Assessment Vector Similarity Search for Nursing Record Support

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing assessment systems for nursing records lack reliability in finding similar patients for assessment creation support due to insufficient consideration of multifaceted nursing information, leading to low search performance.

Innovation Solution

An assessment support system that includes an assessment prediction unit to vectorize patient information, a degree-of-similarity calculation unit to determine similarity, and a search unit to find similar patients based on calculated similarity, utilizing a prediction model trained on actual patient data.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a symptom vector-based similarity detection system is used to find similar patients, then the system can identify patients with similar symptoms, but the search performance is low and reliability is insufficient for assessment creation support

Engineering Contradiction:

Improvereliability of assessment creation supportVSAvoidsearch performance

Core Design Contradiction:

ReliabilityVSMeasurement precision

Solution Approach 1:

The patent transforms the assessment data into vector representations (assessment vectors) in a multi-dimensional space, changing the parameter representation from traditional structured data to continuous vector embeddings. This enables more nuanced similarity calculation by considering multifaceted nursing information including nursing diagnoses, interventions, and patient responses, thereby improving both search performance and reliability for finding similar patients.

Inventive Principle:

Principle #35Parameter changes

2Ease of operation

If only symptom vectors are used for patient similarity detection, then the system is simple to operate, but multifaceted nursing information such as nursing system and medical equipment is not sufficiently considered

Engineering Contradiction:

Improvesimplicity of system operationVSAvoidloss of multifaceted nursing information

Core Design Contradiction:

Ease of operationVSLoss of information

Solution Approach 1:

The patent merges multiple types of nursing information (nursing diagnoses, interventions, patient responses, nursing system data, and medical equipment information) into a unified assessment vector representation. This combining approach preserves comprehensive nursing information while maintaining system operational simplicity through automated vector processing and similarity calculation.

Inventive Principle:

Principle #5Merging (Combining)

3Productivity

If the assessment content of similar patients is used as reference for assessment creation, then the system can provide support, but the content may not be sufficiently relevant due to limited similarity criteria

Engineering Contradiction:

Improveefficiency of assessment creationVSAvoidrelevance of reference assessment content

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The patent replaces traditional mechanical similarity comparison methods with machine learning-based vector embedding and similarity calculation. This substitution enables the system to automatically capture complex relationships in nursing data, improving the relevance of reference assessment content while maintaining high efficiency in assessment creation through automated processing.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS12537101B2System for supporting decision-making regarding assessment using a machine learning-trained model, assessment support method, and recording medium

Publication Date: 2026.01.27 NEC CORP

AI summary

An assessment support system includes: an assessment prediction unit that predicts, based on patient information about a target patient who is a creation target for an assessment in a nursing record, an assessment vector obtained by vectorizing the assessment of the target patient, as a prediction assessment vector; a degree-of-similarity calculation unit that calculates a degree of similarity of the assessment vector to the prediction assessment vector, based on a relationship between the predicted prediction assessment vector and the assessment vector of a patient having the assessment recorded in the nursing record; and a search unit that searches for and outputs at least one similar patient who is similar to the target patient, based on the degree of similarity.