MCI Progression Prediction Using Four Clinical Variables

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Accurately predicting the progression of mild cognitive impairment (MCI) or mild Alzheimer's dementia (mild AD) is challenging due to lighter symptoms, and existing methods often impose a heavy processing load on patients and equipment.

Innovation Solution

An information processing device and method that utilizes a prediction model trained with variables such as orientation, memory, biomarkers, and IADL activities to predict the probability of symptom advancement in MCI or mild AD, using input variables like orientation to time, delayed recall, memory of holidays and medication schedules, and amyloid β status.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If complex prediction models with many variables are used to accurately predict MCI or mild AD progression, then prediction accuracy is improved, but processing load on patients and equipment increases

Engineering Contradiction:
Improveprediction accuracyVSAvoidprocessing load
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent extracts and identifies the four most critical variables from among many potential predictors: orientation to time, delayed recall, memory of IADL events, and amyloid β status. By selecting only these essential variables, the model achieves accurate prediction while minimizing processing load on patients and equipment.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent changes the parameters by focusing on specific, measurable biomarkers and cognitive functions rather than comprehensive assessments. The use of amyloid β status as a binary positive/negative variable and standardized cognitive test scores transforms complex clinical data into manageable parameters that maintain prediction accuracy while reducing complexity.

Inventive Principle:
Principle #35Parameter changes

2Reliability

If comprehensive cognitive assessments are performed to predict disease progression, then prediction reliability is improved, but ease of operation deteriorates due to increased burden on patients

Engineering Contradiction:
Improveprediction reliabilityVSAvoidpatient burden
Core Design Contradiction:
ReliabilityVSEase of operation

Solution Approach 1:

The patent extracts only the four most reliable predictors from comprehensive cognitive assessments, eliminating unnecessary test components. This extraction maintains prediction reliability by focusing on the most discriminative variables while significantly reducing the time and effort required from patients.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent applies partial action by performing only the essential cognitive assessments needed for reliable prediction rather than comprehensive evaluations. The four selected variables provide sufficient predictive power without requiring exhaustive testing, thereby improving ease of operation while maintaining reliability.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP4682905A1Information processing device, information processing method, program, and storage medium
Publication Date: 2026.01.21 EISAI R&D MANAGEMENT CO LTD
  • EP4682905A1 patent drawingFigure 1
  • EP4682905A1 patent drawingFigure 2
  • EP4682905A1 patent drawingFigure 3

AI summary

An information processing device inputs information of a patient to a prediction model trained using training data, in which at least four variables that fall under at least any of first information relating to orientation, second information relating to memory, third information relating to memory of a holiday, a family gathering, a reservation, or schedule of taking medication in IADL, or memory of a meal or actual work of travel in IADL, and fourth information relating to positive/negative of a biomarker are associated with presence/absence of advance of a symptom of mild cognitive impairment (MCI) or mild Alzheimer's dementia (mild AD), with the information of the patient corresponding to each of the at least four variables, and predicts a probability of advance of a symptom of mild cognitive impairment (MCI) or mild Alzheimer's dementia (mild AD) in the patient. The at least four variables include a first variable that is information relating to orientation relating to time out of the first information, a second variable that is information relating to delayed recall out of the second information, a third variable that is information relating to memory regarding a holiday, a family gathering, a reservation, or schedule for taking medication in IADL, out of the third information, and a fourth variable that is information relating to amyloid β positive/negative out of the fourth information.