Radiology departments must manage massive volumes of imaging data, multiple software systems, and growing diagnostic workloads while maintaining speed, accuracy, and quality of care. During WHX, Siemens Healthineers hosted a series of conversations focused on Photon imaging and artificial intelligence in clinical workflows. Professionals pointed to how consolidating data management, image viewing, reporting, and AI into a single workspace can support radiologists across modalities, including use cases relevant to cancer detection. Siemens Healthineer developed a 3D mammography machine that would also use AI-driven capabilities such as automated cancer probability scoring, lesion highlighting, AI-enhanced worklists, advanced reporting tools, and ambient reporting concepts.
“What we’ve done in the last years is basically consolidating all the individual applications and solutions that we have developed and putting this into one solution,” stated Daniel Felke, Head of Sales for Digital Automation at Siemens Healthineers, emphasizing the solution lies in consolidation, accessibility, and artificial intelligence.
Many hospitals and clinics operate with disconnected systems, separate PACS, reporting tools, archives, and vendor-specific software that don’t easily communicate with one another. This fragmentation often limits access to valuable patient data and slows down clinical workflows.
“A lot of customer clinics have issues that they have to deal with, a lot of vendors, with a lot of software, a lot of individual contributions, and we’ve now managed to include everything into one work.”
The approach Daniel describes focuses on bringing imaging, reporting, artificial intelligence, and research tools together into a single digital workspace, regardless of where the data originates.
“It doesn’t matter where the data’s coming from, from a different vendor, from our own CT, if it’s a DICOM or MP three file or multimedia, everything can be handled.”
From Fragmented Systems to One Workspace
A key challenge in modern healthcare is not the lack of data, but the inability to access and connect it. Imaging data, patient history, videos, and external files often exist in silos.
“We know a lot of information is being acquired on the scanners and the patient data, but in the end, not everybody can access the data.”
By associating all acquired information directly with the patient record, including non-traditional data such as videos or recording, clinicians gain a more complete clinical picture.
“Everything that you record and acquire can be associated with the patient.”
This continuity becomes especially valuable when patients return for follow-up care weeks or months later.
Beyond image analysis, AI also transforms reporting. Measurements, annotations, and findings can flow directly from the image viewer into structured reports, reducing manual effort and saving time.
New tools also help summarize complex reports into concise clinical impressions. More recently, ambient reporting introduces a conversational approach to documentation. This structured output not only improves consistency, but enables long-term analytics and research.
AI as a Silent Assistant in Radiology
Rather than replacing radiologists, AI operates in the background, automatically analyzing imaging studies before they are reviewed. This helps prioritize urgent cases and surface findings that might otherwise be overlooked.
“Before the radiologist even started to look at the study, there’s really automated findings that are being generated, which helped them in workflow and coping with all of the data that they have to look at during the day,” explained Felke.
AI-enhanced worklists can flag critical cases first, a crucial feature in large institutions handling hundreds of studies daily.
One example Felke illustrates is how AI can reduce the risk of missed diagnoses.AI systems can detect such findings automatically and draw attention to them, while leaving final decisions to the radiologist.
“You come to the hospital with a broken rib, we get a chest scan, and the attention is towards the broken bones, right? But at the same time, it might be that there’s a lung module.”
Artificial Intelligence Is Transforming Mammography
Mammography remains the main method of breast cancer screening, yet it is also one of the most demanding areas in radiology.
High reading volumes, subtle findings, and the need for early detection place significant pressure on radiologists. That is why AI is being adopted as a clinical support tool to help address these challenges not by replacing radiologists, but by enhancing accuracy, consistency, and efficiency.
One of the primary applications of AI in mammography is lesion detection. AI algorithms are trained on large datasets of annotated mammograms to recognize patterns associated with malignancy, including masses, calcifications, and architectural distortions.
“The images are sent to an AI tool and this tool analyzes the images,” said Ronald Froehlich, Head of Marketing of Xray products at Siemens Healthineers. “It gives different kind of information, the whole examination from a level 0 to 10 of high priority of cancer,”
The AI then marks the areas where the tumor can be, which gives the radiologists the opportunity to analyze the area. The benefit goes beyond only detecting the cancer but also give the radiologists who might be under pressure to view many cases the opportunity to go through efficiently.
