Artificial Intelligence Predicts Patient Response to Peri-Implantitis Therapy

A machine-learning algorithm may be able to assess how patients will respond to treatment for peri-implantitis. A study led by University of Michigan School of Dentistry researchers used a form of artificial intelligence to predict treatment outcomes among this patient group. The investigators used Fast and Robust Deconvolution of Expression Profiles (FARDEEP technology) to analyze tissue samples of 24 patients with peri-implantitis who were undergoing regenerative therapy. The findings showed that low-risk patients exhibited elevated M1/M2 ratios and lower B cell infiltration. Additionally, the team reports that Fusobacterium nucleatum and Prevotella intermedia “were significantly enriched in high-risk individuals.”

Yu Leo Lei, DDS, PhD, senior author and an assistant professor at University of Michigan School of Dentistry, says researchers were surprised to discover the cells associated with more promising outcomes for patients with peri-implantitis were not the cells more adept at tissue repair and wound healing. Rather, this proof-of-concept study shows that “immune cell types that are central to microbial control are strongly correlated with superior clinical outcomes,” he explains. 

“While surgical management can reduce bacterial burdens across all patients, only those with more immune cell subtypes for bacterial control can suppress the recolonization of pathogenic bacteria and show better regenerative outcomes,” Lei notes. The study, “Machine Learning-Assisted Immune Profiling Stratifies Peri-Implantitis Patients With Unique Microbial Colonization and Clinical Outcomes,” appeared in Theranostics.