Evaluating CT/Xray imaging-correlated impedance-based metrics for predictive surgical osseointegration and injury prevention

/ABSTRACT Clinical Importance: Almost 15 million dental implants were placed worldwide in 2021, with 3 million of these procedures taking place in the United States. Implants have been linked to maintaining proper nutrition intake, chewing ability, and quality of life, particularly essential for a rapidly aging population. The demand for implants is projected to increase up to 23% by 2026, with the largest rise expected among those aged 65-74 years – patients with the highest bone quality variability. The procedure itself requires an osteotomy be drilled into the mandible or maxilla (depending on site of implant) so that the implant can be seated within the boney structures of the jaw. A significant challenge to creating the initial osteotomy is ensuring that no critical structures (i.e., nerve or sinus) near to the bone are breached leading to severe post-procedure morbidities. Further, understanding patient- specific bone integrity at the implant site is critical for proper implant selection and use of adjunctive procedures such as bone densification to ensure successful osseointegration and long-term stability. Clinical Limitations: Despite the widespread use of cone beam CT (CBCT) for preoperative anatomic assessment, significant challenges remain in accurately placing the implant and quantifying bone quality from Hounsfield (HU) measurements. Specifically, CBCT struggles to visualize boney microstrucutres, particularly in trabecular cancellous bone which has increasingly been recognized as critical for implant loading, stability and overall osseointegration. Further, while pre-operative imaging and use of guidance has helped significantly improve surgical outcomes, iatrogenic injury to high-risk structures still occurs in ~10-20% of the cases, resulting in significant short- and long-term morbidity. Specific Objectives: This proposal aims to leverage a novel multimodal dataset uniquely integrating radiomics, "impediomics" and intraoperative surgical tracking data. Features from more 600 unique drill depths within 105 osteotomies from an in vivo large animal model yielded a >20k impedance feature dataset with novel insights including ultra-high resolution micro-CT scans of cortical and trabecular bone microarchitecture, impedance-based electrical property features collected in real-time while drilling, and intentional breaches of high-risk nerve and sinus structures. This dataset presents a unique opportunity to evaluate novel potential imaging-correlated impedance-based metrics for predictive surgical osseointegration and injury prevention. We aim to 1) expand this dataset through additional advanced processing and feature extraction to derive impedance-based "impediomics” and radiomics features (Aim 1), 2) explore machine learning models, including multi-feature approaches to detect high-risk anatomic interfaces that could provide real-time feedback during osteotomy to mitigate iatrogenic injury risk (Aim 2), and 3) leverage our novel impedance features and unique ultrahigh-resolution micro-CT images to develop potential metrics for precise site-specific bone quality characterization at prospective implant sites. Such a capability could be transformative for delivering personalized treatment plans that maximize outcomes while mitigating complications for dental implant patients. Project Number: 1R03DE034797-01 | Fiscal Year: 2025 | NIH Institute/Center: National Institute of Dental and Craniofacial Research (NIDCR) | Principal Investigator: Ryan Halter | Institution: RYTEK MEDICAL, INC., LEBANON, NH | Award Amount: $220,000 | Activity Code: R03 | Study Section: Special Emphasis Panel[ZDE1 TO (04)] View on NIH RePORTER: https://reporter.nih.gov/project-details/11120588