NovaScan Announces Publication in Nature Portfolio Journal Demonstrating Rapid AI-Guided Margin Assessment in Mohs Surgery

Published in npj Biomedical Innovations, the peer-reviewed study reports strong performance for NovaScan's high-frequency bioelectrical impedance platform in mapping residual basal cell carcinoma at the surgical margin

CHICAGO, IL — August 17, 2026 — NovaScan, Inc., a clinical-stage oncology device company developing point-of-care cancer assessment technology, today announced the publication of a peer-reviewed study in npj Biomedical Innovations, a Nature Portfolio journal. The article, titled "Machine learning-guided bioelectrical impedance mapping for rapid adjunctive margin assessment in Mohs surgery", evaluates NovaScan's high-frequency bioelectrical impedance spectroscopy platform, coupled with supervised machine learning, for rapid, non-destructive margin assessment in Mohs micrographic surgery.

The study assessed excised dermal specimens from patients undergoing Mohs surgery for nodular basal cell carcinoma. Using NovaScan's technology, impedance measurements were translated into spatial probability heat maps designed to estimate the risk of cancer-positive tissue at the surgical margin. The technology is intended to function as an adjunct to histopathology, providing earlier surgical guidance while preserving frozen section pathology as the definitive standard for margin clearance.

"This publication represents an important milestone for NovaScan and for MarginScan™," said Craig Davis, Chief Executive Officer of NovaScan. "Mohs surgery is highly effective, but the current staging loop depends on frozen section processing and interpretation, which increases waiting time, workflow variability, and bottlenecks in high-volume practices. Our goal is not to replace histopathology, but to provide surgeons with rapid, accurate and actionable information earlier in the procedure, while preserving the existing standard of care."

The study demonstrated excellent diagnostic performance. Under spatial-tolerance scoring designed to account for boundary-level uncertainty in translating histology findings onto the electrode grid, MarginScan achieved a mean ROC AUC of 0.993 ± 0.003, with 95.6% ± 4.7% sensitivity and 96.1% ± 5.0% specificity. Median acquisition time was 3.7 minutes per specimen, with the authors concluding that the technology can be integrated into the Mohs staging workflow without materially extending the grossing-to-histology interval.

"MarginScan was designed to provide a specimen-referenced map that can be interpreted rapidly and intuitively within the clinical workflow," said Donato M. Ceres, Ph.D., Chief Scientific Officer of NovaScan and lead author of the publication. "These results demonstrate the potential of combining high-frequency bioelectrical impedance spectroscopy with AI to provide rapid, spatially resolved information that may improve intraoperative decision-making."

"Mohs surgery is built around complete margin assessment and tissue preservation, but the process is also highly dependent on timing, coordination, and efficient use of the histology workflow," said Manish J. Gharia, M.D., fellowship-trained Mohs surgeon and co-author of the publication. "A rapid, non-destructive tool that provides earlier spatial information about where residual tumor is most likely to be present could be clinically valuable if validated in larger, multi-center studies. Importantly, this approach preserves frozen section histopathology as the definitive standard while potentially giving the surgeon useful information earlier in the staging cycle."

Mohs micrographic surgery remains the gold standard for treating many high-risk and cosmetically sensitive non-melanoma skin cancers. While frozen section histopathology provides definitive margin assessment, processing and interpretation require specialized procedures and infrastructure that result in extended and often unnecessary patient wait times, subsequently contributing to clinic bottlenecks. Because MarginScan is non-destructive, the excised specimen proceeds through the standard frozen section process after scanning. The technology is designed to provide actionable information earlier in the staging cycle while preserving the current standard of care.

"This work provides a strong foundation for NovaScan's clinical and regulatory path in dermatology," said Bob Cook, NovaScan Chief Operating Officer and former Senior Vice President at Castle Biosciences. "The study supports a compelling clinical and operational use case in Mohs surgery while reinforcing the broader potential of NovaScan's platform for rapid point-of-care cancer assessment across multiple tissue types and oncology applications."

NovaScan is preparing for additional clinical studies in non-melanoma skin cancer and is advancing toward a pivotal trial and FDA submission for MarginScan. The company is also developing applications of its platform beyond dermatology, including breast-conserving surgery, supported by a $3.7 million CPRIT Product Development Award.

MarginScan is an investigational device and has not been cleared or approved by the U.S. Food and Drug Administration.

About NovaScan

NovaScan is a clinical-stage oncology device company developing a platform technology that combines high-frequency spectral bioimpedance with AI to provide rapid, accurate cancer assessment at the point of care. NovaScan's lead product, MarginScan™, is being developed for rapid intraoperative margin assessment, with an initial commercial application in non-melanoma skin cancer. The company's broader platform has potential applications across multiple tissue types and oncology settings, including breast-conserving surgery.

Publication

Ceres, D.M., Gharia, M.J., Harvey, K. & Quinn, N. "Machine learning-guided bioelectrical impedance mapping for rapid adjunctive margin assessment in Mohs surgery." npj Biomedical Innovations (2026). DOI: 10.1038/s44385-026-00095-5.

Media Contact

NovaScan, Inc.
Craig Davis
cdavis@novascaninc.com