Medical Image Post-processing Software (uOmnispace.CT)

The Shanghai United Imaging Healthcare Co., Ltd.'s uOmnispace.CT is a Class II medical image post-processing software designed for comprehensive interpretation and analysis of CT scans. Its core methodology involves a versatile platform with diverse clinical applications, significantly enhanced by the recent integration and rigorous validation of deep-learning algorithms to improve efficiency and accuracy across various diagnostic evaluations.

Shanghai United Imaging Healthcare Co., Ltd. has received approval to market its Medical Image Post-processing Software (uOmnispace.CT), determined to be substantially equivalent to previously marketed medical devices. This software, categorized as a Class II medical image management and processing system, supports the interpretation and evaluation of medical examinations within healthcare institutions. The company, located in Shanghai, China, received the decision on May 14, 2025, following a review of their submission dated April 3, 2025.The uOmnispace.CT software operates as a post-processing platform for viewing, manipulating, evaluating, and analyzing medical images. It includes various applications such as colon analysis, dental application, lung density analysis, lung nodule review, vessel analysis, brain perfusion, heart analysis, calcium scoring, dynamic analysis, bone structure analysis, liver evaluation, dual energy imaging, cardiovascular combined analysis, and a newly added body perfusion application. Key modifications to the device include extending its patient population for some features and integrating deep-learning algorithms into several applications, including lung density, vessel, heart, liver, and cardiovascular combined analysis functions. Performance data, including software verification and validation, demonstrated the safety and efficacy of these deep-learning algorithms through rigorous testing against established acceptance criteria across various patient subgroups.

AI Regulatory Concerns

Powered by

Jev reads the full FDA letter and returns calibrated concern probabilities per regulatory category. These are AI estimates, not regulatory advice.

• Q1. How do the integrated deep-learning algorithms compare in performance to traditional methods for specific applications like lung density or vessel analysis in terms of diagnostic accuracy and efficiency?; What specific patient subgroups were included in the validation studies for the deep-learning algorithms, and were there any observed differences in performance across these groups?; Could the document provide more detail on the acceptance criteria used for the rigorous testing of the deep-learning algorithms' safety and efficacy?; What is the intended clinical workflow integration for uOmnispace.CT, particularly considering its new body perfusion application and extended patient populations?; Are there plans for further expansion of deep-learning integration into other existing or new applications within the uOmnispace.CT software, and what are the clinical drivers for such expansions?