The deepcOS® Gate is installed on premise in a clinical institution, and is the core building block on which the rest of deepcOS® is built.
The deepcOS® Gate sits on premise and creates the secure connection to HIT systems such as the PACS, RIS, EMR/EHR, or with the modality. This enables the flow of imaging data to and from the deepcOS® Gate, either automatically or manually.
Before sending studies to the cloud for AI processing, personal information from the DICOM headers is temporarily removed to ensure it remains on premise.
Both DICOM-tag and AI-based orchestration ensure that imaging data is sent to the appropriate AI solution(s) for processing. Orchestration can be automatic or manual, depending on the AI solution and customer preference.
The deepcOS® Gate supports the secure connection with the deepcOS® Cloud to send and receive imaging data. By default, imaging data is not forwarded to other cloud environments and remains within deepcOS®.
After cloud processing, AI results are returned to the deepcOS® Gate where personal data is reassigned before being sent back to the PACS for physicians' review. AI results can then be used as part of the regular radiology reporting process.
The deepcOS® Gate exposes the deepcOS® Clinical API which passes AI output data to downstream systems. This can enable, for example, the display of clinical findings and AI processing status within the existing radiology worklist.
When deepcOS® Gate is connected to the AI Marketplace, institutions can access a myriad of globally leading AI solutions simultaneously.Learn more ->
Enrich the radiology workflow with additional deepcOS® applications, such as an AI-worklist for patient triage and AI Results Explorer.Learn more ->
Export anonymized data from the PACS at scale for research projects, including the development of new AI solutions. Data Privacy Regulations must still be met independently.Learn more ->
deepcOS® Gate exposes the deepcOS® Clinical API, allowing any Healthcare IT system to integrate with deepc and display AI resultsLearn more ->