Dental diagnose on radiography
We announced our radiographs diagnose pre-train module
Abstract
Announce our radiography diagnose Pre-trained model
We spent 5 months training our first model for oral diagnostics, We totally used 3k+ panoramic, 3k+ periapical, and now deployed it to a separate server for experience.
Experience dashboard Panoramic on diagnose
Currently recognizable morphology 
Currently recognizable morphology - Tooth position
- Tooth loss(missing)
- Sinus
- Nerve
- Clinical attachment loss( CAL )
- Radiography bone loss( RBL )
- Cement enamel junction( CEJ )
- Furcation involvements
- Hypercementosis
- Curved tooth root
- Caries
- Chipped
- Apical lesion
- Root canal
- Crown
- Bridge
- Filling
- Enamel pearl
- Wisdom tooth
- Impacted tooth
- Peri-implantitis
- Implant
Periapical on diagnose
Currently recognizable morphology 
Currently recognizable morphology - Maxilla
- Mandible
- Tooth position
- Tooth loss(missing)
- Clinical attachment loss( CAL )
- Radiography bone loss( RBL )
- Cement enamel junction( CEJ )
- Furcation involvements
- Hypercementosis
- Curved tooth root
- Caries
- Chipped
- Apical lesion
- Root canal
- Crown
- Bridge
- Composite
- Amalgam
- Wisdom tooth
- Impacted tooth
- Peri-implantitis
- Implant
Conclusion
Accuracy is something that we’ve been working on for a while now, and our goal is to be able to recognize more morphology with higher accuracy, and as of now, the morphology accuracy for both panoramic and apical is above 92%.
Our mission and vision is to use AI efficiency dental industry practices, AI cannot replace humans (it never has). This is just our first step, and we will provide more AI integrations in the future.
Experience
We welcome you to experience our pre-trained models https://experience.origintechx.ai