About

Colbey Wade Freeman, M.D., is an Assistant Professor of Radiology at the Hospital of the University of Pennsylvania. He is an active radiologist within the Department of Medical Imaging at Penn Presbyterian Medical Center, Pennsylvania Hospital, and Chester County Hospital. Dr. Freeman specializes in advanced ultrasound imaging of the brain, including contrast-enhanced ultrasound and microvascular imaging, as well as clinical imaging informatics and spine imaging. His research focuses on optimizing imaging protocols, developing innovative imaging techniques, and improving clinical interpretation in neuroradiology. He has contributed to numerous publications in the field, emphasizing advancements in vessel wall imaging, neonatal neurovascular anatomy, and spine oncology decision-making.

Research topics

• Medicine
• Immunology
• Radiology
• Pathology

Selected publications

• Matched Analysis of the Risk Assessment and Prediction Tool for Post-Operative Disposition Needs in a Spinal Oncology Population

  Global Spine Journal · 2026-02-10 · 1 citations

  articleOpen access

  Study Design Retrospective cohort study. Objectives As cancer survival improves, metastatic spinal cancer has become increasingly common worldwide. Given the high resource demands of spinal oncology care, tools to optimize perioperative planning are essential. The objective of the study was to assess the effectiveness of the Risk Assessment and Prediction Tool (RAPT) in predicting post-operative needs in patients undergoing surgery for spinal tumors. Methods Consecutive patients (n = 384) undergoing spinal oncology surgery were enrolled and prospectively assessed with RAPT. Coarsened exact matching (CEM) was used to retrospectively isolate risk factors associated with outcomes. Enrolled patients with a low RAPT score (≤9, n = 44) were exact matched against high-scoring patients (10-12, n = 44). The primary outcome of interest was post-acute care disposition; secondary outcomes were 30- and 90- day ED visits, readmissions, and reoperations. McNemar’s test was utilized for matched comparisons. Results A low RAPT score was significantly associated with non-home discharge (OR = 4.33 [1.23, 15.20], P = 0.02) and 30-day readmission (OR = 3.66 [1.02, 13.14], 0.03). Among low-scoring patients, 31.8% required post-acute care (while only 11.3% of high-scoring patients required post-acute care). A low RAPT score was not associated with ER visits, reoperation, or mortality. Isolation of the RAPT walk score alone significantly predicted non-home discharge (OR = 2.8 [1.01, 7.78], P = 0.04). Conclusions When applied prospectively before spinal cancer surgery, the RAPT tool and its subcomponents effectively predict post-acute care needs. Pre-operative prediction of non-home discharge may help guide in-hospital resource allocation and post-acute care of spinal oncology patients.

  PublisherDOI

• Updates on Adult Transcranial Doppler, Gray-Scale, and Contrast-enhanced US Techniques

  Radiographics · 2026-03-12

  articleOpen access

  Transcranial US is a noninvasive, portable, real-time imaging technique that enables targeted evaluation of the intracranial circulation and select brain parenchymal structures through cranial bone windows. Traditionally, transcranial Doppler (TCD) has been used in well-established clinical scenarios, including vasospasm monitoring, stroke risk assessment, and ancillary testing for brain death. TCD encompasses two main techniques: nonimaging TCD (niTCD) and TCD imaging (TCDi). niTCD is a "blind" US technique that relies on standardized depths for vessel localization, guided by characteristic flow waveforms. In contrast, TCDi is a triplex approach that integrates gray-scale (B-mode) US, color Doppler US, and spectral Doppler US into an image-guided examination, enabling direct visualization of the intracranial anatomy and improving diagnostic confidence and reproducibility. In parallel, transcranial B-mode sonography (TCS), incorporated into modern TCDi, has demonstrated promising applications in adult patients, both in the neurointensive care unit and in outpatient practice. In addition, contrast-enhanced US (CEUS) has emerged as a powerful adjunct, markedly enhancing the Doppler signal in patients with poor bone windows, and has expanded the scope of transcranial US toward new exploratory vascular and cerebral perfusion applications. This article provides a comprehensive update on transcranial US in adults, including gray-scale imaging, Doppler techniques, and CEUS applications. With a focus on evidence-based indications and practical techniques, this article aims to support broader awareness and clinical adoption of transcranial US in contemporary adult neuroimaging, with a particular emphasis on TCDi, which remains underrecognized and underused despite its demonstrated clinical value in select indications. © The Author(s) 2026. Published by the Radiological Society of North America under a CC BY 4.0 license.

  PublisherDOI

• Evaluating the dosimetric impact of intrafraction shifts on critical OAR doses for spine SBRT using an In-House simulator

  Technical Innovations & Patient Support in Radiation Oncology · 2026-01-09

  articleOpen access

  This study evaluated the dosimetric impact of intrafraction motion during spine SBRT by developing a novel dose deviation simulator capable of analyzing translations and rotations in all six degrees of freedom. We retrospectively assessed the treatment plans from 18 spine SBRT patients treated at our institution with 24 Gy in two fractions by simulating shifts of 0.5-2 mm (translations) and 0.5-2° (rotations), and calculating changes in D0.03 cc to spinal cord, cord PRV, and thecal sac. Translational shifts, particularly in the vertical direction, produced the greatest observed deviations, with a 2 mm vertical shift yielding median increases, reported as percent of prescription dose, of 20.6% (4.93 Gy) to thecal sac D0.03 cc and 6.5% (1.55 Gy) to spinal cord D0.03 cc. Our simulator's accuracy was validated against treatment planning system-based forward calculations, showing mean differences of -1.3% ± 3.5% between the two systems. This study provides practical reference values to assist clinicians in rapidly evaluating the clinical significance of observed intrafraction motion. These findings may streamline decision-making during treatment and highlight the importance of precise positioning and motion monitoring in spine SBRT.

  PublisherOA PDFDOI

• Association of area deprivation index and median household income with early postoperative outcomes after spine tumor surgery

  Next research. · 2026-04-25

  article
  PublisherDOI

• Disparities in patient enrollment of spinal oncology clinical trials

  Neuro-Oncology Practice · 2025-06-13

  articleOpen access

  Abstract Background Subpopulation underrepresentation in clinical trials contributes to biases in clinical data and systemic healthcare inequities. We aim to evaluate reporting and representation, as well as the effect of geography and socioeconomic trends, in spinal oncology trials. Methods Data were collected from completed spinal oncology trials registered on ClinicalTrials.gov from 2000 to 2023. A total of 42 trials with 5679 participants were included. The demographics of participants were compared with national spinal tumor incidence data and demographic data from patients undergoing spinal oncology surgery at a quaternary care center. Results Only 50% of clinical trials reported race and 28.6% reported the ethnicity of participants, with privately funded trials less likely to report ethnicity (25% vs. 66.7%, P = .02574). When compared with their respective national incidences, Black (4.6% vs. 11.3%, P < .00001), American Indian or Alaska Native (0.2% vs. 0.6%, P = .00084), and Hispanic (4.7% vs 11.4%, P < .00001) patients were significantly underrepresented in trials. Black (4.6% vs. 18.9%, P < .00001) and female (44.5% vs. 48.9%, P = .00438) patients were also underrepresented when compared with the population of patients undergoing spinal oncology surgery. Trials post-2020 had increases in representation of several minority groups compared to pre-2020 trials. Trial sites were mostly located in metropolitan areas, with gaps in the Mountain region and parts of the Southern U.S. Conclusions There has been progress in diversifying spinal oncology trials, but there are still large racial, ethnic, and geographic disparities in the composition of clinical trial patients. Major reporting lapses hinder understanding the gaps in equitable enrollment.

  PublisherOA PDFDOI

• Optimizing Accelerated Protocols for Clinical T1w Vessel Wall Imaging: A Comparison of Deep Learning, Compressed Sensing, and CAIPI Methods

  Proceedings on CD-ROM - International Society for Magnetic Resonance in Medicine. Scientific Meeting and Exhibition/Proceedings of the International Society for Magnetic Resonance in Medicine, Scientific Meeting and Exhibition · 2025-09-16

  article

  Motivation: Clinical intracranial VWI requires high spatial resolution and effective blood/CSF suppression, leading to long scan times, motion artifacts, and increased patient burden. Goal(s): To evaluate optimized protocols for intracranial VWI using accelerated 3D T1w SPACE sequences. Approach: We evaluated DL, CS, and CAIPI acceleration on 20- and 64-channel coils to reduce scan times, comparing them to the clinical GRAPPA protocol. Results: The 64-channel CS protocol with 6x acceleration matched clinical image quality and SNR with reduced scan time. CS and DL sequences had similar image quality and SNR on the 64-channel coil, with DL showing better noise suppression on the 20-channel coil. Impact: This study optimizes intracranial VWI using 3D T1w SPACE with DL, CS, and CAIPI for 20- and 64-channel coils, achieving reduced scan times. The 64-channel CS (6x) matched clinical image quality, while DL provided superior noise suppression on 20-channel coils.

  PublisherDOI

• An Atlas of Neonatal Neurovascular Imaging Anatomy as Depicted with Microvascular Imaging: The Intracranial Arteries

  American Journal of Neuroradiology · 2025-04-22

  reviewOpen access

  Neurosonography is an invaluable diagnostic tool for assessing the neonatal brain and typically includes grayscale and conventional Doppler imaging. Microvascular imaging (MVI) is an emerging imaging technique that offers promising potential in evaluating neonatal intracranial pathology. MVI allows for sensitive and detailed assessment of neurovascular anatomy that can be challenging to visualize with other imaging modalities. This enhanced visualization not only improves the reliability in identifying neurovascular landmarks but also allows more accurate identification and assessment of adjacent parenchymal and extra-axial structures. Our review aims to offer an atlas of neurovascular structures as visualized with MVI, intended as a reference for future research in this field. Here, we focus specifically on the appearance of the intracranial arterial circulation as seen on MVI.

  PublisherDOI

• Added Value of Microvascular Imaging for the Diagnosis and Monitoring of Strokes in Newborns and Infants

  Pediatric Neurology · 2025-01-16

  article
  PublisherDOI

• Innovative Educational Program to Aid Clinical Vessel Wall MR Imaging Interpretation among Neuroradiologists

  American Journal of Neuroradiology · 2025-06-23

  articleOpen access

  Innovations that introduce new knowledge domains face greater barriers to adoption, often requiring investment in infrastructure, training/education, and cultural change. Sustaining and scaling an advanced clinical vessel wall MR imaging program requires technical resources and subspecialized neuroradiologists with advanced cerebrovascular expertise. A multifaceted educational program, including lectures, reporting templates, and an online resource, was implemented within a large academic Neuroradiology Division to address neuroradiology workforce readiness. Seven faculty "superusers" interested in cerebrovascular imaging were identified to facilitate case discussions and provide daily support for colleagues, clinicians, and MR technologists. Impact was assessed through a 12-month pre-/postintervention survey measuring confidence levels in evaluating vessel wall MR imaging examination appropriateness (a), assessing image quality (b), and diagnostic interpretations (c). Results showed division-wide increases in self-reported confidence and statistically significant increases among the superusers. These results show that a structured, expert-led peer-support model can enhance clinical readiness and sustain advanced imaging programs.

  PublisherDOI

• Radiologist-Validated Automatic Lumbar T1-Weighted Spinal MRI Segmentation Tool via an Attention U-Net Algorithm

  Diagnostics · 2025-11-28

  articleOpen access

  Background/Objectives: Spinal MRI segmentation has become increasingly important with the prevalence of disc herniation and vertebral injuries. Artificial intelligence can help orthopedic surgeons and radiologists automate the process of segmentation. Currently, there are few tools for T1-weighted spinal MRI segmentation, with most focusing on T2-weighted imaging. This paper focuses on creating an automatic lumbar spinal MRI segmentation tool for T1-weighted images using deep learning. Methods: An Attention U-Net was employed as the main algorithm because the architecture has shown success in other segmentation applications. Segmentation loss functions were compared, focusing on the difference between BCE and MSE loss. Two board-certified radiologists scored the output of the Attention U-Net versus four other algorithms to assess clinical relevance and segmentation accuracy. Results: The Attention U-Net achieved superior results, with SSIM and DICE coefficients of 0.998 and 0.93, outperforming other architectures. Both radiologists agreed that the Attention U-Net segmented lumbar spinal images with the highest accuracy on the Likert Scale (3.7 ± 0.82). Cohen’s Kappa coefficient was measured at 0.31, indicating a fair level of agreement. MSE loss outperformed BCE with respect to both SSIM and DICE, serving as the loss function of choice. Conclusions: Qualitative observations showed that the Attention U-Net and U-Net++ were the top performing networks. However, the Attention U-Net minimized external noise and focused on internal spinal preservation, demonstrating strong segmentation performance for T1-weighted lumbar spinal MRI.

  PublisherOA PDFDOI

Frequent coauthors

• Misun Hwang

  Children's Hospital of Philadelphia

  25 shared

• Sandra Saade‐Lemus

  Massachusetts General Hospital

  9 shared

• Arastoo Vossough

  Children's Hospital of Philadelphia

  6 shared

• Hao Huang

  Children's Hospital of Philadelphia

  6 shared

• Luis Octavio Tierradentro‐García

  Children's Hospital of Philadelphia

  6 shared

• Raymond W. Sze

  UCSF Benioff Children's Hospital

  6 shared

• Ray Norby

  University of Virginia

  4 shared

• P. Darvishi

  4 shared

Labs

• Radiology at the Hospital of the University of PennsylvaniaPI

Education

• MD

  University of Virginia School of Medicine

  2016

• Bachelor of Science, Biology

  Lebanon Valley College

  2011