New features:

A new release of contextflow iPE v1.0.2 has been deployed to the deepcOS marketplace.

• contextflow iPE v1.0.2 is now available for customer on the deepcOS marketplace, providing access to the latest features and improvements.

contextflow iPE v1.0.2 has been released to the deepcOS marketplace. This product detects incidental Pulmonary Embolisms on chest CTs.

Radiobotics RBfracture v2.6.1 has been released to the deepcOS marketplace. This is an update from the previous version v2.4.0.

• Studies that are confidently normal are now flagged with a white circle in the top left corner, improving workflow efficiency by quickly identifying normal cases.
• Confidence scores are now included on each bounding box, providing additional information to the customer for better decision-making.
• DICOM SR is optionally returned, increasing flexibility in data handling and integration with other systems.

Oxipit ChestEye v2.6.0.5 has been released to the deepcOS marketplace. This is an update on the previous version v2.6.0.3.

• SNOMED codes are now associated with each finding
• There are no changes to DICOM outputs.

Updates:

Migration of Radpair instance to a new European instance for customer.

• The production Radpair DNS for EU access has been migrated to eu.radpair.com and api-eu.radpair.com.
• The old DNS (de.radpair.com including api-de.radpair.com) will be removed after customer confirms the switch, ensuring a seamless transition.
• This migration ensures customer benefits from the latest infrastructure and improved performance within the European region.

contextflow CFA Chest CT v3.3.1 has been released to the deepcOS marketplace. This is an update from the previous version v3.2.0.

• Introduces longitudinal processing capabilities for enhanced analysis.
• Enables optional return of DICOM SEG data upon request.
• Allows customization of the nodule size range reported. Bu default, this is 4mm to 30mm.

New Feature:

The hub now displays the username of the logged-in user, enhancing user experience and personalization.

• The user's email address is used to display a personal identifier in the bottom left corner.
• A profile picture with the first initial of the username is shown next to the name.

New Feature:

Added patient information display to the viewer for enhanced identification and context.

• Implemented the display of Patient Name, Age, Patient ID, and Sex in the viewer launched from the gate manager.
• This addition provides customer with immediate access to essential patient details, improving workflow efficiency.

New feature:

Consolidated "confirm and next" buttons in the DICOM viewer to reduce mouse movement and streamline the evaluation process for customer.

• Introduced a combined "Confirm and Next" button at the bottom of the page to minimize clicks during evaluation.
• Added a "Skip" button alongside the "Confirm and Next" button to provide an alternative action for the customer.
• This change reduces unnecessary mouse movement, making the workflow more efficient, especially when using only one AI engine.

New Feature:

Patient information is now displayed in the viewer, providing essential context for image analysis.

• The viewer now displays patient name, date of birth, patient ID, and sex, improving user workflow by providing immediate access to patient details.
• This enhancement allows customer to quickly verify patient data directly within the viewer, reducing the need to consult external systems.

Fixes:

Internal code was restructured for better maintainability.

• Internal code was restructured into a separate module for improved maintainability.
• No functional changes were introduced with this update.
• The changes were rolled out to all gates.

New Feature:

Introduced a redesigned backend and frontend for improved data reliability and visualization in the analytics sections.

• The new analytics dashboards and charts are now available, providing insights into studies received, processing outcomes, and AI solution performance.
• The updated analytics solution offers comprehensive data on AI landscape, including filters for installation and date range, detailed breakdowns per AI solution and version, and processing time metrics.
• The Solution Page is enhanced with tabs for Overview, Findings, and Concordance, offering operational metrics, AI output data, and comparison metrics between AI and reference data.

Coreline LCS/LAA/CAC v2.0.51.9 has been released to the deepcOS marketplace. This is an update from the previous version v1.1.46.9.

‍

• This version can be used for worklist and reporting integrations.
• VDT is now calculated for decreasing nodules, as well as nodules increasing in size.
• Increased configuration options are available.

Nurea PRAEVAorta2 v1.2.0 has been released to the deepcOS marketplace. This is an update from the previous version v1.1.2.

‍

• This version now offers the ability to perform longitudinal processing
• This version offers increased processing stabiltiy for larger studies.

‍

New Feature:

Implemented audit event logging for Gate Manager logins to enhance traceability and meet security requirements.

• Added logging of login events in Gate Manager, capturing the time, username (email), user agent, and host of each login.
• This enhancement improves traceability and helps in meeting security compliance requirements for customer.
• The new audit logs provide a detailed record of Gate Manager access, aiding in security monitoring and incident response.

Fixes:

Ensuring the referrer is sent via the hub to allow access to release notes.

• The noreferrer attribute was removed from the link to allow the referrer to be passed through.
• The Referrer-Policy header was adjusted to allow at least the origin to be sent for the release notes URL, ensuring that the protected page can verify the origin of the visitor.
• This update allows the customer to restrict access to the release notes page to visitors coming from a specific referring domain URL.

New feature:

Implemented HL7 interface adaptation for a customer to send priority information.

• Implemented the configuration in rockshrimp.
• Implemented the generation of HL7 based on a listener on engine results successful.

Fixes:

System Manager now displays data correctly for all channel partners.

• The system manager was updated to include channel partners in data retrieval, resolving an issue where data was missing for some customers.
• This ensures that all customer data is accurately displayed in the system manager.

New feature:

A user's name will now be displayed in the bottom left corner of the screen in the gate manager. This enhancement provides constant user identification.

• The user's name is extracted from their login email.
• An icon displaying the user's first initial is shown alongside their name for quick recognition.
• This ensures that the logged-in customer is always clearly identified within the gate manager UI.

Fixes:

Data missing in System Manager due to missing channel partner consideration.

• The system manager was updated to include channel partners.
• This ensures complete data visibility for customer.

Fixes:

Fixed issues with gate events page, restoring descriptions and filter functionality.

• Restored the page description for gate events.
• Changed the filter type to a text filter for 'Subject ID' and 'Type' columns.
• Ensured the table fills the entire page, resolving the gap at the bottom.

Fixes:

Fixed issues with the Gate Events page, restoring descriptions and filter functionality.

• Restored the page description for Gate Events.
• Changed the filter type to text for 'Subject ID' and 'Type' columns, enhancing search capabilities.
• Ensured the table fills the page, resolving the previous layout gap.

Smart Soft HealthCare Columbo v2.12.65 has been released to the deepcOS marketplace. This is an update from the previous version v2.11.163.

• This version of the AI solution produces the deepc engine report JSON, which classifies the stenosis level of each vertebra of the spine as none, mild, moderate or severe.

The mediaire mdkee v1.3 AI Solution has been released to the deepcOS AI marketplace. This is the initial release of this AI Solution to deepc.

‍

• This AI solution generates the deepc engine report JSON, which includes all calculated measurements and gradings regarding ligament, meniscus and cartilage damage.

Fix:

Sign buttons are now hidden when no HL7 message has been received, streamlining the user interface and preventing confusion.

• The sign buttons are now hidden when a report is created and no HL7 message was received. This prevents users from attempting actions when the necessary data is missing.
• A check is performed on the Gate to determine if the message exists and forwards this information, allowing the user to only have the option to copy the data.
• The hide_multi_author_actions message is used to control the visibility of the sign button.

The latest version of Image Biopsy Lab LAMA v2.0.3 has been released to the deepcOS marketplace

‍

• This AI solution produces the deepc engine report JSON, which contains all measurements, enabling the opportunity for comprehensive AI based reporting.

New Features:

The release notes page is now accessible from the Gate Manager, ensuring all past release notes are available in one location.

• The new release notes page is accessible from the Gate Manager.
• All previous release notes are available on the new page: https://www.deepc.ai/pages/product-release-notes-and-changelog.
• This update centralizes release information for the customer, improving accessibility.

Fixes:

Added information to ACK messages to include debug information.

• ACK messages now include OBR 3.1, ORC 3.1, OBR 4.1, and OBR 13.
• This allows integration engineers to quickly verify if accession number, ordered exam, and clinical history are transferred to the system during integrations for the customer.

Improvement:

Radpair integration is enhanced to correctly add information from different sources, ensuring comprehensive report updates.

• Implemented logic to add information from different sources on the deepc side, ensuring that reports are updated with all existing information.
• Additive behavior for clinical_notes and findings to keep former information and add any new data.
• Updated replacement logic for studyDescription, modality, BodyPartExamined, SOPClassUID, and ContrastBolusAgent to replace only when appropriate.

Improvements:

UI enhancements to the Job Details page improve usability and clarity.

• The header now displays the AI solution and customer name instead of the job ID, improving context.
• Information is separated into cards (Job Metadata, Execution Timeline, and Data Transfer) for better organization.
• Key fields in Job Metadata can be copied with a single click, streamlining workflows.

Improvements:

Simplified study status logic to improve clarity and usability for customer.

• The study status in the study list has been simplified to 'Incoming' or 'Finished' to provide a clearer understanding of the study's processing state.
• UI enhancements, including highlighting in color, have been implemented to improve the visibility of study statuses.
• The default positioning of columns has been updated to show 'Study Status' in column 2 and 'AI Solutions' in column 3 for better organization.

Updates:

Updated dropdown options in the Evaluations Viewer to better reflect result types and improve data accuracy for analytics dashboards.

• Replaced Incorrect Findings with False Positive to align with standard terminology.
• Replaced Missing Findings with False Negative for consistency in evaluations.
• Removed Findings are Irrelevant option to streamline the evaluation process.

Update:

Enhanced report generation to include findings from HL7 ORM messages, specifically NTE segments, improving data integration for customer.

• Implemented parsing of HL7 ORM messages to extract findings from the NTE segments.
• Integrated extracted findings into the report, providing a more comprehensive view of patient data for the customer.

Fixes:

Enabled HL7 debugging for customer IT.

• Provided access to the gate manager and Gate Events page, displaying HL7 messages and extracted content, to allow the customer to debug their HL7 messages.
• Released the Gate Event page as beta with a default filter for 'gate' events.

Improvement:

Improved error messaging for missing accession numbers, enhancing user experience.

• The page now displays an informative message when an accession number is missing, guiding the user on how to proceed.
• The message advises the user to either provide the missing accession number or contact their IT administrator for assistance.

New Feature:

A disclaimer was added to the DICOM viewer to clarify its intended use.

• A label is now clearly visible in the viewer stating that it is for reviewing and evaluating AI results, and not for diagnostic use. This ensures users understand the viewer's purpose.

New Feature:

A disclaimer was added to the DICOM viewer to clarify its intended use.

• A label is now clearly visible in the viewer stating that it is intended for reviewing and evaluating AI results, and not for diagnostic use by the customer.

Improvement:

A re-structure from the webhook documentation including workflows and examples helps our HIT partners to integrate with our webhook API.

Improvement:

With the introduction of AI reporting, documentation of the integration for our HIT partners is created. It comprises the launch - which needs to be configured in a PACS or RIS, the ingestion of clinical information from a RIS and the return of the final report text to the RIS.

‍

New feature:

Our AI Reporting system is now ready to be tested in German. Speech recognition and final report creation is based on German trained algorithms now.

‍

‍

New feature:

For beta-customers and prospects it is possible to test the AI Reporting module. Several AI algorithms are used for speech recognition, automatic template selection, final report structuring and impression creation. All available in English.

‍

New Features:

This update adds further details to gate-rejections messages and also allows to export data from the studies view.

Studies view: contains information that was previously accessible in AI worklist and study status views

AI worklist and study status views have been merged to studies view.

Export of studies view

Data from the studies view can be exported with a button click: Export Data as CSV will provide an extract of the current view; Export Instance Tags CSV will export data of configuration relevant DICOM tags of all studies on the gate.

Gate rejection-messages per series

For jobs in status gate-rejected, it is now possible to see the reasons for rejection not only per AI-solution-job, but also per series of the job in case the AI-solution operates on multiple series.

‍

‍

Improvement:

The representation of data is re-organized by displaying all information based on studies instead of job, which represent the customer view on the data. Several jobs can be created for one study depending on configured AI solutions or if AI solutions work on series. This distincation is important for IT administrators and maintenance, but not the medical consumer.

‍

New Feature:

Jobs in status AI-rejected and failed, now provide addtional details in the message diaglog (in study status view, jobs table). Messages are shown, if provided by the AI solution.

‍

Improvement:

Each Gate can be enlarged to see the all configured AI solutions and it's traffic. Requested, completed, failed and rejected unique job counts are displayed in a table view and as a graph.A unique job is created depending on the AI solution on patient, study or series level.

‍

Another tab in the analytics sections displays the AI solution performance

AI solution performance

This tab contains a table called AI findings summary, which lists all findings of the system over a selected period of time, it’s AI solution product name, the absolute number of positive predictions and the positive prediction rate. And a graph for positive prediction rate, which is the proportion of jobs that have predicted a present finding per completed job.

‍

New Features:

This update introduces the possibility to review DICOM output of AI solutions directly in the gate, e.g. to evaluate AI solutions outside the PACS.

DICOM viewer

Gate Manager now contains a DICOM viewer to review studies (input and AI output). To open the DICOM viewer, click on the open viewer button in the study status view.

AI solution filter

The new column AI solutions in the AI worklist, contains a list of AI solutions the study has successfully been processed with and allows to filter the AI worklist for studies that have been processed with specific AI solutions.

‍

The initial deepcOS® Hub release allows the following components to be accessible in a single professional unified frontend experience: AI Portfolio, Analytics, System Manager, Developer Docs, Submit AI and Support.

AI Porfolio

This page displayes the latest AI solutions on the deepcOS® platform. It is searchable for specific keywords or filter for solutions concerning modalities, applications, body parts or regulatory clearance. For each solution a product detail page gives more information about the solution and if applicable shows samples and references.

Analytics

Using an interactive dashboard to investigate on their own data, customers can navigate through their usage of deepcOS® Cloud infrastructure. Following charts and data points are available:

• Success rate trend: This displays the trend of proportion of jobs which returned a successful response.
• Gates: List of all gates of the organization or by a customers of the organization (in case of channel partners)
  • Total requests per gate
  • Success Rate per gate: Rate of studies, which were successfully processed and a result created. It does not show if the result was displayed in the PACS.
  • Rejection Rate per gate: Rate of studies, which were rejected by the AI solution itself due to mismatching specifications.
• AI solutions: either all available to the customer or filtered by gate
  • Total requests for the AI solution
  • Success Rate for the solution
  • Rejection Rate for the solution
  • Average Execution Time for the solution in seconds
• Total requests per day

System Manager

System Manager shows a list of Gates for this user and their uptime of the last 24 hours or 7 days in percentage and visually. The user can then click to navigate to the deepcOS Gate manager.

Deverloper Docs

AI solution developers can find here the documentation how to integrate their solution with the deepcOS® platform. The two APIs to consume AI processing and output data are described here in detail for HIT integrations.

Submit AI

The is a submission workflow to support AI partners to submit their new AI solution or an update to the deepcOS® platform.

Support

This page enumerates links for different user groups to reach out to deepc or read the release note for deepcOS® Hub.

‍

New Feature:

New Study Status Display

The study status column in the AI worklist view now contains more detailed study status information including information on the success of the AI processing.

Release Notes

A release note section is now available in the system page to inform users about new or updated features within deepcOS Gate Manager.

‍

New Feature:

This update introduces multi-select options for table filters in AI worklist and study status, and removes the general view from the system page.

Multi-Select Filter Options

Table filter options within the AI worklist and study status now support multi-select for relevant columns, enabling more refined filtering.

Removed General View

The general view has been removed from the system page for a streamlined user experience.

‍

First SW Release

The initial release contains essential components for high-level system monitoring, AI worklist management, and DICOM routing and morphing configurations.

System Monitoring

Provides high-level monitoring of system resources including CPU, memory, and storage usage, along with health status for specific deepcOS Gate services.

AI Worklist

Lists studies recently processed by deepcOS, offering high-level information on each study with search and filter options.

Study Status

Displays the status of a specific study, including processing status per AI solution, DICOM tags, turnaround time, and reasons for rejections if applicable.

Routing Information

Shows routing information, including DICOM endpoints, configured AI solutions per endpoint, and applicable filters for each AI solution.

Morphing Configuration

Displays morphing configuration details used to normalize DICOM tags that form the basis for routing and filtering.

‍