From the analysed documents and approved inputs, it appears that 3 closest documents require separate novelty and inventive-step review. The report preserves the evidence path and amendment decisions behind that assessment.
◍Assessment basis. This agentic assessment uses the approved claim, technical disclosure, search inputs and 3 documents carried into legal analysis. Novelty, inventive step and claim amendment are recorded separately. It reflects what was public as of your filing/priority date, 2026-03-01.
Patentopia Demonstration · sample@patentopia.example · 3 April 2026 Prepared by Christopher Holmgaard · Report ID mock-sample-001
Demonstration record: the claim, documents, evidence excerpts, review identities and decisions on this page are synthetic and show the completed workflow only.
Assessment overview
The record separates novelty, inventive step and claim drafting. Status is shown from traceable evidence and approval receipts; numerical model estimates are not presented as patentability probabilities.
Novelty recordAnalysis completed
5 of 5 counted features have a recorded relationship across 3 documents.
Inventive-step recordAnalysis completed
The recorded Art. 56 analysis is available below.
Claim drafting recordAnalysis available
3 drafting directions retained for traceability; the recorded support, professional review and user decisions produced the clean claim shown below.
Executive assessment
Novelty and inventive step
Moderate
5 of 5 counted features have a recorded relationship across 3 documents. The separate Art. 54 and Art. 56 records appear below.
Claim scope
Moderate
No high-severity breadth weakness was recorded in the submitted claim. Full specification support remains a separate question.
Design-around exposure
Moderate
2 signals that a competitor could design around the claim as drafted; see the suggested amendments.
Enforceability
Not assessed on this run
Not assessed.
Data
Not assessed on this run
Not assessed.
Unity
Not assessed on this run
Not assessed.
Search record
The record below distinguishes database scale, retrieved material and the documents carried into legal analysis. Exclusions are reported explicitly; user-supplied prior art is retained in the analysis set.
SourcesEPO Open Patent Services, Crossref, USPTO, User-supplied prior art
Languages representedEN
Approved / executed queries3 / 3
User-supplied prior art1 forced into review
1
Database-reported matches
Not recorded
2
Retrieved records
121
3
Unique documents after deduplication23 excluded: Duplicate record or patent-family representation
98
4
Ranked candidates73 excluded: Outside the ranked candidate window
25
5
Documents carried into legal analysis22 excluded: Outside the approved closest-document set
3
Recorded provenance: 121 retrieved -> 98 after dedup -> 25 ranked -> 3 closest.
User-supplied prior art
ISO 22523:2006 — External limb prostheses and orthoses — Requirements and test methodsPublication identifier not recorded
Forced inclusion
Yes
Material coverage
Not recorded
Publication status
Date not verified
Matrix status
Retained in the evidence set; outside the closest-document view
Current candidate starting point, subject to verification
This document has the strongest verified relationship to the approved feature model and is the current starting point for the separate novelty and inventive-step analyses.
The claim is represented as patent-relevant technical relationships, not grammatical fragments. Search and legal conclusions use the approved conversion recorded below.
Approved source claim
The complete claim text used for feature traceability and search approval.
A smart adaptive insole system comprising (a) a flexible substrate configured to be inserted into footwear; (b) an array of pressure sensors embedded within the substrate for detecting plantar pressure distribution in real-time; (c) a wireless communication module that transmits pressure data to a mobile device; and (d) a machine learning algorithm running on the mobile device that analyzes gait patterns and generates injury prevention recommendations based on historical pressure data and biomechanical models.
B
An array of pressure sensors embedded within the substrate for detecting plantar pressure distribution in real-time
importance: high
From claim: “an array of pressure sensors embedded within the substrate for detecting plantar pressure distribution in real-time”
ConfirmedExtraction confidence: high
High importanceHigh threat
D
A machine learning algorithm running on the mobile device that analyzes gait patterns
importance: high
From claim: “a machine learning algorithm running on the mobile device that analyzes gait patterns”
ConfirmedExtraction confidence: high
High importanceHigh threat
E
Generates injury prevention recommendations based on historical pressure data and biomechanical models
importance: high
From claim: “generates injury prevention recommendations based on historical pressure data and biomechanical models”
ConfirmedExtraction confidence: high
High importanceHigh threat
C
A wireless communication module that transmits pressure data to a mobile device
importance: medium
From claim: “a wireless communication module that transmits pressure data to a mobile device”
ConfirmedExtraction confidence: high
Medium importanceSubstantial threat
A
A flexible substrate configured to be inserted into footwear
importance: low
From claim: “a flexible substrate configured to be inserted into footwear”
ConfirmedExtraction confidence: high
Low importanceHigh threat
Retrieval validation
PassedRequired diagnostics, feature approval and publication-family verification passed.
Diagnostic
Expected
Observed
Status
Feature A: A flexible substrate configured to be inserted into footwear
Targeted retrieval and completed material review
3 mapped references
Passed
Feature B: An array of pressure sensors embedded within the substrate for detecting plantar pressure distribution in real-time
Targeted retrieval and completed material review
2 mapped references
Passed
Feature C: A wireless communication module that transmits pressure data to a mobile device
Targeted retrieval and completed material review
2 mapped references
Passed
Feature D: A machine learning algorithm running on the mobile device that analyzes gait patterns
Targeted retrieval and completed material review
1 mapped reference
Passed
Feature E: Generates injury prevention recommendations based on historical pressure data and biomechanical models
Targeted retrieval and completed material review
1 mapped reference
Passed
Feature conversionApproved
Publication-family verificationVerified
Evidence objectsVerified
Disclosure matrix and novelty
Partial feature overlap: 5 of 5 counted claim features appear in 3 documents; no single document discloses all of them ; no identified threat to the claim as a whole. Per-document overlap: 1-P (features A, B and C), 2-A (features A, B, D and E) and 3-A (features A and C). No single-document novelty-destroying disclosure (EPC Art. 54) is established on the reviewed evidence; overlap of separate features across different documents is a matter for inventive step (EPC Art. 56), not novelty.
The strategic matrix uses only patent-relevant technical relationships. Select a cell to inspect the publication, locator, reviewed material, language and verification record. A blank cell is not evidence of absence.
Extreme
One document describes every counted feature; your claim may not be new.
High
One document describes most counted features, missing only a minor one.
Substantial
One document describes several features, including an important one.
Moderate
One document describes one important feature, or a few minor ones.
Limited
One document describes part of a single feature.
Minimal
The record maps only a fragment of one counted feature.
No identified threats
The empty state, shown when nothing counts against you at this level.
Disclosure matrix · which document discloses which feature
Strategic disclosure view5 search-relevant features
F verified full; P verified partial; I verified implicit; R imported relationship requiring verification. A blank cell has no verified evidence object and is not proof of absence. Select any cell to inspect its evidence object.
Reviewed prior art · 3 documents
Real-time Gait Analysis Using Wearable Pressure Sensors
Journal article · 2023 · discloses 4 features
High▸
This document discloses 4 claim features, rated high on the named scale.
Feature A · A flexible substrate configured to be inserted into footwear
Disclosed at Section 2.1; page 3
“The insole substrate was fabricated from flexible thermoplastic polyurethane.”
Feature B · An array of pressure sensors embedded within the substrate for detecting plantar pressure ... [truncated]
Disclosed at Section 2.2; page 4
“Sixteen FSR sensors were placed at key plantar regions for continuous pressure distribution measurement at 100 Hz.”
Feature D · A machine learning algorithm running on the mobile device that analyzes gait patterns
Disclosed at Section 3.1; page 7
“A convolutional neural network was deployed on the smartphone for real-time gait phase classification.”
Feature E · Generates injury prevention recommendations based on historical pressure data and ... [truncated]
Disclosed at Section 4.2; page 10
“Longitudinal pressure features were compared with a generic biomechanical risk model to produce injury-risk guidance.”
Inventive step: 1 of 2 assessed features (D) has a recorded obviousness route via 1-P and 2-A (EPC Art. 56). For the remaining 1 feature, the reviewed routes did not establish that obvious-development conclusion.
Inventive step is organised by document-combination route, not by labelling isolated features obvious or non-obvious.
Route 1Recorded as potentially obvious
Starting point
1-P
Secondary reference
2-A
Shared features
Feature D
Distinguishing relationship
Mobile machine-learning analysis of gait patterns
Technical effect
Not verified from the current evidence record
Objective technical problem
How to process collected plantar pressure data to provide actionable injury prevention guidance, rather than raw data visualization or simple gait classification.
Reason to combine
1-P provides the sensor platform and 2-A expressly deploys a gait classifier on a smartphone, giving a direct implementation route for this limitation at its present breadth.
Reason not to combine
Countervailing reason not recorded; professional review required.
Route 2Recorded as not established
Starting point
1-P
Secondary reference
2-A
Shared features
Feature E
Distinguishing relationship
Historical pressure analysis combined with a personalized biomechanical model to generate injury-prevention guidance
Technical effect
Not verified from the current evidence record
Objective technical problem
How to process collected plantar pressure data to provide actionable injury prevention guidance, rather than raw data visualization or simple gait classification.
Reason to combine
No supported reason to combine was established on the reviewed route.
Reason not to combine
2-A uses a generic risk model, but the reviewed record does not establish a reason to adapt the model continuously to an individual across multiple sessions.
Adversarial review
⚑ Substantive claim risks
The following points challenge the technical contribution and claim breadth rather than explaining system limitations.
A broader or better-targeted search may identify the claimed technical relationship in an earlier publication.
The distinguishing relationship may be an arbitrary selection or an expected implementation choice.
The stated technical effect may not be supported across the full breadth of the approved claim.
A functional limitation may state only a desired result without the technical means that achieve it.
Evidence limitations
A focused search for adaptive, user-specific biomechanical models could still identify a stronger combination route. The conclusion is limited to the reviewed material and verified dates listed below.
Feature A is disclosed by "Smart Insole for Plantar Pressure Monitoring" (Art. 54).
Feature B is disclosed by "Smart Insole for Plantar Pressure Monitoring" (Art. 54).
Feature C is disclosed by "Smart Insole for Plantar Pressure Monitoring" (Art. 54).
Feature D is disclosed by "Real-time Gait Analysis Using Wearable Pressure Sensors" (Art. 54).
Feature E is disclosed by "Real-time Gait Analysis Using Wearable Pressure Sensors" (Art. 54).
[exposure] Feature A is mapped in all three reviewed closest-art documents and does not add a distinguishing relationship on this record.
[exposure] Feature D overlaps with the CNN-based gait classifier in 2-A. The claim should be narrowed to distinguish the specific ML approach.
A challenger would combine the smart-insole platform in 1-P with the smartphone classifier in 2-A and argue that longitudinal risk guidance is an expected analytics objective. The strongest response depends on the verified user-specific adaptation and multi-session relationship.
Document 2-A shows that machine learning can be applied to gait data. The reviewed record does not establish a reason to adapt its generic model continuously to an individual across multiple sessions.
The reviewed route treats these as distinguishing relationships requiring professional review: Injury prevention recommendations (not just classification); Historical pressure data analysis over time; Integration of biomechanical models with ML analysis (EPC Art. 56).
Claim amendment workflow
Amendment directions follow from the analysis above. Each direction records the cited threat, minimum wording change, support basis, scope effect and novelty or inventive-step effect.
Amend Feature D · Approved for clean claim
Rationale: Narrowing Feature D to specify the temporal/multi-session aspect and personalized biomechanical model creates distance from 2-A (which uses a generic CNN on single-session data). This is the most impactful amendment because it strengthens the technically distinguishing relationship in the disclosure. Distinguishes from 1-P, 2-A.
Minimum verified amendment:a machine learning algorithm running on the mobile device that analyzes gait patterns → a recurrent neural network running on the mobile device that analyzes temporal gait patterns across multiple sessions using a personalized biomechanical model adapted to the individual user's foot morphology Basis (paragraph [0042]): “a recurrent neural network running on the mobile device that analyzes temporal gait patterns across multiple sessions using a personalized biomechanical model adapted to the individual user's foot morphology”.
Scope effect: Narrows the model architecture, observation period and biomechanical-model relationship.
Novelty / inventive-step effect: Specifying multi-session temporal analysis and a personalized biomechanical model distances the claim from 2-A on inventive step ; 2-A discloses only single-session CNN gait classification.
Distance verification: Verified by Demo patent professional · 2026-04-03T10:13:00Z.
Amend Feature B · Approved for clean claim
Rationale: Specifying sensor count, type (capacitive vs piezoresistive in 1-P), anatomical mapping, and sampling rate creates technical distance from both 1-P and 2-A while narrowing to the actual implementation. Distinguishes from 1-P, 2-A.
Minimum verified amendment:an array of pressure sensors embedded within the substrate for detecting plantar pressure distribution in real-time → an array of at least twelve capacitive pressure sensors arranged in anatomically-mapped zones within the substrate for detecting plantar pressure distribution at a sampling rate of at least 100 Hz Basis (paragraph [0035]): “an array of at least twelve capacitive pressure sensors arranged in anatomically-mapped zones within the substrate for detecting plantar pressure distribution at a sampling rate of at least 100 Hz”.
Novelty / inventive-step effect: Capacitive sensors in anatomically-mapped zones at ≥100 Hz distance the claim from 1-P on novelty ; 1-P discloses generic piezoresistive sensors without a specified sampling rate.
Distance verification: Verified by Demo patent professional · 2026-04-03T10:13:30Z.
Add new feature · Rejected
Rationale: Adding an adaptive/learning element would narrow Feature E toward a relationship that is not mapped in the retained passages from 1-P, 2-A or 3-A. Professional review is still required before relying on that distance. Distinguishes from 1-P, 2-A, 3-A.
Minimum verified amendment: wherein the biomechanical model is continuously updated based on accumulated user data to refine injury risk predictions over time Basis (paragraph [0048]): “the biomechanical model is continuously updated based on accumulated user data to refine injury risk predictions over time”.
Scope effect: Narrows the recommendation function to a continuously updated personalised model.
Novelty / inventive-step effect: The retained passages from 1-P, 2-A and 3-A do not map continuous, user-specific refinement of injury-risk predictions. This is an evidence-limited distance record, not a broader inventive-step conclusion.
Distance verification: Verified by Demo patent professional · 2026-04-03T10:13:45Z.
Final approved claimClean version from the recorded amendment decisions
A smart adaptive insole system comprising (a) a flexible substrate configured to be inserted into footwear; (b) an array of at least twelve capacitive pressure sensors arranged in anatomically-mapped zones within the substrate for detecting plantar pressure distribution at a sampling rate of at least 100 Hz; (c) a wireless communication module that transmits pressure data to a mobile device; and (d) a recurrent neural network running on the mobile device that analyzes temporal gait patterns across multiple sessions using a personalized biomechanical model adapted to the individual user's foot morphology and generates injury prevention recommendations based on historical pressure data and biomechanical models.
Version history
Source claim03/04/2026, 10:00:00
Strategy generated03/04/2026, 10:12:00
Review record2 amendment decisions recorded · 03/04/2026, 10:15:00
Optional assessment terminology
Novelty
Whether all assessed claim limitations are disclosed together in one earlier public document (EPC Art. 54).
Inventive step
Whether the claimed combination follows obviously from the prior-art record for the skilled person (EPC Art. 56).
Defensibility: drafting quality
How well your claim is written to survive challenge. A separate measure from patentability.
Critical / filing date
Documents made public before this date can count against you; later ones can't.
Counted vs generic
Counted features carry weight; generic building blocks are set aside and never scored as threats.
Fail-closed
When date receipts are incomplete, the system withholds a conclusion instead of guessing.
Agentic patentability assessment generated from approved inputs and recorded evidence. It supports patent-professional review and does not replace professional judgment.
Patentopia · patentability assessment · 3 April 2026 Evidence, source and approval status are drawn from the stored assessment record.