Deep Learning Urine Flow Diagnosis Model

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional urodynamic studies for diagnosing lower urinary tract symptoms are invasive, causing discomfort, pain, and risk of infection, and are not suitable for widespread use due to their invasive nature.

Innovation Solution

A deep learning-based method using non-invasive simple urine flow test results to train models for diagnosing lower urinary tract symptoms, extracting character and graph data to identify feature points correlated with symptom causes, and integrating these for accurate diagnosis without invasive procedures.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional urodynamic study is performed to accurately diagnose lower urinary tract symptoms, then diagnostic accuracy is improved, but patient discomfort and risk of infection increase due to invasive catheter insertion

Engineering Contradiction:

Improvediagnostic accuracyVSAvoidpatient discomfort and infection risk

Core Design Contradiction:

Measurement precisionVSObject-affected harmful factors

Solution Approach 1:

The patent creates a virtual model (copy) of the urodynamic study results using deep learning algorithms trained on data from actual urodynamic studies. This virtual model can predict diagnostic outcomes without requiring actual invasive procedures, thereby maintaining diagnostic accuracy while eliminating patient discomfort and infection risks associated with real catheter insertion

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The patent replaces the mechanical invasive system (catheter insertion and physical pressure measurement) with an information processing system (deep learning algorithms analyzing non-invasive urine flow test data). This substitution maintains the diagnostic function while eliminating the harmful mechanical intrusion into the patient's body

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Reliability

If invasive urodynamic study is used to determine surgical candidates, then diagnostic reliability is improved, but patient mental stress and embarrassment increase

Engineering Contradiction:

Improvediagnostic reliabilityVSAvoidmental stress and embarrassment

Core Design Contradiction:

ReliabilityVSObject-affected harmful factors

Solution Approach 1:

The patent creates a virtual representation of the invasive diagnostic process using deep learning models trained on urodynamic study data. This virtual model reproduces diagnostic reliability by analyzing patterns from training data, allowing accurate surgical candidate identification without subjecting patients to the embarrassing and stressful invasive procedure

Inventive Principle:

Principle #26Copying

3Ease of operation

If simple urine flow test is used instead of urodynamic study, then patient comfort is improved, but diagnostic capability deteriorates due to lack of pressure measurement

Engineering Contradiction:

Improvepatient comfortVSAvoiddiagnostic capability

Core Design Contradiction:

Ease of operationVSMeasurement precision

Solution Approach 1:

The patent introduces deep learning algorithms as an intermediary that bridges the gap between simple urine flow test data and accurate diagnostic capability. The algorithm processes the limited non-invasive data and extracts diagnostic information that would traditionally require invasive pressure measurements, thereby restoring diagnostic capability while maintaining patient comfort

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent transforms the approach by changing from direct physical measurement parameters (pressure, flow rate) to derived informational parameters (patterns, correlations, relationships) that can be extracted from simple urine flow test data through deep learning analysis. This parameter transformation enables accurate diagnosis without invasive measurements

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS20230309931A1Deep learning-based simple urine flow test result learning method and lower urinary tract symptom diagnosis method

Publication Date: 2023.10.05 SAMSUNG LIFE PUBLIC WELFARE FOUND

AI summary

A simple urine flow test result learning method and a lower urinary tract symptom diagnosis method is provided, and more particularly, a simple urine flow test result learning method and a lower urinary tract symptom diagnosis method of training a neural network using simple urine flow test results, which are non-invasive data, and diagnosing lower urinary tract symptoms using the trained neural network, wherein the lower urinary tract symptom diagnosis method prevents pain and shame from occurring in a patient during a diagnosis process of lower urinary tract symptoms and reduces the risk of secondary infection occurring through an invasive diagnosis method, by generating a trained model using results of a simple urine flow test, which is a non-invasive test method, based on deep learning, and diagnosing lower urinary tract symptoms using the trained model.