Post

From Interaction to Data: Designing a Computable Dental Record System

From Interaction to Data: Designing a Computable Dental Record System

Introduction

Traditional odontograms focus on visual interaction.

However, in forensic identification systems operating at scale,
interaction alone is not sufficient.

The key question becomes:

How can a user interaction be transformed into structured, computable data?

This post explores the design approach for converting user input
into a consistent and queryable data representation.


Problem: Interaction Does Not Equal Data

A naive approach treats user interaction as simple UI events.

This approach is intuitive, but fundamentally limited:

  • It does not encode standardized meaning
  • It cannot be reliably compared across cases
  • It lacks consistency between different users

In forensic scenarios, where data must be analyzed and compared at scale,
this becomes a critical limitation.

In large-scale disaster situations, even small inconsistencies in records
can delay identification or introduce critical errors.

This is not merely a usability issue —
it is a problem of reliability.


Design Choice: A Minimal Data Structure

Each user interaction is represented using three components:

  • Tooth ID (FDI system)
  • Region (C, TL, TR, BR, BL)
  • Code (hierarchical classification)

This structure defines a minimal unit of information.

A localized finding with standardized meaning

By constraining input into this format,
the system ensures consistency without eliminating expressiveness.


From Interaction to Record

A user interaction follows a transformation process:

  • User Click
  • Region Selection
  • Code Assignment
  • Structured Record Generation

Each step reduces ambiguity and introduces semantic meaning.

The result is not a visual artifact,
but a data point that can be stored, queried, and compared.


Why Structure Matters (Trade-off Analysis)

There are multiple ways to represent user input:

Option 1: Free-form interaction

  • ✔ Flexible
  • ✘ Inconsistent
  • ✘ Difficult to compare across cases

Option 2: Fully constrained system

  • ✔ Highly consistent
  • ✔ Suitable for large-scale analysis
  • ✘ Reduced flexibility

Final Decision: Structured Interaction Model

The system adopts a structured interaction model.

This design makes a deliberate trade-off:

It sacrifices flexibility
in exchange for consistency, comparability, and computability

This balance is essential for systems that must operate reliably
across large datasets and multiple users.


System Implications

The structured model enables:

  • Standardized data input across users
  • Consistent encoding of dental findings
  • Reliable comparison between cases
  • Direct compatibility with algorithmic processing

Most importantly, it allows dental data to be:

queried, compared, and computed at scale.


Limitations

This approach has inherent limitations.

  • It depends on correct user input
  • Some complex conditions are simplified
  • Flexibility is reduced compared to free-form input

These limitations are intentionally accepted.

The goal is not to maximize input freedom,
but to ensure consistency and computational usability.


Future Extensions

The current design serves as a foundation.

Possible extensions include:

  • Validation mechanisms for input consistency
  • Integration with radiographic image analysis
  • Mapping structured data to machine learning models

These extensions build upon the same principle:
transforming interaction into computable data.


Conclusion

A digital odontogram should not only capture interaction —
it should encode meaning.

By transforming user input into structured data,
the system moves beyond visualization
and becomes a computational tool.

This transition is not optional,
but necessary for scalable and reliable forensic identification systems.

This post is licensed under CC BY 4.0 by the author.