ML Ontology Mapping Brain Structures to Mental Functions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional neuroimaging techniques rely on expert-determined frameworks, leading to subjective biases and limited novelty and replicability in linking brain structures to mental functions.
Innovation Solution
A machine learning-based approach using natural language processing and machine learning models to generate an ontology that maps brain structures to mental functions by analyzing co-occurrences of textual and spatial data, optimizing the number of domains and mental function terms using unsupervised and supervised learning techniques, and applying this ontology to process electronic medical records.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If expert-determined frameworks are used to link brain structures to mental functions, then the process relies on established knowledge, but subjective biases are introduced and replicability is limited
Solution Approach 1:
The patent replaces the manual, expert-driven framework construction process with an automated machine learning system. The ML model objectively processes neuroimaging data to generate brain-structure-to-function mappings, eliminating subjective biases while maintaining scientific rigor through algorithmic consistency and reproducibility
Solution Approach 2:
The system enables self-service by allowing the data and algorithms to automatically generate the ontology without requiring continuous expert intervention. The machine learning model independently processes the corpus of data and produces reproducible results, reducing reliance on subjective expert judgment while maintaining high reliability
2Adaptability or versatility
If conventional neuroimaging techniques are used, then the methodology is established and reliable, but novelty in linking brain structures to mental functions is limited
Solution Approach 1:
The patent changes the fundamental parameters of the analysis by using machine learning algorithms to process neuroimaging data differently from conventional methods. This enables the discovery of novel brain-function relationships that were not apparent through traditional expert-driven approaches, while the systematic nature of ML ensures methodological stability and reproducibility
3Measurement precision
If a detailed ontology with many domains and mental function terms is created, then the mapping precision is improved, but the complexity of the ontology increases
Solution Approach 1:
The patent creates a dynamic ontology structure where domains and mental function terms can be adaptively selected based on the specific research questions and data characteristics. The machine learning model can optimize the ontology's granularity and complexity for different applications, achieving high mapping precision without requiring a fixed, overly complex structure for all use cases
Data Source
AI summary
A method may include applying, to a corpus of data, a first machine learning technique to identify candidate domains of an ontology mapping brain structure to mental function. The corpus of data may include textual data describing a plurality of mental functions and spatial data corresponding to a plurality of brain structures. A second machine technique may be applied to optimize a quantity of domains included in the ontology and/or a quantity of mental function terms included in each domain. The ontology may be applied to phenotype an electronic medical record and predict a clinical outcome for a patient associated with the electronic medical record. Related systems and articles of manufacture, including computer program products, are also provided.