Population-level segmentation of the penis in magnetic resonance imaging using deep learning

A deep learning model segments the penis in DIXON magnetic resonance imaging with 92% accuracy, enabling population-scale phenotyping.

viernes, 3 de julio de 2026 • 2 min read • Q2BSTUDIO Team

Automation of penile measurement in magnetic resonance imaging

Artificial intelligence is radically transforming the field of medical imaging, enabling automated analyses that previously required hours of manual work. One of the most striking advances is the automatic segmentation of complex organs using deep learning, a technique now applied even to traditionally difficult-to-measure anatomical structures, such as the penis. Until recently, clinical evaluations were based on poorly reproducible external measurements that completely ignored the internal portion of the organ. Magnetic resonance imaging (MRI) offers a complete three-dimensional view, but large-scale manual segmentation was unfeasible.

Thanks to architectures like nnU-Net and datasets annotated by experts, it is now possible to train models that achieve accuracies comparable to those of a specialist. In a recent study with over 34,000 participants from the UK Biobank, an automated system achieved a Dice coefficient of 0.92 and remarkable longitudinal reproducibility. This opens the door to population-scale phenotyping in urology and reproductive health, allowing the association of total penile volume —including its internal and external components— with genetic, endocrine, and metabolic factors. The possibility of applying such tailored applications in hospital or research settings requires a robust and flexible technological infrastructure.

At Q2BSTUDIO, we develop custom software to integrate artificial intelligence models into real clinical workflows, ensuring scalability, security, and performance. Our AWS and Azure cloud services enable the deployment of these systems in the cloud with high availability, while our cybersecurity solutions protect sensitive patient data. Additionally, through business intelligence services with Power BI, we transform segmentation results into interactive dashboards for researchers and physicians. The enterprise AI we design includes AI agents capable of automating complex image analysis and biomarker extraction processes.

This approach not only accelerates research in urology but also lays the groundwork for more precise and personalized diagnostic tools. If your organization seeks to implement similar automated segmentation solutions or needs a comprehensive medical data analysis platform, at Q2BSTUDIO we offer enterprise AI tailored to your needs, combining deep learning, cloud computing, and business intelligence. Likewise, we develop custom applications that integrate these models into production environments, ensuring effective adoption in the healthcare sector.

A BREAK?

Play for a moment before you go

OUR SERVICES

How we can help you

Do you have a project in mind?

Tell us your vision and we'll turn it into a software solution. Whatever the scope, we make your idea real.