Periodontitis is a globally prevalent inflammatory oral disease, affecting approximately 50% of the population worldwide and imposing a substantial burden on patients′ health and quality of life. Early and accurate diagnosis is critical for preventing disease progression; however, conventional diagnostic approaches often rely on subjective clinical assessments, which only primarily evaluate the cumulative state of the disease, thus limiting their ability to achieve precise early detection. In recent years, the rapid advancement of artificial intelligence (AI) in medical diagnostics has demonstrated significant promise, particularly through the integration of multimodal data to enable more comprehensive information capture and analysis. Multimodal data fusion, which combines diverse inputs such as imaging, clinical parameters, and biomarkers, offers transformative potential for AI-powered periodontitis diagnostics. This innovative approach aims to overcome the limitations of traditional methods, significantly enhancing diagnostic accuracy and predictive capabilities. This manuscript reviews the primary diagnostic techniques for periodontitis, explores recent advances in AI applications within this domain, and emphasizes the potential of multimodal data in facilitating precision diagnosis. Furthermore, it provides new insights and supports for personalized treatment strategies.

Periodontitis is a globally prevalent inflammatory oral disease, affecting approximately 50% of the population worldwide and imposing a substantial burden on patients′ health and quality of life. Early and accurate diagnosis is critical for preventing disease progression; however, conventional diagnostic approaches often rely on subjective clinical assessments, which only primarily evaluate the cumulative state of the disease, thus limiting their ability to achieve precise early detection. In recent years, the rapid advancement of artificial intelligence (AI) in medical diagnostics has demonstrated significant promise, particularly through the integration of multimodal data to enable more comprehensive information capture and analysis. Multimodal data fusion, which combines diverse inputs such as imaging, clinical parameters, and biomarkers, offers transformative potential for AI-powered periodontitis diagnostics. This innovative approach aims to overcome the limitations of traditional methods, significantly enhancing diagnostic accuracy and predictive capabilities. This manuscript reviews the primary diagnostic techniques for periodontitis, explores recent advances in AI applications within this domain, and emphasizes the potential of multimodal data in facilitating precision diagnosis. Furthermore, it provides new insights and supports for personalized treatment strategies.

Based on the transcriptome analysis data of a Bacillus licheniformis, a novel bidirectional promoter was identified from the strain and its transcriptional strength was analyzed. The expression level of a Bacillus clausii derived alkaline protease gene driven by the bidirectional promoter was studied by using the known strong constitutive promoter pShuttle-09 as a control. Three recombinant expression vectors and the corresponding recombinant bacteria were constructed. Under the control of the new promoter pLA and its reverse promoter pLB, the alkaline protease expression level respectively reached 164 U/mL and 111 U/mL. The results indicated that the transcription strength of pLA was significantly higher than that of pShuttle-09 and pLB, and both the pLA and pLB promoters could initiate the expression of the alkaline protease. Thus, it provides a new expression element for the heterogenous genes in Bacillus sp. and a new idea for the co-expression of two genes in one prokaryotic strain.
Bacillus subtilis ; Promoter Regions, Genetic