Abstracting and Indexing

• PubMed NLM
• Google Scholar
• Semantic Scholar
• Scilit
• CrossRef
• WorldCat
• ResearchGate
• Academic Keys
• DRJI
• Microsoft Academic
• Academia.edu
• OpenAIRE
• Scribd
• Baidu Scholar

Introduction: Conventional contrast dosing in dynamic CT and MRI imaging relies on fixed protocols that often overlook individual patient variations in tissue perfusion and vascular dynamics. Recent advancements in radiomics and artificial intelligence (AI) offer a novel opportunity to personalize contrast administration in real time. This study aimed to evaluate the feasibility, diagnostic impact, and safety of real-time radiomic-guided adaptive contrast dosing in dynamic CT and MRI scans among patients with hepatic, renal, and soft tissue lesions at Katihar Medical College, India.

Methods: We prospectively enrolled 120 patients scheduled for dynamic contrast-enhanced imaging. Participants received either a conventional fixed-dose protocol or an AI-driven adaptive dosing protocol in which real-time radiomic texture analysis modulated contrast volume and injection rate on a per-patient basis. Key endpoints were lesion-to-parenchyma contrast (LPC) ratio, radiomic texture features (entropy and kurtosis), total contrast volume, diagnostic confidence, and inter-observer agreement.

Results: The adaptive dosing cohort achieved significantly higher LPC ratios than the fixed-dose group (1.78 ± 0.12 vs 1.42 ± 0.15; p < 0.001). Entropy increased and kurtosis decreased, indicating sharper lesion delineation. Additional hepatic and renal lesions were identified only in the adaptive arm, prompting changes in patient management. Mean contrast usage fell by 12 %, and no instances of nephropathy or serious adverse reactions occurred. Diagnostic confidence and reader agreement both improved markedly.

Conclusion: Real-time, radiomic-guided contrast adjustment boosts diagnostic accuracy, lowers contrast requirements, and maintains patient safety, supporting a personalised imaging paradigm.