Deep DivesApril 15, 2026

Before Dental AI, There Was Biogeneric

How Sirona taught CAD software to infer teeth from the patient—not just select them from a library.

ByNorbert Ulmer

Editor in Chief

Key Findings

• From selection to inference: Biogeneric shifted CAD/CAM dentistry from library-based tooth selection to patient-specific anatomical inference.
• Mathematical foundations: The technology relies on mathematical morphology and statistical modeling rather than modern neural networks.
• Contextual design: Multiple design modes (Individual, Copy, Reference) allowed software to interpret biological context from adjacent or opposing teeth.
• Workflow compression: Biogeneric significantly reduced chairside design time by providing anatomically plausible first drafts.
• AI lineage: Modern AI crown-design systems inherit Biogeneric's core ambition: generating restorations that are contextually appropriate rather than merely manufacturable.

The Old CAD Problem: A Tooth From the Shelf

Early dental CAD/CAM systems were revolutionary, but they were still constrained by a simple design logic. A user could select a tooth form from a library, scale it, adjust it, move it, stretch it, and manually refine the occlusion. This was digital, but it still carried the mindset of analog substitution: choose a form, then adapt it.

That approach worked, but it had obvious limitations. A library tooth is not the patient's tooth. It may be anatomically plausible, but it is not necessarily biologically or functionally native to the patient's arch. The dentist or technician still had to do the intellectual work of making the restoration fit the individual case.

Biogeneric changed the starting point. Instead of asking, "Which library tooth should we use?" it asked, "What does this patient's dentition suggest the missing tooth should look like?" That is a subtle difference. It is also the beginning of a new era.

The Scientific Move: Tooth Morphology Becomes Mathematics

The scientific foundation of Biogeneric is associated most closely with Albert Mehl, Volker Blanz, Reinhard Hickel, and related work in mathematical tooth morphology. Their research treated natural teeth not as isolated artistic forms but as measurable, comparable, computable structures. Stone replicas of natural molars could be scanned in three dimensions. Their surfaces could be mapped. Average forms could be calculated. Variations could be described statistically. Tooth morphology could become a mathematical object.

That mattered because once anatomy becomes computable, missing anatomy can be reconstructed. This is the key to understanding Biogeneric. It was not simply a smarter tooth library. It was an attempt to model the relationship between teeth, to find patterns in natural morphology, and to use those patterns to generate patient-specific restorations.

The language used in some early product materials could sound almost biological, suggesting a kind of "genetic blueprint" of morphology and occlusion. But the more careful interpretation is this: Biogeneric was based on statistical and geometric modeling of tooth form. It did not read DNA. It read shape.

The Commercial Turning Point

Biogeneric's roots appeared in research and early software development before it became a broader commercial design engine. By the late 2000s and early 2010s, Sirona began integrating Biogeneric more deeply into CEREC and inLab workflows.

The significant product moment came when Biogeneric moved beyond limited occlusal reconstruction and into broader restorative indications: crowns, veneers, inlays, onlays, bridges, and anatomically sized restorations.

For chairside dentistry, this meant CEREC users could generate more natural restoration proposals with less manual sculpting. For laboratories, inLab brought similar logic into the lab CAD environment, where technicians could begin with more anatomically informed proposals rather than static database forms.

This was not a minor software upgrade. It was a philosophical upgrade. The software was no longer merely a design surface. It was becoming a design participant.

The Different Biogeneric Modes Reveal the Deeper Idea

Biogeneric was not one single function. It evolved into several design modes, each answering the same underlying question in a slightly different way: where should the software get its anatomical signal?

Biogeneric Individual

Biogeneric Individual used the patient's own scanned anatomy to generate the restoration proposal. The software analyzed the preparation and surrounding dentition, especially neighboring teeth, to infer a suitable occlusal form. Its premise was straightforward: the mouth contains clues. If enough of those clues are captured, the missing tooth surface can be reconstructed in a way that is more patient-specific than a generic library tooth.

Biogeneric Copy

Biogeneric Copy was useful when the existing anatomy was worth preserving. If a pre-operative tooth, old crown, provisional, or wax-up had acceptable function or esthetics, the software could transfer parts of that occlusal surface into the new restoration and complete or enhance the rest. This matters clinically because not every restoration should be invented from scratch. Sometimes the best design strategy is preservation.

Biogeneric Reference

Biogeneric Reference allowed the user to select another tooth as the morphologic guide. This could be a contralateral tooth, an antagonist, or another suitable reference tooth of the same class. The design logic here was powerful: if the restoration site itself lacks enough information, borrow intelligence from another part of the patient's dentition.

Copy & Mirror and Jaw-Oriented Design

Later software logic extended this idea further, using symmetry, reference anatomy, and broader arch relationships. The design task became less about one isolated tooth and more about the tooth's place in a system. That evolution matters because modern AI crown design is moving in the same direction. The crown is not just a shape. It is a relationship: to the preparation, the margin, the neighbors, the opposing arch, the material, and the functional history of the patient.

Biogeneric Versus Modern AI

Modern AI crown-design systems use very different methods. Today's research explores generative adversarial networks, transformers, point-cloud models, mesh completion, biomechanical optimization, and large datasets of clinical scans and technician-designed restorations. These systems can learn patterns from data in ways Biogeneric did not.

Biogeneric, by contrast, was rooted in mathematical morphology, statistical modeling, and algorithmic reconstruction. It encoded relationships derived from natural tooth form. It did not behave like a modern neural network.

But the conceptual overlap is undeniable. Both Biogeneric and modern AI try to solve the same core problem: "Given the available patient-specific information, generate the missing restoration."

That is why Biogeneric belongs in the history of dental AI—not because it was deep learning, but because it reframed restorative CAD as inference.

The Real Legacy: Software Begins to Read the Mouth

The most important thing about Biogeneric is not the brand name. It is the question the software introduced into dentistry.

Before Biogeneric, CAD software largely helped users manipulate digital objects. With Biogeneric, the software began to interpret biological context. It looked at the mouth and made a proposal.

That may sound ordinary today, but it was not ordinary then. It marked a shift from CAD as drawing software to CAD as restorative intelligence. This is the lineage modern AI inherits.

When today's AI crown-design systems analyze a preparation, margin, adjacent teeth, antagonists, and prior technician designs, they are pursuing a more advanced version of the same ambition: to generate a restoration that is not merely manufacturable, but contextually appropriate.

Clinical and Industry Implications

For Dentists

For dentists, especially CEREC users, Biogeneric reduced the cognitive burden of chairside design. A restoration proposal that already looked anatomically plausible meant less time pushing cusps, dragging fissures, adjusting ridges, and manually shaping occlusion. The dentist still had to evaluate the proposal, but the starting point was better.

The core lesson is that digital design quality begins before the design phase. A better algorithm cannot compensate for poor input forever. Clean margins, complete scans, accurate bite records, and thoughtful pre-operative data remain essential. Dentists must focus on feeding the software the right clinical evidence, acting as an editor of machine-generated proposals.

For Dental Laboratories

For laboratories, Biogeneric foreshadowed the same structural shift labs are experiencing today with AI-assisted design. The technician's value did not disappear; it moved. Instead of spending as much time creating basic tooth morphology from scratch, the technician could increasingly focus on review, correction, esthetics, material constraints, contacts, occlusion, case planning, and exceptions.

Labs should measure software by correction time, not demo quality. The question is how much expert intervention is required before the design is clinically and manufacturably acceptable. Standardizing incoming scan requirements ensures that AI-assisted workflows receive the high-quality data they need to function effectively.

For Dental AI Companies and Manufacturers

For AI companies, Biogeneric serves as a historical benchmark. It proves that dentistry has been chasing patient-specific computational design for decades. A modern AI company must demonstrate how their systems move beyond statistical modeling into true learning. The real value lies in trustworthy automation: better first proposals, fewer corrections, shorter design cycles, lower remake risk, and clearer human oversight.

Limitations and What Remains Uncertain

While Biogeneric was a major step forward, its output depended heavily on scan quality, available anatomy, accurate bite records, proper model orientation, and the suitability of the selected design mode. If the neighboring anatomy was poor, the pre-operative scan was missing, or the occlusion was unstable, the algorithm still needed significant human direction.

Furthermore, occlusion cannot be reduced fully to morphology. A crown can look anatomically convincing and still require adjustment. Contact strength, dynamic movements, parafunction, material thickness, preparation design, and clinical context all matter. This highlights the ongoing necessity for human oversight in any automated restorative workflow.

Definitions

• Biogeneric: A CAD/CAM software feature developed by Sirona that uses mathematical morphology to infer and generate patient-specific tooth designs based on surrounding anatomy.
• Mathematical Morphology: A theory and technique for the analysis and processing of geometrical structures, used in dentistry to calculate average tooth forms and statistical variations.
• CAD/CAM: Computer-Aided Design and Computer-Aided Manufacturing, referring to software and machinery used to design and mill dental restorations.
• Generative AI: Artificial intelligence capable of generating text, images, or other data using generative models, often in response to prompts or contextual data.

Frequently Asked Questions

Citation-Ready Summary

Summary: Long before the advent of modern machine-learning workflows, Sirona's Biogeneric technology introduced a foundational shift in restorative CAD software by moving from static library-tooth selection to patient-specific anatomical inference. Using mathematical morphology and statistical modeling, Biogeneric treated tooth design as a prediction problem based on surrounding clinical context. This shift reduced routine design burdens for clinicians and technicians, foreshadowing the current era of AI-assisted crown generation where software serves as an active design participant rather than a passive drawing tool.

Internal Link Suggestions

• The Dental Lab Technician Isn't Disappearing. The Job Is Being Rewritten.
• The Dental Lab Won't Just Make Crowns. It Will Learn From Them.
• The Dental AI Race Isn't About Technology. It's About Who Controls the Workflow.

References / Source Notes

Core scientific foundation of Biogeneric

1. Mehl, A.; Blanz, V.; Hickel, R. "Biogeneric tooth: a new mathematical representation for tooth morphology in lower first molars." European Journal of Oral Sciences, 2005. Link
2. Mehl, A.; Blanz, V. "New procedure for fully automatic occlusal surface reconstruction by means of a biogeneric tooth model." International Journal of Computerized Dentistry, 2005. Link
3. Mehl, A. "A new mathematical process for the calculation of average forms of teeth." Journal of Prosthetic Dentistry, 2005. Link
4. Prof. Dr. Dr. Albert Mehl — University of Zurich publication profile. Link

Sirona / CEREC / inLab product-history sources

5. Dental Tribune. "CEREC Biogeneric: now for crowns, veneers and anatomical bridges." 2010. Link
6. Dental Tribune. "Sirona introduces latest inLab 3-D software upgrade V3.80." 2010. Link
7. DrBicuspid. "Sirona upgrades inLab software." August 3, 2010. Link
8. Dental Product Shopper. "inLab 3D Software Upgrade: V3.80." Link
9. Sirona / CEREC Biogeneric PDF: "CEREC Biogeneric: natural occlusions with just one click." Link

CEREC and Dentsply Sirona software manuals

10. Dentsply Sirona. CEREC Premium SW 4.5 Operator's Manual. Link
11. Dentsply Sirona. CEREC SW 4.5.x User's Manual. Link
12. Dentsply Sirona. CEREC Software 4.6.x Operator's Manual. Link
13. Dentsply Sirona. CEREC SW 5 Operator's Manual. Link
14. Sirona Dental CAD/CAM System CEREC SW 4.3 Manual. Link
15. Patterson Support. "Copy & Mirror Design Mode." Link

Clinical evaluation and morphology-comparison studies

16. Arslan, Y. et al. "Evaluation of biogeneric design techniques with CEREC CAD/CAM system." Journal of Advanced Prosthodontics, 2015. Link
17. Wang, F. et al. "Comparison and evaluation of the morphology of crowns generated by Biogeneric design modes." PLOS One, 2020. Link
18. Rim, K. et al. "Biogeneric tool to respect morphology and occlusion." Prosthodontics-related publication, 2024. Link

Patent sources

19. EP2363094B1 — Method, device, and program related to dental/tooth modeling. Link
20. EP2961344A1 / EP2961344B1 — Method for constructing dental surfaces of a dental prosthetic element and producing dental restorations. Link

Sirona / Dentsply Sirona corporate-history sources

21. Dentsply Sirona. "History." Link
22. Dentsply Sirona Investor Relations. "Dentsply Sirona Completes $14.5 Billion Merger of Equals." February 29, 2016. Link
23. Dental Tribune. "At age 37, CEREC advances the restorative capabilities of dentists as never before." 2022. Link

Modern AI crown-design comparison sources

24. Ding, H. et al. "Morphology and mechanical performance of dental crown designed by 3D-DCGAN." Dental Materials, 2023. Link
25. Hosseinimanesh, G. et al. "Personalized dental crown design: A point-to-mesh completion network." Medical Image Analysis, 2025. Link
26. Hosseinimanesh, G. et al. "From Mesh Completion to AI Designed Crown." arXiv, 2025. Link
27. Hosseinimanesh, G. et al. "Improving the quality of dental crown using a Transformer-based method." arXiv, 2023. Link
28. Wei, L. et al. "VBCD: A Voxel-Based Framework for Personalized Dental Crown Design." arXiv, 2025. Link
29. Wei, L. et al. "MADCrowner: Margin Aware Dental Crown Design with Template Deformation and Refinement." arXiv, 2026. Link
30. 2024 review: "AI-powered technologies in CAD/CAM restorative procedures." Link