Neural Network Dynamic Input Layer Selection for Medical Diagnosis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current neural network systems for medical diagnosis face challenges in reducing false positive diagnostic results, especially when dealing with large datasets and complex medical images, which affects their accuracy and computational efficiency.
Innovation Solution
The method involves dynamically selecting input layers based on a data qualification threshold, deactivating less relevant layers, and recursively adjusting the initial weight matrix through backpropagation to minimize false positive patient diagnostic states, using a supervised classification approach within a machine learning neural network framework.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all input layers are activated in the neural network, then comprehensive medical data is processed, but computational complexity and processing time increase
Solution Approach 1:
The patent implements dynamic input layer selection where the neural network adaptively activates only the most relevant input layers based on the specific medical case being analyzed. This dynamic approach allows the system to maintain high diagnostic accuracy by selecting appropriate data sources while reducing computational complexity by deactivating unnecessary layers, directly resolving the contradiction between comprehensive data processing and computational efficiency
2Reliability
If comprehensive medical data from multiple sources is processed, then diagnostic accuracy improves, but false positive results increase
Solution Approach 1:
The patent employs feedback mechanisms through iterative training processes where the neural network learns from both correct and incorrect diagnoses. The system continuously adjusts its processing by incorporating feedback from diagnostic outcomes, refining its ability to distinguish true positives from false positives while maintaining comprehensive data analysis, thus improving reliability without proportionally increasing false positive rates
3Loss of information
If all input layers are processed, then complete medical information is analyzed, but system response time decreases
Solution Approach 1:
The patent segments the medical data processing into distinct input layers that can be independently activated. By dividing the comprehensive medical information into modular segments (different data sources and types), the system can selectively process only the necessary segments for each diagnostic case, maintaining information completeness where needed while reducing overall processing time through parallel and selective execution
Data Source
AI summary
Method and system of deploying a machine learning neural network (MLNN). The method comprises receiving a set of input features associated with data representative of a patient medical state at input layers of a trained MLNN, the trained MLNN comprising an output layer interconnected to the input layers via intermediate layers configured in accordance with an initial matrix of weights, a subset of the input features being activated responsive to a data sufficiency threshold reached in conjunction with deactivating, from the intermediate layers, a remainder of the input layers, the trained MLNN produced in accordance with adjusting the initial matrix of weights in diminishment of false positives in providing, at the output layer, a patient state diagnosis, and generating, at the output layer, a medical state diagnosis in accordance with the diminishment of false positives.