Toothlens Implementation Playbook (DSO)

Executive Summary

Toothlens is a diagnostic imaging intelligence platform that analyzes dental radiographs in real-time, flagging pathology, treatment opportunities, and quality issues before they become clinical or compliance problems. For DSOs managing 10+ locations, this tool standardizes diagnostic consistency across multiple clinicians, reduces missed findings, and creates a data layer for treatment acceptance and clinical governance. Deployment at scale drives per-location efficiency gains of 8-15% in diagnostic accuracy and accelerates case acceptance by 20-30% through consistent case presentation.

Pre-Implementation Checklist

• DICOM/Imaging System Audit: Document all imaging hardware and software across all 10+ locations (sensors, intraoral cameras, CBCT systems, RVG/Schick versions). Confirm which systems export DICOM or standard image formats. Identify any legacy analog or proprietary systems requiring workarounds.

• EHR Integration Mapping: Verify your primary EHR (Dentrix, Eaglesoft, Curve, etc.) and document current imaging import workflows. Confirm who has admin access at each location and whether your IT infrastructure supports bidirectional data flow without bottlenecks.

• Clinical Governance & Protocol Review: Establish who owns diagnostic protocols across the DSO (Chief Clinical Officer, Regional Directors, or individual practice doctors?). Draft a standardized response protocol for Toothlens alerts before implementation—decide in advance how findings are documented, communicated, and reviewed.

• IT Network & Security: Confirm all 10+ locations have HIPAA-compliant cloud access, bandwidth adequate for real-time image transmission (minimum 10 Mbps), and current firewall/VPN configurations. Identify any sites with restricted internet or offline-heavy workflows requiring custom setup.

• Staff Skill Inventory: Catalog tech-level staff at each location (clinical coordinators, imaging techs, office managers). Identify 1-2 "power users" per location for intensive training who can become internal champions and troubleshoot peer issues.

• Budget & Procurement Sign-Off: Secure finance approval for per-location licensing, potential hardware upgrades, and 6-month support/integration budget. Confirm any volume discounts or enterprise licensing arrangements with Toothlens beforehand.

• Change Management Sponsor: Assign a Chief Clinical Officer or VP of Clinical Operations as executive sponsor to communicate rollout timeline, address physician resistance, and enforce adoption discipline.

Implementation Timeline

**Phase 1: Pilot & Validation (Weeks 1–3)**

• Week 1: Deploy Toothlens at 2 high-volume, tech-forward locations. Conduct hands-on training with imaging techs, clinical coordinators, and 2–3 clinicians per site. Run parallel workflows (manual diagnostic review + Toothlens output) for 1 week to validate accuracy and workflow fit. Document all integration hiccups with your Toothlens implementation manager.

• Week 2–3: Review pilot data (case volume, alert frequency, clinician feedback). Adjust alert sensitivity thresholds, documentation templates, and case routing based on real workflows. Confirm DICOM export quality and image-to-alert latency. Conduct debrief with pilot site leads to refine training scripts and identify location-specific protocol tweaks.

**Phase 2: Staged Rollout to Remaining Locations (Weeks 4–8)**

• Week 4: Deploy to 3–4 additional mid-size locations (mix of tech adoption maturity). Assign pilot-site power users as peer mentors via video calls. Validate that standard training deck and templates work across different practice cultures.

• Week 5–6: Roll out to remaining 5–6 locations in 2-location cohorts. Use a cascading "train-the-trainer" model: Toothlens implementation team trains regional leads, who train site staff. Establish a centralized Slack/Teams channel for cross-location issue resolution. Run weekly sync calls with all practice managers to track adoption metrics.

• Week 7–8: Monitor alert volume, clinician engagement, and integration stability across all locations. Patch any systemic issues (e.g., DICOM export delays, EHR sync bugs). Begin collecting baseline diagnostic accuracy and case acceptance data.

**Phase 3: Optimization & Enforcement (Weeks 9–12)**

• Week 9–10: Conduct site-by-site audit of Toothlens adoption: Are clinicians reviewing alerts? Are findings being documented consistently? Are alerts reducing missed pathology? Host regional "deep-dive" training sessions for underperforming locations.

• Week 11–12: Lock in standard protocols (e.g., all periapical findings >3mm must trigger a clinical note; all CBCT pathology triggers a treatment case). Publish DSO-wide diagnostic guidelines referencing Toothlens classifications. Plan quarterly clinical review meetings to audit Toothlens performance and adjust sensitivity settings.

Key Metrics to Track

• Diagnostic Accuracy Rate: % of Toothlens-flagged findings confirmed by clinician review vs. missed findings on manual-only cases. Target: 92%+ agreement by week 12; monitor drift monthly.

• Alert Sensitivity/Specificity: Average alerts per 100 radiographs and % of alerts clinically actionable. Target: 15–25 alerts/100 radiographs; <10% false positives by week 8.

• Clinician Adoption: % of clinicians actively reviewing Toothlens alerts within 24 hours of imaging. Target: 85%+ by week 10; track via EHR audit logs.

• Case Acceptance Lift: Treatment case case-acceptance rate pre- vs. post-Toothlens (measured by treatment-plan acceptance %). Target: +15–20% lift within 90 days, driven by consistent case presentation of Toothlens-identified pathology.

• Missed Pathology Reduction: Count of undiagnosed pathology identified post-treatment (emergency RCTs, late-stage perio disease) in Toothlens-enabled workflows vs. baseline period. Target: 30–40% reduction within 180 days.

• Imaging Tech Efficiency: Average time to process radiograph (capture → Toothlens export → clinician alert) and % of images processed on first capture. Target: <3 minutes per image; 95% first-capture quality by week 6.

Common Pitfalls

• Assuming "Set and Forget" Adoption: Many DSOs deploy Toothlens and expect clinician buy-in without ongoing reinforcement. Reality: Early enthusiasm fades if alerts aren't systematically reviewed or documented. Mitigation: Establish weekly DSO-wide clinical huddles (30 min) reviewing interesting Toothlens cases. Tie clinician performance dashboards (% alerts reviewed, finding confirmation rate) to quality scorecards and compensation.

• Ignoring Regional Imaging Quality Variance: Some locations may have older sensors, undertrained imaging techs, or poor positioning discipline. Toothlens accuracy degrades on poor-quality images. Mitigation: Conduct an imaging-quality audit pre-implementation. Invest in sensor upgrades or positioning jigs at underperforming locations before Toothlens rollout. Set minimum image-quality standards (e.g., Perfectly Positioned Index >80%).

• Misaligning Alert Sensitivity Too High: Deploying with hyper-sensitive settings generates alert fatigue, and clinicians start ignoring Toothlens output by week 3. Mitigation: Begin with conservative (specific) settings; iteratively increase sensitivity after clinician feedback confirms value. Start at 60–70% sensitivity; increase to 75–85% after 4 weeks of validation.

• Skipping EHR Integration Documentation: Toothlens findings aren't consistently documented in treatment plans or clinical notes, fragmenting the clinical record and reducing case acceptance leverage. Mitigation: Create templated clinical-note macros (e.g., "Toothlens-identified class II caries mesial surface #14; treatment plan outlined") and train all clinicians to use them. Audit notes monthly for consistency.

• Underestimating Change Resistance from Older Clinicians: Senior doctors may perceive Toothlens as a threat to clinical autonomy or "cookbook dentistry." Mitigation: Frame Toothlens as a diagnostic safety net, not a replacement for clinical judgment. Share peer testimonials from respected practices. Offer 1-on-1 demo sessions emphasizing how Toothlens surfaces findings they'd otherwise