CranioCatch
Step-by-step implementation guide — pre-implementation checklist, onboarding, staff training, go-live runbook, and ROI tracking.
CranioCatch — Implementation Playbook (DSO)
CranioCatch Implementation Playbook for DSOs
Diagnostic Imaging AI for Multi-Location Dental Organizations
1. Executive Summary
What CranioCatch Does
CranioCatch is an FDA-cleared AI-powered diagnostic imaging platform that automatically analyzes dental radiographs (panoramic, periapical, bitewing, and CBCT) to detect pathologies, anatomical structures, and treatment-relevant findings in seconds. The software overlays AI-detected annotations directly onto images, providing dentists with a comprehensive diagnostic second opinion that highlights caries, periapical lesions, bone loss, impacted teeth, and over 70 additional detectable conditions.
Why DSOs Benefit from Diagnostic Imaging AI
Deploying diagnostic AI at scale delivers compounding advantages unavailable to single-location practices:
Standardization of Diagnostic Quality: Across 15–50 locations, diagnostic accuracy varies based on provider experience, fatigue, and time pressure. CranioCatch establishes a consistent diagnostic baseline, reducing the "provider lottery" effect that impacts patient outcomes and case acceptance variability.
Data Aggregation for Strategic Decision-Making: Centralized AI deployment generates enterprise-wide insights—prevalence patterns by region, undiagnosed pathology rates, provider-level detection benchmarks—enabling evidence-based resource allocation and quality improvement initiatives.
Scalable Training Accelerator: New associates and recent graduates benefit from AI-assisted learning across all locations simultaneously, reducing ramp-up time to full productivity and improving retention.
Liability Risk Reduction: Documented AI-assisted diagnosis creates an additional layer of defensibility and demonstrates standard-of-care adherence across your portfolio.
Case Acceptance Improvement: Patients presented with visual AI annotations showing detected pathology demonstrate higher treatment acceptance rates—a margin improvement that compounds across your patient volume.
Expected Timeline: Decision to Full Deployment
| DSO Size | Pilot Phase | Full Deployment | Total Timeline |
|---|---|---|---|
| 15–25 locations | 4–6 weeks | 10–14 weeks | 14–20 weeks |
| 26–40 locations | 6–8 weeks | 14–18 weeks | 20–26 weeks |
| 41–50 locations | 6–8 weeks | 18–24 weeks | 24–32 weeks |
Timeline assumes adequate IT infrastructure and no major PMS migration occurring simultaneously.
2. Pre-Implementation Checklist (Weeks 1–2)
Technical Requirements
Hardware Requirements (Per Location)
☐ Workstations: Windows 10/11 Pro, minimum 8GB RAM, i5 processor or equivalent (16GB RAM recommended for CBCT analysis)
☐ Display: 1920x1080 resolution minimum; medical-grade monitors not required but recommended for optimal visualization
☐ Imaging sensors: Existing digital sensors compatible (CranioCatch is sensor-agnostic)
Network Requirements
☐ Minimum 50 Mbps download/10 Mbps upload per location for cloud-based deployment
☐ Stable internet connectivity with <100ms latency to cloud endpoints
☐ Firewall configuration allowing outbound HTTPS (port 443) to CranioCatch cloud services
⚠️ Locations with satellite internet or rural connectivity often experience timeout issues—flag for Wave 3 or later
Software Requirements
☐ Compatible imaging software (see Integration section)
☐ Current browser: Chrome 90+, Edge 90+, or Firefox 88+
☐ PDF viewer for report export functionality
Enterprise-Level Requirements
Network Standards Across Locations
☐ Document current network architecture per location (direct internet vs. MPLS/SD-WAN)
☐ Identify locations routing through centralized data centers—these may require additional firewall rules
☐ Verify VPN configurations don't block required endpoints
🔵 Request CranioCatch IP whitelist documentation for enterprise firewall configuration
Hosting Model Decision
🟣 Executive Decision Required: Cloud vs. Hybrid Deployment
| Factor | Cloud-Only | Hybrid (Edge + Cloud) |
|---|---|---|
| Upfront cost | Lower | Higher |
| Ongoing cost | Subscription-based | Licensing + maintenance |
| Network dependency | High | Moderate |
| CBCT processing speed | Cloud-dependent | Faster local processing |
| Data sovereignty | Vendor-managed | Greater control |
Recommendation for most DSOs: Cloud-only deployment with hybrid evaluation for high-volume CBCT locations after initial rollout stabilizes.
Identity Management
☐ Confirm SSO compatibility (SAML 2.0 supported)
🔵 Provide CranioCatch with Identity Provider metadata
☐ Define role-based access control structure (Admin, Provider, Viewer)
☐ Establish centralized credentialing process—avoid per-location account creation
Centralized Administration
☐ Designate enterprise administrator accounts (recommend 2–3 individuals)
☐ Define location-level admin permissions vs. central admin permissions
☐ Establish audit log access protocols for compliance team
Vendor Onboarding Steps
Key Contacts to Establish
| Role | Contact Purpose | Expected Response Time |
|---|---|---|
| Enterprise Account Manager | Commercial terms, escalations, renewals | Same business day |
| Technical Implementation Lead | Integration, configuration, troubleshooting | 2–4 hours during rollout |
| Training Coordinator | Training scheduling, materials access | 24 hours |
| Support Desk (Tier 1) | Day-to-day technical issues | 1–2 hours |
| Clinical Success Manager | Workflow optimization, adoption metrics | Weekly cadence |
🔵 Vendor Onboarding Checklist:
☐ Schedule kickoff call with CranioCatch enterprise team (Week 1, Day 1–2)
☐ Execute Master Services Agreement with BAA addendum
☐ Complete security questionnaire exchange (your InfoSec team → CranioCatch, and reverse)
☐ Establish Slack/Teams channel or ticketing system integration for ongoing communication
☐ Receive dedicated implementation project plan from CranioCatch
☐ Confirm escalation matrix and after-hours support availability
Data/Access Prerequisites
☐ Compile location list with addresses, time zones, and primary contacts
☐ Document Practice Management System (PMS) and version per location
☐ Document imaging software and version per location
☐ Inventory CBCT units by manufacturer/model (if applicable)
☐ Identify imaging archive storage method (local, cloud, third-party PACS)
🔵 Obtain CranioCatch API credentials for test environment
☐ Prepare sample image sets from 3–5 representative locations for integration testing
☐ Verify DICOM export capability from imaging systems
Internal Stakeholder Alignment
Stakeholder Alignment Map
| Stakeholder Level | Role | Involvement Type | Timing |
|---|---|---|---|
| Board/Investors | Awareness, ROI validation | Quarterly updates | Pre-decision, post-Wave 1 |
| C-Suite (CEO, CFO, CDO) | Budget approval, strategic alignment | Decision-making | Weeks 0–1 |
| VP of Operations | Implementation ownership | Accountable | Throughout |
| Chief Dental Officer | Clinical validation, provider communication | Responsible | Weeks 1–8 |
| IT Director | Technical architecture, security | Responsible | Weeks 1–4 |
| Regional Managers | Location coordination, change management | Informed → Responsible | Week 2 onward |
| Office Managers | Local implementation, staff coordination | Responsible | Week 3 onward |
| Lead Providers | Clinical adoption, peer influence | Consulted | Week 2 onward |
| Compliance/Legal | HIPAA, BAA, liability review | Consulted | Week 1 |
Approval Requirements
🟣 Approvals to Secure Before Proceeding:
☐ Budget approval for licensing, implementation, and training costs (CFO)
☐ Clinical workflow approval and liability acknowledgment (CDO/Chief Dental Officer)
☐ Security and compliance sign-off on BAA and data handling (Compliance/Legal)
☐ IT architecture approval for integration approach (IT Director)
☐ Executive sponsor designation for board-level accountability (CEO or CDO)
Baseline Metrics to Capture
⚠️ Critical: Capture these metrics BEFORE go-live. Without baseline data, ROI demonstration is impossible.
Clinical Metrics (Per Location)
| Metric | How to Capture | Target Locations |
|---|---|---|
| Caries detection rate per exam | Chart audit of 50 random exams from past 90 days | All Wave 1 + sample of others |
| Periodontal disease documentation rate | Same chart audit | All Wave 1 |
| Pathology referral rate | Count of referrals made for diagnosed pathology | All Wave 1 |
| Time from image capture to diagnosis documented | Workflow observation, 10 patients per location | Wave 1 only |
Business Metrics (Enterprise-Wide)
| Metric | How to Capture | Frequency |
|---|---|---|
| Case acceptance rate by location | PMS reporting, treatment planned vs. completed | Monthly average, trailing 6 months |
| Average treatment plan value | PMS reporting | Monthly average |
| Claim denial rate for diagnostic codes | Billing system reporting | Monthly average |
| Patient NPS by location (if tracked) | Existing survey data | Latest quarter |
Standardization Protocol
☐ Create standardized data collection template for all locations
☐ Assign central team member to validate data quality before accepting submissions
☐ Document any methodology variations that could affect cross-location comparison
☐ Store baseline data in centralized system accessible to operations and CDO
3. Location Readiness Assessment
Scoring Framework
Rate each location 1–5 on the following factors, then calculate a composite score.
Factor 1: IT Infrastructure Maturity (Weight: 25%)
| Score | Criteria |
|---|---|
| 5 | Fiber internet (100+ Mbps), workstations <3 years old, current PMS version, dedicated IT support |
| 4 | Cable internet (50+ Mbps), workstations 3–5 years old, PMS within 2 versions of current |
| 3 | Mixed connectivity (25–50 Mbps), some workstations >5 years, PMS slightly outdated |
| 2 | Inconsistent connectivity (<25 Mbps), workstations 5–7 years old, PMS significantly outdated |
| 1 | Unreliable internet, legacy hardware, PMS no longer supported |
Factor 2: Staff Tenure and Adaptability (Weight: 20%)
| Score | Criteria |
|---|---|
| 5 | <15% annual turnover, previous successful tech adoption, team expresses enthusiasm for AI |
| 4 | 15–25% turnover, neutral to positive tech adoption history |
| 3 | 25–35% turnover, mixed tech adoption outcomes |
| 2 | 35–50% turnover, previous tech implementations struggled |
| 1 | >50% turnover, active resistance to technology changes |
Factor 3: Patient Volume (Weight: 15%)
| Score | Impact Assessment |
|---|---|
| 5 | Top 20% of portfolio by volume—highest ROI potential, can absorb some workflow disruption |
| 4 | 60–80th percentile—strong ROI potential, manageable risk |
| 3 | 40–60th percentile—balanced risk/reward |
| 2 | 20–40th percentile—lower immediate ROI, may be good for controlled pilot |
| 1 | Bottom 20%—limited ROI, deprioritize unless strategically important |
Factor 4: Tech Stack Compatibility (Weight: 25%)
| Score | Criteria |
|---|---|
| 5 | PMS and imaging system with documented CranioCatch integration, API access confirmed |
| 4 | PMS/imaging system on supported list, integration not yet deployed at scale |
| 3 | PMS supported, imaging system requires workaround (DICOM export) |
| 2 | PMS requires custom integration work, imaging compatibility uncertain |
| 1 | Legacy PMS or imaging system without clear integration path |
Factor 5: Local Champion Availability (Weight: 15%)
| Score | Criteria |
|---|---|
| 5 | Tech-forward dentist AND office manager, both volunteered, history of successful change leadership |
| 4 | Strong champion in one role (provider or manager), supportive counterpart |
| 3 | Willing champion identified but unproven in this capacity |
| 2 | No natural champion, but no active resistance |
| 1 | Key influencers skeptical or resistant, high risk of implementation friction |
Composite Score Calculation
Formula: (Factor 1 × 0.25) + (Factor 2 × 0.20) + (Factor 3 × 0.15) + (Factor 4 × 0.25) + (Factor 5 × 0.15) = Composite Score
Readiness Tiers
| Composite Score | Tier | Rollout Wave |
|---|---|---|
| 4.0–5.0 | High Readiness | Wave 1 (Pilot) |
| 3.0–3.9 | Moderate Readiness | Wave 2 |
| 2.0–2.9 | Low Readiness | Wave 3 |
| <2.0 | Remediation Required | Hold for infrastructure/staffing improvements |
Recommended Rollout Sequence
Wave 1 Selection: Choose 2–3 locations scoring 4.0+ with geographic distribution representing your portfolio diversity (urban/suburban, different PMS systems, different imaging equipment)
Wave 2 Selection: All locations scoring 3.0–3.9, prioritizing those with highest patient volume within this tier
Wave 3 Selection: Locations scoring 2.0–2.9, grouped by shared barrier (e.g., all locations needing network upgrade deployed together after infrastructure remediation)
Hold List: Locations below 2.0 require investment before CranioCatch deployment—create remediation plan with timeline
4. Rollout Strategy
Wave Structure
Recommended Wave Configuration
| Wave | Location Count | Duration | Cumulative Coverage |
|---|---|---|---|
| Wave 1 (Pilot) | 2–3 locations | 4–6 weeks | 5–10% of portfolio |
| Wave 2 (Expansion) | 5–8 locations | 4 weeks | 25–35% of portfolio |
| Wave 3 (Scale) | 8–12 locations | 4 weeks | 55–70% of portfolio |
| Wave 4 (Completion) | Remaining | 4 weeks | 100% |
For 15–20 location DSOs, Waves 3 and 4 can be combined.
Wave 1 Pilot Selection Criteria
🟣 Executive Decision Required: Final pilot location selection
Must-Have Criteria:
☐ Composite readiness score ≥4.0
☐ Office manager has capacity to participate in daily check-ins for 2 weeks
☐ At least one provider willing to serve as clinical champion
☐ No major competing initiatives (renovation, new provider onboarding, PMS migration)
☐ Geographic accessibility for in-person support if needed
Nice-to-Have Criteria:
☐ Different PMS systems across pilot locations to test integration variety
☐ Mix of patient demographics representative of broader portfolio
☐ Location that struggled with previous tech implementation (tests change management approach)
☐ Presence of associate dentists (tests AI-assisted learning use case)
Timeline Per Wave
Wave 1 (Pilot): Weeks 3–8
- Week 3: Configuration and integration at pilot locations
- Week 4: Staff training and parallel testing
- Week 5: Go-live with intensive monitoring
- Weeks 6–7: Stabilization and workflow optimization
- Week 8: Pilot retrospective and Wave 2 preparation
Wave 2: Weeks 9–12
- Week 9: Configuration and integration (5–8 locations simultaneously)
- Week 10: Champion-led training using pilot learnings
- Week 11: Go-live with daily monitoring
- Week 12: Stabilization
Wave 3+: Repeat 4-week cycle
Go/No-Go Criteria Between Waves
🟣 Wave Advancement Requires Executive Approval
Go Criteria (All Must Be Met):
☐ ≥90% of pilot locations actively using tool for all eligible radiographs
☐ No unresolved critical technical issues (system downtime, integration failures)
☐ Provider satisfaction score ≥3.5/5.0 in pulse survey
☐ Training completion rate 100% for all roles at pilot locations
☐ No patient safety incidents attributable to AI implementation
☐ Documentation shows workflow efficiency neutral or improved
No-Go Triggers (Any One Stops Advancement):
⚠️ Integration failure causing data integrity issues
⚠️ Provider revolt—>50% of pilot providers refusing to use system
⚠️ Critical security vulnerability identified
⚠️ Vendor unable to provide adequate support response times
Rollback Plan
If a Wave Must Be Paused:
Immediate Communication (Within 2 hours of decision)
- Central team notifies all affected locations
- Regional managers brief office managers
- Vendor account manager engaged for technical assessment
System Deactivation (Within 24 hours)
- Disable CranioCatch integration at affected locations
- Revert to pre-implementation diagnostic workflow
- Preserve all configuration settings for restart
Impact Containment
- Unaffected waves continue on schedule unless root cause is systemic
- Create firewall between technical issue resolution and ongoing rollout
Root Cause Analysis (Within 72 hours)
- Document failure mode
- Assign corrective action owner
- Establish restart criteria
Restart Protocol
- Vendor certifies issue resolved
- Central team validates fix in test environment
- Restart with extended monitoring period
5. Configuration & Integration (Weeks 2–3)
Practice Management System Integration
Dentrix Integration (Patterson)
🔵 Vendor Involvement Required
Prerequisites:
☐ Dentrix G7.3 or later installed
☐ Dentrix Smart Image with TWAIN or ScanX interface
☐ Local administrator credentials available
Step-by-Step Process:
- ☐ Install CranioCatch desktop agent on imaging workstation (installer provided by CranioCatch)
- ☐ Configure Dentrix image export path to CranioCatch monitored folder
- 🔵 CranioCatch technical team provides API bridge configuration
- ☐ Map Dentrix patient ID to CranioCatch patient record
- ☐ Configure write-back of AI annotations to patient chart
- ☐ Test bidirectional data flow with 5 test patients
- ☐ Validate image quality settings for optimal AI analysis
Common Issues:
⚠️ Image compression in Dentrix may reduce AI accuracy—disable lossy compression
⚠️ Multi-doctor charts require provider-specific routing configuration
Eaglesoft Integration (Patterson)
🔵 Vendor Involvement Required
Prerequisites:
☐ Eaglesoft 21.0 or later
☐ Eaglesoft Advanced Imaging module enabled
☐ SQL database access for patient record sync
Step-by-Step Process:
- ☐ Enable Eaglesoft API access (requires Patterson support)
- 🔵 CranioCatch configures middleware connector
- ☐ Install CranioCatch agent with Eaglesoft-specific configuration
- ☐ Map imaging workflow to trigger automatic CranioCatch analysis
- ☐ Configure annotation overlay display within Eaglesoft viewer
- ☐ Test full workflow: image capture → AI analysis → chart update
- ☐ Verify audit trail captures AI-assisted diagnosis flag
Common Issues: ⚠️ Eaglesoft's image handling varies by acquisition device—test each sensor type separately
Open Dental Integration
Prerequisites:
☐ Open Dental version 21.1 or later
☐ Open Dental API key (Service Account)
☐ eConnector service running
Step-by-Step Process:
- ☐ Generate API key in Open Dental (Setup → Advanced Setup → API)
- 🔵 Provide API key to CranioCatch for cloud integration setup
- ☐ Install CranioCatch imaging monitor on acquisition workstation
- ☐ Configure OpenDentalImages folder path synchronization
- ☐ Map patient identifier and provider ID fields
- ☐ Test with existing patient records (do not use production-creating new patients)
- ☐ Configure procedure code auto-entry for AI-assisted exams
Common Issues:
⚠️ Open Dental's multi-database architecture requires careful location-specific configuration
⚠️ Version mismatches between locations require individual integration testing
Imaging System Integration
Carestream Dental Integration
☐ Enable DICOM export from Carestream software
🔵 CranioCatch provides DICOM listener endpoint configuration
☐ Configure automatic export on image acquisition
☐ Validate image metadata (patient ID, date, image type) transfers correctly
☐ Test panoramic, bitewing, and periapical image types separately
Dexis Integration
☐ Verify Dexis version (9.0+ recommended)
☐ Configure Dexis Imaging export settings
☐ Install CranioCatch integration plugin (if available for version)
☐ Test TWAIN image capture workflow
Sirona/SIDEXIS Integration
☐ Enable DICOM server mode in SIDEXIS
🔵 Configure CranioCatch as DICOM destination
☐ Test 2D and CBCT image routing separately
☐ Validate CBCT processing time expectations (may require hybrid deployment)
Test Environment Setup
Centralized Test Environment (Recommended)
☐ Create dedicated CranioCatch test tenant separate from production
☐ Configure test PMS instance (or isolated database) per PMS type in portfolio
☐ Upload sample images representing each imaging system in use
☐ Establish test patient records that mirror production data structure
🔵 Request CranioCatch test API credentials with same permissions as production
Validation Checklist
☐ Image upload completes without error
☐ AI analysis returns within expected timeframe (<10 seconds for 2D, <60 seconds for CBCT)
☐ Annotations display correctly on image viewer
☐ Patient record linkage is accurate
☐ Provider attribution captures correctly
☐ Report generation functions as expected
☐ Audit log captures all required events
⚠️ Test with minimum viable network connectivity to simulate worst-case location
Data Migration / Historical Ingestion
🟣 Executive Decision Required: Historical image analysis scope
Options:
| Approach | Pros | Cons |
|---|---|---|
| Prospective only (new images from go-live) | Simplest, lowest cost | No AI insight on existing patient base |
| Selective retrospective (high-risk patients) | Targeted value, manageable volume | Requires patient identification criteria |
| Full retrospective (all images from past X months) | Maximum insight, enables trend analysis | Significant cost, long processing time |
If Retrospective Analysis Selected:
☐ Define scope: which patients, which image types, how far back
☐ Estimate image volume for CranioCatch capacity planning
🔵 Negotiate retrospective analysis pricing and timeline with CranioCatch
☐ Create workflow for reviewing AI findings on historical images
☐ Establish provider communication process for flagged historical pathology
⚠️ Patient notification and re-engagement for discovered historical pathology requires clinical and legal protocol
Security and HIPAA Compliance Verification
Enterprise-Level HIPAA Checklist
☐ Business Associate Agreement (BAA) executed with CranioCatch
☐ BAA covers all intended use cases (cloud storage, AI processing, potential offshore support access)
☐ Data encryption in transit verified (TLS 1.2 minimum)
☐ Data encryption at rest verified (AES-256 or equivalent)
☐ CranioCatch SOC 2 Type II report reviewed by security team
☐ Data retention and deletion policies documented and acceptable
☐ Breach notification procedures defined and SLA acceptable
☐ Subprocessor list reviewed (where does CranioCatch data flow?)
Access Control Checklist
☐ Role-based access control configured by location and role
☐ Admin access limited to authorized individuals only
☐ SSO integration enforces organizational password/MFA policies
☐ Session timeout configured to organizational standards
☐ Audit logs capture access events for compliance review
☐ Offboarding process defined (credential revocation timing)
Data Governance
☐ Data ownership clause confirmed (your data remains yours)
☐ AI model training policy reviewed (does CranioCatch use your data to train models?)
🟣 If yes, determine if opt-out is available and whether to exercise it
☐ Data export capability confirmed (you can retrieve your data if leaving)
☐ Data residency location documented (important for state-specific requirements)
Standardized vs. Location-Specific Configuration
Standardize Centrally
| Setting | Standard Configuration |
|---|---|
| Detection sensitivity | Medium (balance false positives vs. missed findings) |
| Mandatory pathology categories | All FDA-cleared categories enabled |
| Report format | Standard template across all locations |
| Annotation color scheme | Consistent color coding for pathology types |
| Audit logging level | Full capture for all clinical interactions |
| User role definitions | Standard roles: Admin, Provider, Hygienist, Viewer |
| Integration parameters | Standardized API configuration per PMS type |
Allow Local Discretion
| Setting | Local Customization Rationale |
|---|---|
| Provider-specific display preferences | Individual providers may prefer different annotation densities |
| Specialty-specific detection emphasis | Pediatric vs. general vs. endo may weight findings differently |
| Report language/template variations | Patient communication style may vary by market |
| Training pace and timing | Accommodate local scheduling constraints |
6. Team Training Plan
Train-the-Trainer Model
Champion Selection Criteria
Provider Champion (1 per location):
☐ Dentist with demonstrated technology adoption comfort
☐ Peer respect and clinical credibility among other providers
☐ Available for 4 hours of central training plus ongoing support role
☐ Willingness to provide feedback and suggest workflow improvements
☐ Not departing organization within 12 months
Office Manager Champion (1 per location):
☐ Operational competency with current technology stack
☐ Change management experience (has led previous implementations)
☐ Strong communication skills for staff training delivery
☐ Available for 3 hours of central training plus ongoing support role
☐ Authority to adjust workflows within their location
Champion Responsibilities
During Implementation:
- Complete central training certification
- Customize training delivery for their team
- Serve as first-line troubleshooting contact
- Participate in daily check-ins during go-live week
- Collect and communicate staff feedback
Ongoing:
- Train new hires within 1 week of start date
- Conduct quarterly refresher sessions
- Escalate unresolved issues to regional manager
- Participate in cross-location champion community
Standardized Training Materials (Created Centrally)
☐ Training slide deck with speaker notes
☐ Role-specific quick reference guides (see Day 1 cheat sheets below)
☐ Video library: 3–5 minute modules per key workflow
☐ FAQ document updated after each wave
☐ Common objection responses for provider resistance
☐ Patient communication scripts
☐ Training completion certification quiz
Role-Specific Training Outlines
Dentists/Providers
Training Time: 45 minutes initial + 15 minutes shadow session
Training Format: Live demo (20 min) + hands-on practice (15 min) + shadow session with first 5 patients (15 min)
Content:
- What CranioCatch detects and its accuracy validation
- How AI annotations appear on images
- Interpreting confidence scores and clinical significance
- When and how to override or dismiss AI findings
- Documentation requirements for AI-assisted diagnosis
- Patient communication: presenting AI findings
- Workflow integration: when to review AI output in exam flow
Common Resistance Points:
| Objection | Response |
|---|---|
| "I don't need AI to tell me what I can already see" | "Think of it as a second pair of eyes that never gets tired. It's caught findings in studies that experienced providers missed. It's a safety net, not a replacement for your expertise." |
| "This will slow me down" | "After the first week, most providers report neutral or improved speed. The AI pre-highlights areas to focus on, reducing systematic scan time." |
| "What about liability if the AI misses something?" | "You remain the diagnosing clinician. The AI is a decision support tool. Document your clinical judgment when you disagree with AI findings." |
| "My patients won't trust computer diagnosis" | "Patient response data shows the opposite—visualizing findings on screen increases trust and case acceptance. You're not deferring to AI, you're showing transparency." |
Day 1 Cheat Sheet for Providers:
┌─────────────────────────────────────────────────────────────────┐
│ CRANIOCATCH PROVIDER QUICK REFERENCE │
├─────────────────────────────────────────────────────────────────┤
│ TO VIEW AI FINDINGS: │
│ 1. Open patient image as usual │
│ 2. Click "AI Analysis" button (⚡ icon) or wait for auto-load │
│ 3. Colored overlays show detected findings │
│ │
│ COLOR CODING: │
│ 🔴 Red = Caries/Decay 🟡 Yellow = Bone loss │
│ 🔵 Blue = Restoration 🟣 Purple = Periapical pathology │
│ 🟢 Green = Anatomy ⚪ Gray = Calculus │
│ │
│ CONFIDENCE SCORES: │
│ >90% = High confidence 70-89% = Review recommended │
│ <70% = Possible finding, clinical judgment essential │
│ │
│ TO OVERRIDE/DISMISS: │
│ Click finding → "Dismiss" or "Modify" → Add clinical note │
│ │
│ PATIENT COMMUNICATION: │
│ "This imaging technology helps me identify areas of concern. │
│ Let me show you what it's highlighting and explain what │
│ I'm seeing clinically..." │
│ │
│ SUPPORT: [Champion Name] | Escalation: [Phone] | Vendor: [#] │
└─────────────────────────────────────────────────────────────────┘
Hygienists
Training Time: 30 minutes
Training Format: Live demo (15 min) + hands-on practice (15 min)
Content:
- When hygienists interact with CranioCatch (if imaging duties apply)
- How to explain AI analysis to patients during cleaning
- Flagging findings for dentist
AI-generated implementation guide based on public vendor information. Verify specifics directly with CranioCatch.