Skip to main content
2024 AAOMS 106th Annual Meeting Main banner
Icon Legend
This session is not in your schedule.
This session is in your schedule. Click again to remove it.
Presentation Icons
Ticketed <br><i>This session/event requires a ticket.</i>
Ticketed
This session/event requires a ticket.
In-person only<br><i>This session is offered in-person only.</i>
In-person only
This session is offered in-person only.
On-demand<br><i>On-demand session available beginning Sept. 9.</i>
On-demand
On-demand session available beginning Sept. 9.
Recorded<br><i>This session will be recorded and available beginning Sept. 30.</i>
Recorded
This session will be recorded and available beginning Sept. 30.
Sold Out<br><i>This ticketed session/event is sold out.</i>
Sold Out
This ticketed session/event is sold out.
2024 Oral Abstract Award
2024 Oral Abstract Award
Poster Icons
Poster Winner
Poster Winner
Back

Trauma Management

The Utilization of Artificial Intelligence [AI] in Identifying Maxillofacial Fractures From CT Scan Imaging

(75) The Utilization Of Artificial Intelligence [AI] In Identifying Maxillofacial Fractures From CT Scan Imaging

Thursday, September 12, 2024
2:30 PM – 4:00 PM EDT

    Co-Author(s)

  • Waleed Zaid, DDS, MSc, FRCDC, FACS (he/him/his)

    Associate Professor
    Louisiana State University Health Sciences Center New Orleans
    New Orleans, Louisiana, United States

    • Poster Presenter(s)

  • James Vegrzyn, DMD, MD, MBA

    Resident
    Louisiana State University Health Sciences Center New Orleans
    New Orleans, Louisiana, United States

    • Co-Author(s)

  • RA

    Rami Abdulla, n/a

    Student
    Louisiana State University Health Sciences Center New Orleans
    New Orleans, Louisiana, United States

  • AS

    Anne Scruggs, n/a

    Resident
    Louisiana State University Health Sciences Center New Orleans
    New Orleans, Louisiana, United States

Disclosure(s):
James Vegrzyn, DMD, MD, MBA: No financial relationships to disclose
Abstract: Maxillofacial fractures are a prevalent cause for hospital admissions worldwide. Computed tomography (CT) scanning is an indispensable diagnostic tool in assessing maxillofacial trauma. Traditionally, neuroradiologists evaluate CT scans to delineate the presence and extent of maxillofacial fractures and serve a critical role in surgical treatment planning. However, the scarcity of neuroradiologists and oral and maxillofacial surgeons, particularly in remote regions, necessitates patient transfers to more specialized facilities, placing a strain on healthcare resources.

This study investigates the efficacy of an artificial intelligence (AI) model in accurately diagnosing maxillofacial fractures, positing that AI can significantly improve the capabilities of general radiologists in detecting fractures resulting from maxillofacial trauma.

We conducted a single-center retrospective analysis of the trauma registry from Our Lady of the Lake Regional Medical Center in Baton Rouge, Louisiana, of maxillofacial trauma patients from July 2019 to July 2022. Our population of interest includes patients with isolated mandible fractures. Patients with concomitant facial fractures were excluded.

Patient demographics were recorded and maxillofacial Digital Imaging and Communications in Medicine (DICOM) scans anonymized. The AI model, leveraging SqueezeNet architecture, was trained on de-identified DICOM data to discern isolated mandible fractures using 2D images derived from 3D reconstructions. For unilateral fractures, the contralateral uninjured side served as a control.

The diagnostic ability of the AI model was gauged against the gold standard of neuroradiologist interpretations of the maxillofacial CT scans. From a cohort of 95 patients with neuroradiologist-confirmed isolated mandible fractures, the AI algorithm demonstrated an impressive 86%-90% accuracy rate in fracture detection.

Our findings suggest that AI can significantly enhance the diagnostic process for maxillofacial fractures, providing invaluable support to general radiologists. We advocate for expanded research encompassing the full spectrum of facial skeletal injuries to fully realize the potential of AI in this domain.

References

Artificial Intelligence: The Future of Maxillofacial Prognosis and Diagnosis?

Rekawek, Peter et al. Journal of Oral and Maxillofacial Surgery, Volume 79, Issue 7, 1396 – 1397

Use of Artificial Neural Network in Differentiation of Subgroups of Temporomandibular Internal Derangements: A Preliminary StudyBas, Burcu et al. Journal of Oral and Maxillofacial Surgery, Volume 70, Issue 1, 51 - 59