies across various medical applications is substantial.

The broader implications of iScreen's launch extend beyond immediate patient care. It signals a shift in how healthcare systems may approach diagnostics, particularly in emergency medicine. As AI tools like iScreen become more prevalent, they could reshape the competitive landscape of medical imaging and diagnostics, prompting other hospitals and healthcare providers to invest in similar technologies. This trend may also lead to a reevaluation of training programs for radiologists, as the role of AI in diagnostics continues to evolve.

Looking ahead, the successful integration of iScreen and similar AI tools could redefine operational workflows in hospitals, enhancing the speed and accuracy of diagnoses while allowing healthcare professionals to focus on more complex decision-making tasks. The ongoing collaboration between SGH and Iota MedTech suggests a commitment to continuous improvement and innovation in healthcare, paving the way for future advancements that could further transform patient care.