AI Predicts With 94% Accuracy Who Won't Respond to Therapy | Nijat Ahmadov, Nucs AI

Only 45% respond to radioligand therapy. Nijat Ahmadov on finding the right patient before treatment starts. Book Ian for your event → iankhan.com | booking@iankhan.com SUMMARY Nijat Ahmadov co-founded Nucs AI to solve patient selection in radioligand therapy. He explains why selection is the central challenge, the thousands of scans used to train multiple models, why PSMA PET is among the hardest modalities for AI, and lesion-level response prediction as the next goal. KEY FINDINGS • Radioligand therapy as a live treatment with a selection problem • Only 45% of patients responding to the therapy • 94% accuracy identifying who will not respond • Over 90% accuracy detecting and segmenting tumour lesions • Thousands of scans used across a multi-model training process • PSMA PET as one of the most challenging modalities to analyse with AI • Lesion-level response assessment as the next development goal • How the healthcare industry is adopting AI as a working tool NOTABLE QUOTES "Right now only 45% of the patients respond to the radioligand therapy." — Nijat Ahmadov GUEST Nijat Ahmadov, Co-Founder & CEO, Nucs AI FULL TRANSCRIPT iankhan.com/futurist/nijat-ahmadov-nucs-ai-patient-selection Recorded at RSNA 2025 for The Futurist. ──────────────────── ABOUT IAN KHAN — HEALTHCARE AI KEYNOTE SPEAKER Most AI keynotes end with a better understanding. Mine end with a measured readiness score and a ninety-day plan. Your leadership team is assessed before the event, the room sees its own co