The Missing Knowledge Layer
for Surgical AI
SurgicalDataOS transforms cataract surgery videos into machine-readable knowledge that powers AI, robotics, simulation, research and next-generation ophthalmic intelligence.
A representation framework for machine-understandable cataract surgery.
The Problem
Current Surgical AI is trained on pixels.
Not on surgical knowledge.
Object detection and segmentation tell a model where things appear in a frame. They do not encode why an action occurs, when a phase transitions, or how instrument motion relates to tissue response. Robotic cataract surgery demands causal, temporal, and intent-aware representations — not bounding boxes alone.
- Detects objects in isolated frames
- No temporal or causal structure
- Cannot reason about surgical intent
- Fails under domain shift and occlusion
- Representation of operative semantics
- Temporal, causal, and intent-aware reasoning
- Enables robotic planning and skill assessment
- Generalizes across surgeons and equipment
From Surgical Video to Machine Knowledge
What is SurgicalDataOS?
Every procedure is represented as a sequence of Machine Knowledge Objects (MKOs)—computational representations that preserve what the surgeon observed, interpreted, decided and performed. Together these MKOs form a machine-readable Knowledge Graph capable of supporting explainable AI, robotics, simulation, education and collaborative research.
Relational Knowledge Model
Knowledge Objects Connected by Typed Relationships
Operative video informs structured Knowledge Objects. In a full representation, each object links to others through typed relationships — temporal, causal, hierarchical, and more — preserving procedural and clinical reasoning.
- Temporal
- Causal
- State-transition
- Hierarchical
- Dependency
- Alternative-strategy
- Corrective / revision
- Evidence / provenance
Interactive graphs on this site are conceptual illustrations only. They are not complete clinical knowledge graphs or validated operative models.
What Relationships Preserve
Relationships between Knowledge Objects carry procedural, causal, and clinical meaning — not labels alone.
ObservationInterpretation
What is seen is assigned clinical significance.
InterpretationDecision
Meaning informs the chosen operative strategy.
DecisionAction
Strategy becomes an executed intervention.
ActionOutcome
The immediate operative state produced by the act.
EventCorrective action
A transition triggers a deliberate surgical response.
Knowledge ObjectEvidence / provenance
Each object remains traceable to video, author, and review history.
Alternative actionDecision context
Clinically valid branches can coexist from the same state.
AI-Assisted, Surgeon-Validated Capture
Knowledge is proposed by AI, confirmed by surgeons, and stored as linked Knowledge Objects — not generated autonomously.
01
AI proposes
AI analyzes operative video and proposes structured annotations and candidate Knowledge Objects.
02
Surgeon reviews
The operating surgeon reviews, corrects, and enriches proposals with clinical context and meaning.
03
Knowledge Object validated
Confirmed information is organized into a Knowledge Object with observation, interpretation, decision, action, and outcome.
04
Graph enriched
Validated objects connect through typed relationships, preserving sequence, causality, and surgical reasoning.
Beyond Computer Vision
Existing datasets describe what is visible.
SurgicalDataOS represents what is happening.
Computer vision identifies objects in individual frames. Operative intelligence requires temporal events, anatomical relationships, instrument interactions, intent and procedural decision making. SurgicalDataOS transforms video into machine-understandable knowledge.
Traditional Computer Vision
Visual Perception
Where is it?
- Detection
- Segmentation
- Tracking
- Classification
- Bounding Boxes
Describes pixels and objects within individual frames.
Knowledge Objects
Surgical Understanding
What is happening?
- Phase
- Action
- Instrument
- Anatomy
- Tissue
- Complication
Represents surgical workflow, anatomical context and procedural meaning.
Knowledge Layer
Machine Reasoning
What should happen next?
- Knowledge Graph
- Decision Layer
- Surgical Context
- Robot-ready Representation
- Foundation Model Input
Enables reasoning, planning and robotic execution.
Knowledge Object Explorer
Knowledge Object Explorer
See how SurgicalDataOS transforms a single surgical event into structured machine-readable knowledge.
Reference Video
Title
Phaco Fixation Established
Representative Frame
40
Frame Range
0–82
Timestamp Range
0.00–1.37 seconds
Phase
Nucleus Management
Stage
Fragmentation
Observation
The phaco tip is embedded within the nucleus, providing stable fixation. The nucleus remains intact, and the chopper has not yet initiated the chopping manoeuvre.
Interpretation
Stable fixation has been achieved, creating the mechanical conditions required for controlled advancement of the chopper toward the equator.
Decision
Maintain secure nuclear fixation while preparing to position the second instrument for nucleus fragmentation.
Action
The phaco tip maintains stable purchase on the nucleus while fixation is preserved.
Event
A stable mechanical relationship is established between the nucleus and the phaco tip, preparing the nucleus for controlled fragmentation.
Outcome
The nucleus is securely stabilised, allowing safe progression to the chopping phase.
Cognitive Intent
Create a stable mechanical foundation from which a controlled and reproducible nuclear fracture can be initiated.
Knowledge Extract
Successful vertical chop begins with secure nuclear fixation. Stable fixation is the prerequisite that enables controlled force transmission during the subsequent chopping manoeuvre.
Reference Dataset
Title
Nucleus Fragmentation
Procedure
Phacoemulsification
Technique
Vertical Chop
Duration
13.7
FPS
60
Current Knowledge Object Number
Knowledge Object 1 of 7
Knowledge Graph
From Knowledge Objects to Knowledge Graphs
Every Knowledge Object becomes a connected node within the SurgicalDataOS ontology, linking surgical observations, decisions, actions, tissues, instruments and outcomes into a machine-readable graph for AI, robotics and simulation.
Conceptual illustration only: this interactive graph is not a complete clinical knowledge graph, a validated operative model, or a production clinical or autonomous-surgery system.
Applications
One knowledge layer, many frontiers
Structured Knowledge Objects may support downstream uses across education, research, and future computational systems — the platform is not an annotation product.
Artificial Intelligence
Validated Knowledge Objects may supply clinically grounded material for computer vision, multimodal models, and surgical reasoning research.
Robotic Surgery
Structured knowledge may inform future research into assistance and shared autonomy — not autonomous or regulatory-ready systems on its own.
Simulation
Simulation environments may incorporate procedural sequence together with documented interpretation and decision context.
Surgical Education
Teaching and peer learning from linked observation, interpretation, decision, action, and outcome.
Research
A common Knowledge Object framework for reproducible studies of technique, workflow, and outcomes.
Clinical Decision Support
Future systems may draw on structured surgical knowledge for explanation and traceability — subject to clinical validation.
Future Platform
Infrastructure for knowledge at scale
Annotation Studio
Multi-level labeling with AI assist, consensus workflows, and real-time validation against the representation schema.
Dataset Marketplace
Discover, license, and version curated cataract datasets with full provenance and quality metrics.
Knowledge Graph Explorer
Traverse surgical entities, query temporal relationships, and export subgraphs for model training.
Validation Dashboard
Inter-annotator agreement, schema compliance, and automated quality gates before dataset release.
API
Programmatic access to annotations, graph queries, and streaming video pipelines for research integration.
Scope
SurgicalDataOS is a conceptual and technical demonstrator for converting operative video into structured, surgeon-validated Knowledge Objects. References to robotics, autonomous surgery, foundation models, or clinical decision support describe potential downstream uses of structured knowledge — not current validated clinical products, autonomous control, or regulatory-ready systems. These capabilities would require independent technical, clinical, safety, and governance evidence.
About
“The knowledge layer for machine-understandable surgery.”
SurgicalDataOS was initiated by Dr. Merine Paul, a practicing ophthalmologist and cataract surgeon from India, with a clinical interest in how surgical expertise, decision-making, and operative experience can be represented in forms that support computational intelligence.
The initiative explores how surgical knowledge can be structured for applications in surgical AI, education, simulation, research, and future robotic systems.
SurgicalDataOS is currently an independent research initiative and working demonstrator.
Collaboration
Building Surgical Intelligence Together
SurgicalDataOS is an open research initiative that welcomes collaboration with surgeons, AI researchers, robotics companies, academic laboratories and industry partners interested in advancing machine-understandable intelligence. We believe the future of surgical AI will be built through open scientific collaboration, shared representations and rigorous validation.
Contact
Build the knowledge layer with us
Whether you are advancing AI research, developing robotic platforms, or curating clinical datasets — we want to hear from you.