π λͺ©μ°¨Contents
- π₯ νμ νλΈ (Phase C, NEW!)π₯ Patient Hub (Phase C, NEW!)
- μμ€ν ꡬ쑰 μ΄ν΄System Architecture
- π©Ί ν΄λ¦¬λ λμ보λ (v3 Β· μ΄λ©μΌ μ μ‘)π©Ί Clinic Dashboard (v3 Β· Email)
- κ΄λ¦¬μ νμ΄μ§ λλ¬λ³΄κΈ°Admin Panel Tour
- Linkage Matrix νΈμ§Editing Linkage Matrix
- λ°°ν¬ μν¬νλ‘Deployment Workflow
- μμ νλ κ°μ΄λClinical Tuning Guide
- μ°κ΅¬ κ·Όκ±° μ°Έκ³ Research References
- λ¬Έμ ν΄κ²°Troubleshooting
π₯ νμ νλΈ β Phase C ν΅ν© μμ€ν (NEW!)Patient Hub β Phase C Integrated System (NEW!)
νμ νλΈλ Face Reading + ADHD Catcher + Symptom Catcher + QEEG + μμ λ ΈνΈλ₯Ό ν νμ μ€μ¬μΌλ‘ ν΅ν©νλ μ μμ€ν μ λλ€. κΈ°μ‘΄ Clinic Dashboardμ νμ₯ν. Patient Hub is a new system that unifies Face Reading + ADHD Catcher + Symptom Catcher + QEEG + Clinical Notes around each individual patient. An extended version of the existing Clinic Dashboard.
https://neurocatchers.com/symptom_catcher/face-clinic/patient-hub-v1.htmlλΉλ°λ²νΈ:Password:
bnm2026
β¨ μ£Όμ κΈ°λ₯Key Features
- ν΅ν© νμ DB: μ΄λ©μΌ, μ ν, μλ μμΌ, μ νΈ μΈμ΄, μ£ΌνΈμ λ± μμ κΈ°λ³Έ μ 보Unified patient DB: email, phone, DOB, preferred language, chief complaint
- λͺ¨λ λͺ¨λ¬λ¦¬ν° νλμ: 6κ° ν (Overview / Face / ADHD / Symptom / QEEG / Notes)All modalities at a glance: 6 tabs (Overview / Face / ADHD / Symptom / QEEG / Notes)
- νλ νμλΌμΈ: μ΅κ·Ό 15κ° νλ μκ°μ ν΅ν© νμActivity timeline: recent 15 activities chronologically unified
- μμ λ ΈνΈ: intake/progress/assessment/plan/discharge μ’ λ₯λ³Clinical notes: intake/progress/assessment/plan/discharge types
- μ΄λ©μΌ μ μ‘: λ²νΌ ν λ²μΌλ‘ Face Reading κ²°κ³Ό HTML μ΄λ©μΌEmail delivery: HTML email of Face Reading results in one click
ποΈ DB ν μ΄λΈ ꡬ쑰DB Table Structure
patientsβ μ€μ νμ μ 보 (νμ₯λ¨)central patient info (extended)face_sessionsβ Face Readingadhd_sessionsβ ADHD νκ° (μ€ν€λ§ μ€λΉ, μ°κ²° λκΈ°)ADHD assessments (schema ready, wiring pending)symptom_sessionsβ Symptom νκ° (μ€ν€λ§ μ€λΉ)Symptom assessments (schema ready)qeeg_sessionsβ QEEG (future)clinical_notesβ μμ λ ΈνΈclinical notesemail_logsβ μ΄λ©μΌ μ μ‘ μΆμ email send trackingpatient_activity_summary(VIEW) β νμλ³ νλ μμ½per-patient activity summary
π λ°°ν¬ λ¨κ³ (μΈλΌνΌμ€νΈκ° ν μΌ)Setup Steps
- Supabase SQL Editorμμ
schema_phase_c_central_db.sqlμ€νRunschema_phase_c_central_db.sqlin Supabase SQL Editor - Resend κ³μ μμ± + API Key λ°κΈCreate Resend account + API key
- Supabase CLIλ‘ Edge Function λ°°ν¬ (
send-face-reading-email)Deploy Edge Function via Supabase CLI - GitHubμ
patient-hub-v1.htmlμ λ‘λ β μλ λ°°ν¬Uploadpatient-hub-v1.htmlto GitHub β auto-deploys - μμΈ κ°μ΄λ:
A2_EMAIL_DEPLOY_GUIDE.mdλ°CLINICIAN-MANUAL.htmlDetailed guides:A2_EMAIL_DEPLOY_GUIDE.mdandCLINICIAN-MANUAL.html
ποΈ μμ€ν ꡬ쑰 μ΄ν΄System Architecture
3λ¨κ³ κ³μ° νμ΄νλΌμΈ3-Stage Calculation Pipeline
μ΄ μμ€ν μ λͺ¨λ κ°μ€μΉλ νΈμ§ κ°λ₯ν΄μ. μ°κ΅¬ κ²°κ³Όμ μμ κ΄μ°°μ λ°λΌ μ§μμ μΌλ‘ νλν μ μλλ‘ μ€κ³λμ΄μ. νλμ½λ©λ AI λͺ¨λΈμ΄ μλμμ. All weights are editable in this system. Designed for continuous tuning based on research and clinical observation. Not a hardcoded AI model.
νμΌ κ΅¬μ‘°File Structure
π©Ί ν΄λ¦¬λ λμ보λ (Phase B)Clinic Dashboard (Phase B)
μ μ URLAccess URL
π λ‘κ·ΈμΈLogin
νμ¬ λΉλ°λ²νΈ: bnm2026 (Phase B μμ, μΆν κ°ν μμ )
Current password: bnm2026 (Phase B temporary, to be hardened)
π₯ νμ λ±λ‘Patient Registration
- μΌμͺ½ ν¨λ "β μ νμ λ±λ‘" ν΄λ¦Click "β Add New Patient" on left panel
- μ΄λ¦κ³Ό λ©λͺ¨ μ λ ₯ (λͺ¨λ μ νμ¬ν)Enter name and notes (both optional)
- "λ±λ‘" ν΄λ¦ β μλμΌλ‘ λ€μ ID λ°κΈ (BNM-002, BNM-003...)Click "Add" β next ID auto-generated (BNM-002, BNM-003...)
- νμμκ² μ½λ μλ €μ£ΌκΈ° β νμκ° face-readingμ μ λ ₯νλ©΄ μλ μ μ₯Give code to patient β results auto-save when they use face-reading
π νμ μμΈ νλ©΄Patient Detail View
- μ΅κ·Ό λΆμ μμ½: κ°μ₯ μ΅κ·Ό μΈμ μ Big Five 5κ° μ μLatest Summary: Most recent session's Big Five scores
- μ£Όμ μ¦μ Top 3: κ°μ₯ κ°ν μ¦μ κ²½ν₯Top 3 Symptoms: Strongest tendencies
- μκ°λ³ λ³ν κ·Έλν: Big Five / μ¦μ / κ°μ 3κ°μ§ μ°¨νΈ μ ν κ°λ₯Trends Chart: Switch between Big Five / Symptoms / Emotion
- μ 체 μΈμ κΈ°λ‘: λͺ¨λ λΆμ μκ°μ + μΈμ μμ κ°λ₯All Sessions: Chronological + deletable
πΎ λ°μ΄ν°λ² μ΄μ€ (Supabase)Database (Supabase)
Supabase νλ‘μ νΈ URL: Supabase project URL:
λ κ°μ ν μ΄λΈ: Two tables:
patientsβ νμ μ 보 (id, name, notes, created_at)Patient infoface_sessionsβ λΆμ μΈμ (Big5, risks JSON, emotion, wuxing, FACS λ±)Analysis sessions (Big5, risks JSON, emotion, wuxing, FACS etc)
π μμΌλ‘ (Phase C)Roadmap (Phase C)
- μ§μ§ μΈμ¦ μμ€ν (μ΄λ©μΌ + λΉλ°λ²νΈ)Real authentication (email + password)
- Row Level Security κ°νHardened Row Level Security
- QEEG κ²°κ³Όμ ν΅ν©Integration with QEEG results
- AI μ μν νλ‘ν μ½ μΆμ²AI adaptive protocol recommendations
- HIPAA BAA (νμμ Supabase Pro $25/μ)HIPAA BAA (Supabase Pro if needed, $25/mo)
ποΈ κ΄λ¦¬μ νμ΄μ§ λλ¬λ³΄κΈ°Admin Panel Tour
face-admin-v3.1.htmlμ 8κ° νμΌλ‘ ꡬμ±λμ΄ μμ΄μ:
face-admin-v3.1.html has 8 tabs:
| νTab | μν Purpose |
|---|---|
| π Overview | μμ€ν μν, κ³μ° 곡μ, μ°κ΅¬ κ·Όκ±° νλμSystem stats, formula, research basis at a glance |
| πΈ Big5 β Sym | 16κ° μ¦μ Γ 5κ° μ±κ²© λ§€νΈλ¦μ€ (80μΉΈ)16 symptoms Γ 5 Big Five matrix (80 cells) |
| π Feat β Big5 | 18κ° μΌκ΅΄ νΉμ§ Γ 5κ° μ±κ²© λ§€νΈλ¦μ€ (90μΉΈ)18 facial features Γ 5 Big Five matrix (90 cells) |
| π Feat β Sym | μΌκ΅΄ νΉμ§ β μ¦μ μ§μ μ°κ²° (sparse)Facial features β symptoms direct (sparse) |
| π₯ Symptoms | μ¦μ μΆκ°/μμ /μμ Add/edit/delete symptoms |
| β¨ Rules | 컀μ€ν μμ κ·μΉ (μ‘°κ±΄λΆ κ°μ€μΉ λΆμ€νΈ)Custom clinical rules (conditional boosts) |
| βοΈ Fusion | Ξ±, Ξ² κ°μ€μΉ μ‘°μ Ξ±, Ξ² weight adjustment |
| πΎ I/O | JSON λ€μ΄λ‘λ/μ λ‘λ/μ΄κΈ°νJSON download/upload/reset |
π Linkage Matrix νΈμ§Editing Linkage Matrix
κ°μ€μΉ ν΄μλ²How to Read Weights
- μμ (+): νΉμ±μ΄ λμμλ‘ μ¦μ μνλ β (μ: μ κ²½μ¦ 0.55 β μ°μΈ)Positive (+): Higher trait β higher risk (e.g., N=0.55 β depression)
- μμ (β): νΉμ±μ΄ λμμλ‘ μ¦μ μνλ β (λ³΄νΈ μμΈ, μ: μ±μ€μ± β0.25 β μ°μΈ)Negative (β): Higher trait β lower risk (protective, e.g., C=β0.25 β depression)
- 0: μ°κ΄ μμZero: No relationship
- λ²μ: -1.0 ~ +1.0 (κ΄λ‘μ μΌλ‘ Β±0.7 μ΄μμ κ°ν μ°κ΄, Β±0.3~0.7μ μ€κ°, Β±0.3 μ΄νλ μ½ν¨)Range: -1.0 to +1.0 (convention: |0.7|+ strong, |0.3-0.7| moderate, |0.3| or less weak)
μμ μ½λ©Color Coding
- πΈ λΆν: μμ μ°κ΄ (νΉμ±ββμ¦μβ)πΈ Pink: Positive association
- π 보λΌ: μμ μ°κ΄, ν° μ λκ° (λ³΄νΈ μμΈ)π Purple: Negative, large magnitude (protective)
- πΏ λ―ΌνΈ: μμ μ°κ΄, μ€κ°πΏ Mint: Negative, moderate
μ μ¦μ μΆκ°Adding New Symptoms
- π₯ Symptoms νμΌλ‘ μ΄λGo to π₯ Symptoms tab
- "μ μ¦μ μΆκ°" μΉμ μμ ID (snake_case), νκ΅μ΄λͺ , English name, Category μ λ ₯In "Add New Symptom" section, enter ID (snake_case), Korean name, English name, Category
- ID λͺ
λͺ
κ·μΉ: μλ¬Έ μλ¬Έμ + μΈλμ€μ½μ΄ (μ:
social_anxiety)ID naming: lowercase + underscore (e.g.,social_anxiety) - "β μΆκ°" β μλμΌλ‘ λͺ¨λ λ§€νΈλ¦μ€μ ν΄λΉ μ»¬λΌ μμ±λ¨"β Add" β Column auto-created in all matrices
- πΈ Big5 β Sym νμΌλ‘ κ°μ μ μ¦μμ Big5 κ°μ€μΉ μ λ ₯ (μ°κ΅¬ κ·Όκ±° μ°Έκ³ )Go to πΈ Big5 β Sym tab, enter Big Five weights (see research references)
- "πΎ μ μ₯" λλ¬ λ°μClick "πΎ Save"
컀μ€ν μμ κ·μΉ (Custom Rules)Custom Clinical Rules
λ§€νΈλ¦μ€λ‘ νννκΈ° μ΄λ €μ΄ λ³΅ν© μ‘°κ±΄ ν¨ν΄μ μ¬μ©ν΄μ. μ: μ±μ€μ±β + μ κ²½μ¦β + mouth pressβ = burnout κ°μ€μΉ +0.25 For complex conditional patterns hard to express in matrices. Example: high C + high N + high mouth press = burnout boost +0.25
μ¬μ© κ°λ₯ν condition keys:
Available condition keys:
C_high, C_low, N_high, N_low, E_high, E_low, O_high, O_low, A_high, A_low, mouth_press_high, AU15_high, AU4_high, AU1_high, asymmetry_high, sanjeong_imbalance_high, fWHR_high, tired_eyelids_high
π λ°°ν¬ μν¬νλ‘Deployment Workflow
Linkage Matrix JSON λ°°ν¬ (μ 체 μν¬νλ‘)Linkage Matrix JSON Deployment
μμΈ λ¨κ³λ³ κ°μ΄λDetailed Step-by-Step
Step 1-2: Adminμμ νΈμ§ & λ€μ΄λ‘λEdit & Download in Admin
- λΈλΌμ°μ μμ
face-admin-v3.1.htmlμ΄κΈ°Openface-admin-v3.1.htmlin browser - μνλ νμμ μμ Make changes in desired tab
- κ° μΉμ μ "πΎ μ μ₯" ν΄λ¦ (λΈλΌμ°μ localStorage μ μ₯)Click "πΎ Save" in each section (saves to browser localStorage)
- πΎ I/O νμΌλ‘ μ΄λ β π₯ λ€μ΄λ‘λ ν΄λ¦Go to πΎ I/O tab β click π₯ Download
face-linkage-matrix.jsonνμΌμ΄ λ€μ΄λ‘λλ¨face-linkage-matrix.jsonfile downloaded
Step 3: GitHub μ λ‘λGitHub Upload
- GitHub μ μ₯μ μ΄κΈ°:
bostonneuromind/neurocatchersOpen GitHub repo:bostonneuromind/neurocatchers - κ²½λ‘λ‘ μ΄λ:
symptom_catcher/face-reading/config/Navigate to:symptom_catcher/face-reading/config/ - "Add file" β "Upload files" ν΄λ¦Click "Add file" β "Upload files"
- λ€μ΄λ‘λν
face-linkage-matrix.jsonλλκ·Έ μ€ λλ‘Drag and drop the downloadedface-linkage-matrix.json - Commit message μμ± (μ: "Update linkage matrix - added social_anxiety")Write commit message (e.g., "Update linkage matrix - added social_anxiety")
- "Commit changes" ν΄λ¦Click "Commit changes"
Step 4: μλ λ°°ν¬ νμΈVerify Auto-Deploy
- GitHub Actions νμμ λ°°ν¬ μ§ν νμΈ (μ½ 2-5λΆ)Check GitHub Actions tab for deployment progress (2-5 min)
- μ΄λ‘μ μ²΄ν¬ β λ¨λ©΄ λ°°ν¬ μ±κ³΅Green checkmark β means deployment succeeded
- μ€μ μ¬μ΄νΈ νμΈ:
https://neurocatchers.com/symptom_catcher/face-reading/face-reading-v3.2.htmlVerify live:https://neurocatchers.com/symptom_catcher/face-reading/face-reading-v3.2.html - λΈλΌμ°μ μΊμ λ¬Έμ μ Ctrl+Shift+R (κ°μ μλ‘κ³ μΉ¨)For cache issues: Ctrl+Shift+R (hard refresh)
face-linkage-matrix-YYYYMMDD.json νμμΌλ‘ λ°±μ
μ λ³λ μ μ₯νμΈμ. λ¬Έμ μκΈ°λ©΄ λ‘€λ°± κ°λ₯ν΄μ.
Before major changes, save a backup as face-linkage-matrix-YYYYMMDD.json. Enables rollback if issues arise.
π― μμ νλ κ°μ΄λClinical Tuning Guide
νλ μμΉTuning Principles
- μ°κ΅¬ μ°μ : κ°λ₯ν ν μ°κ΅¬ κ·Όκ±°(Kotov 2010, Kachur 2020 λ±) μ¬μ©Research first: Use research basis (Kotov 2010, Kachur 2020) when possible
- μμ κ΄μ°°λ‘ 보μ: μ€μ μ¬λ‘μμ λ°λ³΅ κ΄μ°°λλ ν¨ν΄λ§ custom rulesμ μΆκ°Clinical observation supplements: Only add repeatedly observed patterns to custom rules
- μκ² μμ: μ κ°μ€μΉλ Β±0.2 μ λλ‘ μμ, μ μ§μ μ‘°μ Start small: Begin new weights around Β±0.2, adjust gradually
- ν λ²μ νλ: μ¬λ¬ λ³κ²½μ λμμ νλ©΄ ν¨κ³Ό μΆμ λΆκ°One at a time: Multiple simultaneous changes prevent effect tracking
- κ²μ¦ μΌμ΄μ€: μλ €μ§ νλ‘νμΌμ ν μ€νΈ μ¬μ§μΌλ‘ κ²°κ³Ό κ²μ¦Validation cases: Verify with known-profile test photos
Fusion κ°μ€μΉ (Ξ±, Ξ²) μ‘°μ Fusion Weights (Ξ±, Ξ²)
| μλ리μ€Scenario | Ξ± (personality) | Ξ² (direct) |
|---|---|---|
| κ· ν κΈ°λ³Έκ° (κΆμ₯)Balanced default (recommended) | 0.60 | 0.40 |
| μ±κ²© κΈ°λ° λͺ¨λΈ κ°μ‘°Emphasize personality model | 0.75 | 0.25 |
| μ§μ μΌκ΅΄ μ νΈ κ°μ‘°Emphasize direct facial signals | 0.40 | 0.60 |
| PTSDΒ·μν λ± μΌκ΅΄ νΉμ΄ ν¨ν΄ λ§μ μ¦μSymptoms with face-specific patterns | 0.45 | 0.55 |
νλ 체ν¬λ¦¬μ€νΈTuning Checklist
- β λ³κ²½ μ JSON λ°±μ νλμ?β Backed up JSON before change?
- β λ³κ²½μ μ°κ΅¬ κ·Όκ±° λλ μμ κ·Όκ±°κ° μλμ?β Research or clinical basis for change?
- β νλμ μ¦μ/νΉμ±λ§ λ³κ²½νλμ?β Only one symptom/trait modified?
- β λ³κ²½ ν μ¬μ©μ νμ΄μ§μμ μ€μ ν μ€νΈνλμ?β Tested on actual user page?
- β Commit messageμ λ³κ²½ μ΄μ λͺ μνλμ?β Commit message explains reason?
π μ°κ΅¬ κ·Όκ±° μ°Έκ³ Research References
ν΅μ¬ λ¬Έν (νλ κ·Όκ±°)Core Literature (tuning basis)
- Ekman, P. & Friesen, W.V. (1978). Facial Action Coding System (FACS). Consulting Psychologists Press. β FACSμ μλ³Έ 체κ³β Original FACS system
- Kotov, R. et al. (2010). Linking 'big' personality traits to anxiety, depressive, and substance use disorders: A meta-analysis. Psychological Bulletin, 136(5), 768-821. β Big5 Γ μ¦μ μ°κ΄μ±μ λ©νλΆμ ν΅μ¬β Core meta-analysis for Big5 Γ symptom
- Kachur, A. et al. (2020). Assessing the Big Five personality traits using real-life static facial images. Scientific Reports, 10, 8487. β μΌκ΅΄ β Big5 μμΈ‘ (N=12,447)β Face β Big5 prediction (N=12,447)
- Cai, L. & Liu, X. (2022). Identifying Big Five personality traits based on facial behavior analysis. Frontiers in Public Health, 10, 1001828.
- Widiger, T.A. (2009). Neuroticism. In Handbook of Individual Differences in Social Behavior.
- Miller, J.D. & Lynam, D. (2001). Structural models of personality and their relation to antisocial behavior: A meta-analytic review. Criminology, 39(4), 765-798.
- Min, S.K. (2008). Clinical correlates of Hwa-Byung and a proposal for a new anger disorder. Psychiatry Investigation, 5(3), 125-141. β νλ³ μμ μ°κ΅¬ (νκ΅ λ¬Έν μ¦νκ΅°)β Hwabyeong clinical research (Korean culture-bound syndrome)
- Martel, M. (2010). Research on ADHD and Big Five personality.
- CarrΓ©, J.M. & McCormick, C.M. (2008). In your face: facial metrics predict aggressive behaviour in the laboratory. β fWHR μ°κ΅¬β fWHR research
Kotov (2010) μ£Όμ ν¨κ³Ό ν¬κΈ°Kotov (2010) Key Effect Sizes
| μ¦μDisorder | N | E | C | A |
|---|---|---|---|---|
| Depression | +1.65 β | -0.83 | -0.50 | -0.14 |
| Anxiety | +1.90 β | -0.53 | -0.26 | -0.12 |
| ADHD | +0.45 | -0.11 | -1.20 β | -0.20 |
| Substance use | +0.65 | +0.10 | -1.20 β | -0.68 |
| BPD (Widiger 2009) | +2.30 β | +0.02 | -0.80 | -0.90 |
ν¨κ³Ό ν¬κΈ° (Cohen's d): Β±0.2 μμ, Β±0.5 μ€κ°, Β±0.8 νΌ. β = κ°μ₯ κ°ν μ°κ΄. νμ¬ μμ€ν μ μ΄ κ°μ -1~+1λ‘ μ κ·νν΄ μ¬μ©. Effect sizes (Cohen's d): Β±0.2 small, Β±0.5 medium, Β±0.8 large. β = strongest links. System normalizes these to -1~+1 range.
π§ λ¬Έμ ν΄κ²°Troubleshooting
λ°°ν¬ ν λ³κ²½μ¬νμ΄ μ 보μ¬μChanges not showing after deployment
- GitHub Actionsμμ μ΄λ‘ μ²΄ν¬ νμΈ (λ°°ν¬ μλ£?)Check green checkmark in GitHub Actions (deployment complete?)
- λΈλΌμ°μ κ°μ μλ‘κ³ μΉ¨:
Ctrl+Shift+R(Win) /Cmd+Shift+R(Mac)Hard refresh:Ctrl+Shift+R(Win) /Cmd+Shift+R(Mac) - CloudFront μΊμκ° 5-10λΆ λ¨μμμ μ μμ β κΈ°λ€λ €μ£ΌμΈμCloudFront cache may persist 5-10 min β wait
- μ¬μ ν μ 보μ΄λ©΄ μν¬λ¦Ώ μ°½(incognito)μμ νμΈIf still not showing, try incognito window
Admin νμ΄μ§ μμ μ΄ μ¬μ©μ νμ΄μ§μ λ°μ μ λΌμAdmin edits not reflecting in user page
- μμΈ: Adminμ λΈλΌμ°μ localStorageμλ§ μ μ₯. JSON λ€μ΄λ‘λ & GitHub μ λ‘λ λΉ λ¨λ¦Ό.Cause: Admin only saves to browser localStorage. Missed JSON download & GitHub upload.
- ν΄κ²°: πΎ I/O νμμ π₯ λ€μ΄λ‘λ β GitHubμ
config/face-linkage-matrix.jsonμ μ λ‘λSolution: πΎ I/O tab β π₯ Download β upload to GitHub'sconfig/face-linkage-matrix.json
"μΌκ΅΄μ μ°Ύμ§ λͺ»νμ΄μ" μ€λ₯κ° μμ£Ό λμ"Face not found" error is frequent
- μμΈ 1: μ¬μ§μ΄ μ΄λ‘κ±°λ νλ¦Ών¨ β μ¬μ©μμκ² μ¬μ§ κ°μ΄λ μλ΄Cause 1: Dark/blurry photo β guide user to photo guide
- μμΈ 2: μλͺ¨μ΅μ΄κ±°λ μΌκ΅΄μ΄ μλ Έμ β μ λ©΄ μ¬μ§ κΆμ₯Cause 2: Side view or cropped face β request front-facing
- μμΈ 3: MediaPipe λͺ¨λΈ λ‘λ μ€ν¨ β λ€νΈμν¬ νμΈ, νμ΄μ§ μλ‘κ³ μΉ¨Cause 3: MediaPipe load fail β check network, refresh
JSON λΆλ¬μ€κΈ° μ€λ₯JSON Import Error
- νμ κ²μ¦ μ€ν¨:
symptomsμbigfive_to_symptomμ΄ μμ΄μΌ ν¨ (μ΅μ νμ νλ)Format check fail: must havesymptomsandbigfive_to_symptom(min required) - JSON syntax μ€λ₯: μ½€λ§Β·λκ΄νΈ νμΈ. jsonlint.comμμ κ²μ¦JSON syntax error: check commas/brackets. Validate on jsonlint.com
πΏ λλ§Ίλ λ§Closing Note
μ΄ μμ€ν μ μ΄μμλ λꡬμμ. μμ μ¬λ‘κ° μμΌμλ‘, μ°κ΅¬κ° μ λ°μ΄νΈλ μλ‘ κ³μ μ κ΅ν΄μ§λλ‘ μ€κ³λμ΄μ. λΉμ μ νλμ΄ νμΒ·λ΄λ΄μμ λ κΉμ μ΄ν΄λ‘ μ΄μ΄μ§κΈ°λ₯Ό ποΈ This system is a living tool. Designed to grow more refined as clinical cases accumulate and research updates. May your tuning lead to deeper understanding of patients and clients ποΈ
β Boston Neuromind Β· Canton, MA
v3.1 Β· 2026.04
β Boston Neuromind Β· Canton, MA
v3.1 Β· April 2026