The Fraunhofer Institute for Medical Image Computing (MEVIS) is working with physicians to develop artificial intelligence (AI) software to aid in diagnosis and therapy planning.
Fraunhofer MEVIS researchers will present their first promising results of their AI venture later this month in Chicago at the RSNA 2017 annual meeting.
Computer assistants and physicians are seeking to train artificial intelligence algorithms to improve their performance over time. One such initiative is a pilot project hosted by the Asklepios Clinic Barmbek in Hamburg, Germany.
A team of researchers and physicians are looking to measure as precisely as possible liver volume during therapy in CT or MRI scans. Among the objectives is to determine how repeated irradiation changes the size of an organ, a process that currently is time-consuming.
Improved versions of the AI platform can evolve from the computer assistant that the Fraunhofer experts developed in collaboration with physicians. The participating physicians actively partake in the development as part of their everyday work by feeding new cases into the software, reviewing how the assistant processes new datasets, and correcting the results.
By next year, the researchers hoped to have developed the platform to a point where it can become a marketable product.
![A normal mammogram confirmed by three-year radiologic follow-up illustrates reader-marked regions of interest (ROIs) during (A) unaided (round 1) and (B) artificial intelligence (AI)–assisted (round 2) reading. Each colored dot represents an ROI for recall by a human reader. Readers could mark more than one ROI per case, represented by multiple dots of the same color. During AI-assisted reading, the AI system displayed three visible prompts: two with suspicion of malignancy scores of 35% (left mediolateral oblique [L MLO] and craniocaudal [L CC]) and one with a suspicion of malignancy score of 10% (right craniocaudal [R CC]), shown as polygonal overlays. Without AI, six of 10 readers (60%) marked a false-positive ROI. With AI assistance, this fell to two of 10 (20%). R MLO = right mediolateral oblique.](https://img.auntminnieeurope.com/mindful/smg/workspaces/default/uploads/2026/07/2026-07-14-radiology-mammogram-ai-auto-bias.H0bYO8QlWs.jpg?auto=format%2Ccompress&dpr=2&fit=crop&h=167&q=70&w=250)
