We’re thrilled to share new research led by Dr. Shuhua (Steve) Zheng, one of our outstanding trainees in Radiation Oncology at Northwestern University, that may shift how we think about preoperative decision-making in pancreatic cancer. In this study, Dr. Zheng and team developed PanScore—a deep learning model trained on genomic features to predict occult liver metastases in patients with pancreatic ductal adenocarcinoma (PDAC). Using over 4,000 cases across independent cohorts, PanScore stratifies patients into biologically distinct risk groups. By integrating common mutations and CNVs (including KRAS, TP53, CDKN2A, and SMAD4), this tool can flag patients unlikely to benefit from surgery alone, potentially guiding more personalized neoadjuvant strategies. This work exemplifies the intersection of machine learning, translational genomics, and clinical oncology—and we couldn’t be prouder of Dr. Zheng’s vision and execution. Full article: sciencedirect.com/science/ar… Let’s reimagine PDAC risk stratification, one algorithm at a time. #PancreaticCancer #AIinMedicine #RadiationOncology #DeepLearning #Genomics #PrecisionOncology #MentorshipMatters #NorthwesternMedicine

It's out! Using deep learning, we addressed one of radiation oncology’s toughest challenges: accurate and automated tumor delineation. Our model segments lung tumors in motion with near-expert accuracy—scalable, reproducible, and ready for the clinic. rdcu.be/et5U0

Watch our model in action below. It’s fast, accurate, and designed to change how we target tumors in real-world settings. Proud of our team @NorthwesternMed @LurieCancer This is worth a look: rdcu.be/et5U0 #AIinHealthcare #LungCancer #RadiationOncology #DeepLearning #TumorSegmentation #MedicalImaging #4DImaging #InnovationInOncology