Computational Neurosurgery Research

Neuroinformatics and Clinical Data Science

Every neurosurgical procedure generates information that can improve care for future patients. Our team transforms those clinical experiences into evidence that guides decision-making, advances surgical practice, and improves patient outcomes. To achieve this, we develop and manage large-scale clinical data resources, including national neurosurgical registries and multi-institutional datasets that capture outcomes across a broad spectrum of neurological diseases.

These resources have helped define outcomes after surgery for spinal disorders, one of the most common reasons patients seek neurosurgical care. By studying thousands of patients treated for degenerative spine disease, spinal deformity, spinal tumors, and other complex spinal conditions, we investigate how patient characteristics, surgical techniques, and emerging technologies influence recovery, complications, quality of life, and long-term outcomes. This work has contributed to the evaluation of minimally invasive and robotic spine surgery while helping establish evidence-based treatment strategies for spinal care.

The same commitment to combining clinical innovation and rigorous outcomes research extends to spinal cord injury. Through translational studies and clinical trials investigating regenerative therapies, including cell-based treatments, we seek to better understand the mechanisms of neurological recovery and develop new approaches for restoring function after severe injury.

Artificial Intelligence and Medical Image Computing

Medical images and other biomedical data contain a wealth of information that often remains difficult to capture through conventional approaches. Our Neural Image Computing lab  develops machine learning, computer vision, and medical image computing methods that transform these data into quantitative and clinically meaningful representations of patients. Through applications in disease screening, computational phenotyping, biomarker discovery, and outcome assessment, we seek to uncover patterns that improve diagnosis and deepen our understanding of human disease.

Developmental and congenital conditions have provided a particularly important setting for this work. By integrating imaging and phenotypic data, we characterize anatomical variation, identify features associated with genetic disorders, and develop reproducible measurements that support patient evaluation, treatment planning, and outcome assessment. These approaches enable the systematic study of conditions that have traditionally relied on subjective or qualitative evaluation.

The resulting technologies have helped establish computational phenotyping and quantitative anatomy as powerful tools for clinical research. From syndrome recognition to the characterization of craniofacial and developmental disorders, our work transforms anatomy into a measurable source of biological and clinical information, creating new opportunities to study disease and evaluate treatments.