Welcome
QUIN Lab
Quantitative Intelligent Imaging
Advancing pediatric imaging through AI, quantitative MRI, and intelligent image analysis
The Quantitative Intelligent Imaging (QUIN) Lab at Boston Children's Hospital and Harvard Medical School develops new imaging and artificial intelligence technologies to improve the diagnosis, characterization, and monitoring of disease in children.
We bring together expertise in medical imaging, MRI physics, machine learning, computer vision, and clinical medicine to develop imaging methods that are faster, more quantitative, more robust, and more clinically useful.
Our research spans three closely connected goals: creating next-generation MRI acquisition and reconstruction techniques, developing AI methods that learn from both medical images and clinical information, and translating these technologies into quantitative imaging biomarkers that can improve patient care.
Our Research
Artificial Intelligence & Multimodal Learning
We develop machine learning methods that integrate medical images, radiology reports, and other clinical information. Our work includes vision-language models, foundation models, generative AI, weakly and semi-supervised learning, and methods that reduce dependence on large manually annotated datasets.
Next-Generation MRI
We develop fast and motion-robust MRI methods designed for the challenges of pediatric imaging. Our research includes advanced image reconstruction, motion correction, quantitative MRI, super-resolution, diffusion imaging, and techniques that reduce scan time and the need for sedation.
Quantitative Imaging Biomarkers
We develop computational methods that transform medical images into reproducible measures of anatomy, tissue microstructure, physiology, and function. These quantitative biomarkers are designed to improve disease characterization and enable more objective monitoring of patients over time.
AI for Pediatric Disease
Working closely with radiologists and clinical collaborators, we translate our methods to important pediatric applications involving the brain, abdomen, kidneys, gastrointestinal tract, lungs, musculoskeletal system, and other organs.
From Methods to Clinical Impact
Our research begins with fundamental advances in imaging and computation but is driven by clinical problems. By bringing engineers, computer scientists, MR physicists, radiologists, and clinicians together, we aim to develop technologies that can move beyond proof-of-concept studies and ultimately improve how children are imaged, diagnosed, and treated.
Research Environment
QUIN is part of the Computational Radiology Laboratory at Boston Children's Hospital and is affiliated with Harvard Medical School. Our multidisciplinary environment provides access to advanced MRI systems, large clinical imaging datasets, high-performance computing resources, and close collaborations across pediatric subspecialties.
Latest Research
Isik, H.U., Ozaydin, M.A., Kurugol, S. and Ertekin, Ş., 2026. ARC-CT: Anatomy-Routed Contrastive Vision-Language Learning for 3D Chest CT. arXiv preprint arXiv:2608.28455. (Oral presentation at MICCAI Thoracic Image Analysis Workshop)
Timms, L., Utkur, M., Ariyurek, C., Hewlett, M., Kurugol, S. and Afacan, O., 2026. Fast, Robust T2‐IVIM Quantitative MRI With Distortion and Motion‐Corrected Multi‐Echo EPI. Magnetic resonance in medicine, 95(5), pp.2527-2537.
Spieker, V., Huang, W., Ariyurek, C., Timms, L., Rueckert, D., Afacan, O., Schnabel, J.A. and Kurugol, S., 2026. Primitive Representation Learning for Unsupervised Dynamic Contrast Enhanced MRI Reconstruction. arXiv preprint arXiv:2608.18055. (MICCAI Off-Grid: 1st Workshop on Continuous Representations and Grid-Free Methods in Medical Imaging)
Join QUIN
We welcome collaborations with researchers and clinicians interested in medical imaging, artificial intelligence, and pediatric disease. Opportunities are also available for postdoctoral fellows, graduate students, visiting researchers, and research trainees.