🧠 Boston Neuromind
USPTO κ°€μΆœμ› μ§„ν–‰ 쀑 Β· 2026USPTO Provisional Applications Β· 2026

Boston Neuromind νŠΉν—ˆ 포트폴리였 Boston Neuromind Patent Portfolio

μ •μ‹ κ³Ό 진단과 λ””μ§€ν„Έ μ •μ‹ κ±΄κ°•μ˜ νŒ¨λŸ¬λ‹€μž„μ„ λ°”κΎΈλŠ” 4건의 μ•Œκ³ λ¦¬μ¦˜ νŠΉν—ˆ. 8개 데이터 μ†ŒμŠ€λ₯Ό λ² μ΄μ§€μ•ˆ μΆ”λ‘ μœΌλ‘œ ν†΅ν•©ν•˜κ³ , 18개 DSM-5-TR 진단을 λ™μ‹œμ— μ‚°μΆœν•˜λ©°, 인간-AI λŒ€ν™”μ˜ 신뒰도λ₯Ό κ°κ΄€μ μœΌλ‘œ μΈ‘μ •ν•˜λŠ” μ‹œμŠ€ν…œ. Four algorithm patents reshaping the paradigms of psychiatric diagnosis and digital mental health: a system that fuses eight data sources through Bayesian inference, simultaneously yields eighteen DSM-5-TR diagnoses, and objectively measures the reliability of human-AI conversation.

4μ•Œκ³ λ¦¬μ¦˜ νŠΉν—ˆAlgorithm Patents
8데이터 μ†ŒμŠ€ 톡합Data Sources Fused
18DSM-5-TR 진단DSM-5-TR Diagnoses
103인용 λ…Όλ¬ΈCited References

πŸ“‹ 이 포트폴리였λ₯Ό 처음 λ³΄μ‹œλŠ” λΆ„κ»˜πŸ“‹ If You're New to This Portfolio

μ•„λž˜ 4건의 νŠΉν—ˆλŠ” λͺ¨λ‘ μ‹€μ œ μž‘λ™ν•˜λŠ” μ‹œμŠ€ν…œ(agedlearning.com)에 κ΅¬ν˜„λ˜μ–΄ 있으며, Boston Neuromind LLC의 μž„μƒ λ°μ΄ν„°λ‘œ κ²€μ¦λ˜κ³  μžˆμŠ΅λ‹ˆλ‹€. 핡심은 DMDA(Patent 4)이며, λ‚˜λ¨Έμ§€ 3건은 DMDAλ₯Ό λ‘˜λŸ¬μ‹Ό 보쑰 발λͺ…μœΌλ‘œ μ‹œλ„ˆμ§€λ₯Ό μ΄λ£Ήλ‹ˆλ‹€. 각 νŠΉν—ˆλŠ” μ²­κ΅¬ν•­Β·λ„λ©΄Β·μ„ ν–‰κΈ°μˆ  비ꡐ·관련 λ…Όλ¬Έ(References)을 ν¬ν•¨ν•œ USPTO κ°€μΆœμ› ν‘œμ€€ ν˜•μ‹μœΌλ‘œ μž‘μ„±λ˜μ–΄ μžˆμŠ΅λ‹ˆλ‹€. All four patents below are implemented in a real, working system (agedlearning.com) and are being validated with clinical data from Boston Neuromind LLC. DMDA (Patent 4) is the core, and the other three patents are supporting inventions that synergize around it. Each patent is written in USPTO provisional-application format, including claims, drawings, prior-art comparisons, and references.

πŸ›οΈ 핡심 νŠΉν—ˆ (Flagship)Flagship Patent

PATENT 4 Β· FLAGSHIP

DMDA β€” 4-Tier Progressive 닀쀑 μ†ŒμŠ€ 진단 μ•Œκ³ λ¦¬μ¦˜ DMDA β€” 4-Tier Progressive Multi-Source Diagnostic Algorithm

DMDA = DSM-5-TR Multi-Source Diagnostic AlgorithmDSM-5-TR Multi-Source Diagnostic Algorithm

μ„€λ¬Έ, 타이핑 νŒ¨ν„΄, 폰 μˆ˜λ™ 데이터, qEEG, HRV, μ–Όκ΅΄, μŒμ„±, ERP β€” 8개 이질적 데이터 μ†ŒμŠ€λ₯Ό λ² μ΄μ§€μ•ˆ 닀쀑 λͺ¨λ“ˆ μœ΅ν•©(Bayesian Multi-Module Fusion)으둜 톡합. 데이터 μ†ŒμŠ€κ°€ 좔가될 λ•Œλ§ˆλ‹€ 진단 신뒰도가 60% β†’ 95%둜 단쑰 μ¦κ°€ν•˜λŠ” μˆ˜ν•™μ  보μž₯(μ—”νŠΈλ‘œν”Ό 정리). 18개 DSM-5-TR 진단을 λ™μ‹œμ— 사후 ν™•λ₯ λ‘œ μ‚°μΆœν•˜λŠ” 세계 졜초 μ‹œμŠ€ν…œ. Integrates eight heterogeneous data sources β€” survey, typing pattern, phone-passive data, qEEG, HRV, face, voice, and ERP β€” via Bayesian Multi-Module Fusion. As data sources are added, diagnostic confidence increases monotonically from 60 % to 95 % with a mathematical guarantee (entropy theorem). The first system in the world to simultaneously output posterior probabilities for eighteen DSM-5-TR diagnoses.

8 데이터 μ†ŒμŠ€Data Sources μ„€λ¬Έ + 타이핑 + 폰 + qEEG + HRV + μ–Όκ΅΄ + μŒμ„± + ERP survey + typing + phone + qEEG + HRV + face + voice + ERP
4 Progressive TierProgressive Tiers Tier 1 (60-75%) β†’ Tier 4 (95-98%)
18 DSM-5-TR 진단DSM-5-TR Diagnoses ADHD, MDD, GAD, PTSD, OCD, Bipolar...
94% μ—”μ§„ 정확도Engine Accuracy Boston Neuromind μž„μƒ 데이터 검증 validated on Boston Neuromind clinical data
USPTO Class: G16H 50/20 FDA: 510(k) Class II 경둜pathway 청ꡬ항Claims: 10 References: 35
전체 λͺ…μ„Έμ„œ 보기 β†’View Full Specification β†’

πŸ“‘ 보쑰 νŠΉν—ˆ 3건Three Supporting Patents

PATENT 1

Fischer 동적 기술 μˆ˜μ€€ 적응 (Fischer DSL Adaptation) Fischer Dynamic Skill Level Adaptation

Harvard Mind, Brain & Education의 Kurt Fischer κ΅μˆ˜κ°€ μ •λ¦½ν•œ 13단계 동적 기술 이둠을 μ μš©ν•΄, μ‚¬μš©μžμ˜ μ–Έμ–΄ λ³΅μž‘λ„(μ–΄νœ˜Β·κ΅¬λ¬ΈΒ·κ²°μ†)λ₯Ό μ‹€μ‹œκ°„ μΆ”μΆœν•˜κ³  AI 응닡 λ³΅μž‘λ„λ₯Ό λ™μ μœΌλ‘œ μ‘°μ •ν•˜λŠ” μ•Œκ³ λ¦¬μ¦˜. 발λͺ…μžκ°€ Fischer κ΅μˆ˜μ—κ²Œ 직접 사사받은 ν•™μˆ  μžμ‚°μ„ μ•Œκ³ λ¦¬μ¦˜μœΌλ‘œ κ΅¬ν˜„. An algorithm that applies the 13-level Dynamic Skill Theory of Harvard's Prof. Kurt Fischer (Mind, Brain & Education): real-time extraction of a user's linguistic complexity (lexical, syntactic, cohesive features) is used to dynamically adjust AI-response complexity. The inventor β€” directly mentored by Prof. Fischer β€” implements this academic legacy as an algorithm.

μ–Έμ–΄ νŠΉμ„± μΆ”μΆœκΈ°Linguistic Feature Extractor μ–΄νœ˜ λ‹€μ–‘μ„±, ꡬ문 λ³΅μž‘λ„, 응집성 lexical diversity, syntactic complexity, cohesion
13 동적 기술 μˆ˜μ€€Dynamic Skill Levels Sensorimotor β†’ Reflective
USPTO: G06F 40/35 / G06N 5/04 청ꡬ항Claims: 9 Refs: 18
λͺ…μ„Έμ„œ 보기 β†’View Spec β†’
PATENT 2

QEEG-λŒ€ν™” νŒŒλΌλ―Έν„° λ§€ν•‘ QEEG-to-Conversation Parameter Mapping

19채널 μ •λŸ‰λ‡ŒνŒŒ(qEEG)μ—μ„œ μΆ”μΆœλœ λ°”μ΄μ˜€λ§ˆμ»€(Theta/Beta Ratio, Frontal Alpha Asymmetry, Peak Alpha Frequency λ“±)λ₯Ό AI λŒ€ν™” μ‹œμŠ€ν…œμ˜ νŒŒλΌλ―Έν„°(응닡 속도, μ •μ„œ 톀, 인지 λΆ€ν•˜)에 직접 λ§€ν•‘ν•˜λŠ” 세계 졜초의 μ•Œκ³ λ¦¬μ¦˜. BCN(Board Certified in Neurofeedback) μžκ²©μ„ κ°€μ§„ 발λͺ…μžλ§Œ 섀계 κ°€λŠ₯ν•œ 독점 기술. The first algorithm in the world to directly map biomarkers extracted from 19-channel quantitative EEG (qEEG) β€” including Theta/Beta Ratio, Frontal Alpha Asymmetry, and Peak Alpha Frequency β€” onto the parameters of an AI conversation system (response latency, affective tone, cognitive load). A proprietary technology that only an inventor holding Board Certification in Neurofeedback (BCN) can design.

5 QEEG λ°”μ΄μ˜€λ§ˆμ»€Biomarkers TBR, FAA, PAF, COH, PAP
4 λŒ€ν™” νŒŒλΌλ―Έν„°Conversation Parameters 속도, 톀, λΆ€ν•˜, 곡감 깊이 latency, tone, load, empathy depth
USPTO: A61B 5/377 / G06N 3/04 청ꡬ항Claims: 8 Refs: 23
λͺ…μ„Έμ„œ 보기 β†’View Spec β†’
PATENT 3

μ‹œκ°„μ  μΌμΉ˜μ„± 탐지 (Temporal Congruence Detection) Temporal Congruence Detection

μ–Όκ΅΄ ν‘œμ •(FACS), μŒμ„± 운율, ν…μŠ€νŠΈ λ‚΄μš©μ˜ μ‹œκ°„μ  뢈일치 νŒ¨ν„΄μ„ 탐지해, μžκΈ°λ³΄κ³ λ‘œλŠ” λ“œλŸ¬λ‚˜μ§€ μ•ŠλŠ” κ°€λ©΄ 우울(masked depression), μ•Œλ ‰μ‹œν‹°λ―Έμ•„, μžμ‚΄ μœ„ν—˜ μ‹ ν˜Έλ₯Ό 포착. 동적 μ‹œκ°„ μ›Œν•‘(DTW)κ³Ό 닀쀑 λͺ¨λ‹¬ μ •μ„œ μœ΅ν•©μ„ κ²°ν•©ν•œ μ•Œκ³ λ¦¬μ¦˜. "말과 ν‘œμ •κ³Ό 글이 같은 λ°©ν–₯을 ν–₯ν•˜λŠ”κ°€?"λΌλŠ” μ§ˆλ¬Έμ— κ°κ΄€μ μœΌλ‘œ λ‹΅ν•œλ‹€. An algorithm that detects temporal incongruence across facial expressions (FACS), vocal prosody, and textual content to capture masked depression, alexithymia, and suicide-risk signals invisible to self-report. It combines Dynamic Time Warping (DTW) with multi-modal affect fusion to objectively answer the question: "Do voice, face, and text point in the same direction?"

3 λͺ¨λ‹¬λ¦¬ν‹°Modalities μ–Όκ΅΄(FACS) + μŒμ„± + ν…μŠ€νŠΈ face (FACS) + voice + text
DTW μ‹œκ°„ μ •λ ¬Temporal Alignment μ‹œκ°„ μ°¨ ≀ 200ms 뢄석 time-lag ≀ 200 ms analysis
USPTO: G06V 40/16 / G10L 25/63 청ꡬ항Claims: 9 Refs: 27
λͺ…μ„Έμ„œ 보기 β†’View Spec β†’

πŸ‘¨β€πŸ”¬ 발λͺ…μž 자격Inventor Credentials

이 4건의 νŠΉν—ˆλŠ” λ‹€μŒκ³Ό 같은 ν¬μ†Œν•œ 자격 μ‘°ν•© μœ„μ—μ„œλ§Œ 섀계 κ°€λŠ₯ν•©λ‹ˆλ‹€ These four patents can only be designed atop the following rare combination of credentials

BCN Board Certified in Neurofeedback β€” qEEG μž„μƒ λΆ„μ„μ˜ λ―Έκ΅­ 졜고 μˆ˜μ€€ 자격 Board Certified in Neurofeedback β€” the highest U.S. credential for clinical qEEG analysis
Harvard Visiting Scholar Mind, Brain & Education / Kurt Fischer ꡐ수 직접 사사 Mind, Brain & Education / directly mentored by Prof. Kurt Fischer
Instructional Design PhD λ―Έκ΅­ Top 20 λŒ€ν•™μ› 박사 ν•™μœ„ Doctoral degree from a top-20 U.S. graduate school
Mental Health Master's GPA 4.0/4.0 (μƒμœ„ 1-3%) GPA 4.0/4.0 (top 1–3 %)
3+ λ…„ μž„μƒ κ²½ν—˜Years of Clinical Experience μ‹€μ œ ν΄λΌμ΄μ–ΈνŠΈ 데이터 기반 μ•Œκ³ λ¦¬μ¦˜ 검증 algorithms validated on real client data
MA μŠˆνΌλ°”μ΄μ € ν•˜ μž„μƒMA Supervised Clinician 맀사좔세츠 μ£Ό 합법적 μž„μƒ ν™œλ™ + 2-3λ…„ ν΄λΌμ΄μ–ΈνŠΈ retention licensed clinical practice in Massachusetts; 2–3-year client retention

🎯 μΆœμ› μ „λž΅Filing Strategy

μ™œ 4건을 ν•¨κ»˜ μΆœμ›ν•˜λŠ”κ°€?Why File All Four Together?

πŸ’° μ˜ˆμƒ λΉ„μš©Estimated Costs

ν•­λͺ©Item 개수Count 건당 λΉ„μš©Per-Item Cost μ†Œκ³„Subtotal
Patent 4: DMDA ⭐ Flagship 1 $2,000–3,000 $2,000–3,000
Patent 1: Fischer DSL1$1,500–2,500$1,500–2,500
Patent 2: QEEG-Conversation1$1,500–2,500$1,500–2,500
Patent 3: Temporal Congruence1$1,500–2,500$1,500–2,500
총 USPTO Provisional μΆœμ› λΉ„μš©Total USPTO Provisional Cost 4 β€” $6,500–10,500

* μœ„ λΉ„μš©μ€ κ°€μΆœμ›(provisional) 기쀀이며, 1λ…„ λ‚΄ 정식(non-provisional) μ „ν™˜ μ‹œ μΆ”κ°€ λΉ„μš© λ°œμƒ. DMDAλŠ” FDA 510(k) 경둜 별도 λΉ„μš© 별도 μ‚°μ •. * Costs above are for provisional filings; conversion to non-provisional within one year incurs additional expense. DMDA also requires a separately budgeted FDA 510(k) pathway.