AI helps spot pancreatic tumors in ultrasound

Kari Jaaskelainen

12 Jan 2026 — 1 min read

AI helps spot pancreatic tumors in ultrasound

Pancreatic cancer is aggressive and hard to detect early. This study tested a Vision Transformer–based deep learning model that segments tumors in endoscopic ultrasound (EUS) images.

Training: 17,367 EUS images from two public datasets using 5-fold cross-validation.
External test: 350 images from a separate public dataset, with radiologist-drawn masks.
Performance: Dice ≈ 0.65, IoU ≈ 0.61, sensitivity ≈ 72%, specificity ≈ 98%.

Results were consistent across datasets, but about 9.7% of cases showed multiple mistaken outlines. The authors note that varied data sources and limited external validation mean the model needs further refinement, standardized datasets, and prospective studies.

Why it matters: More objective EUS tumor segmentation could support faster, more consistent assessments and research—helping clinicians, not replacing them.

Read the preprint: https://arxiv.org/abs/2601.05937v1

Paper: https://arxiv.org/abs/2601.05937v1

Register: https://www.AiFeta.com

#AI #MedicalImaging #PancreaticCancer #Ultrasound #EUS #DeepLearning #VisionTransformer #HealthcareAI #Radiology #Segmentation

Meet GRACE: a moral governor for safer, more transparent AI

AI agents are getting powerful—so how do we make sure they do the right thing, not just the effective thing? Meet GRACE, a reason-based moral governor that keeps AI behavior aligned with human norms by separating moral reasoning from goal-driven decision-making. * Moral Module: uses deontic logic and explicit reasons

Why some fine-tuned LLMs miss phishing—and how to fix it

Not all fine-tuned LLMs spot phishing equally. A new study tests Llama 3.1 8B, Gemma 2 9B, and Mistral on high-stakes phishing detection—and uses SHAP and mechanistic interpretability to reveal why models do (or don’t) generalize. * Architecture × data diversity matters: Gemma 2 9B hits state-of-the-art performance (F1

AI that explains itself—by following the science

AI that explains itself—by following the science Black-box AI can be powerful, but it often can’t tell us why it made a decision. Concept Bottleneck Models (CBMs) try to fix that by predicting human-understandable concepts first, then the final answer. The catch: standard CBMs ignore domain-specific cause-and-effect and

Meet GenomAgent: A Team of AI Specialists for Smarter Genomics Q&A

TL;DR: Finding reliable answers in genomics is hard. GenomAgent turns one big AI into a coordinating team of specialists—and beats the current leader by 12% on a key benchmark. Genomic facts live across many databases. Standard chatbots struggle because they can’t flexibly query those sources. GeneGPT added

AI helps spot pancreatic tumors in ultrasound

Read more

Meet GRACE: a moral governor for safer, more transparent AI

Why some fine-tuned LLMs miss phishing—and how to fix it

AI that explains itself—by following the science

Meet GenomAgent: A Team of AI Specialists for Smarter Genomics Q&A