Chromatograph Mass Spectrometry Feature Analysis for Peak Importance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods require significant manual effort to identify feature amounts influencing specific physical properties in chromatograph mass spectrometry data, particularly when dealing with complex and unseparated peaks, with limited automation assistance for operators.
Innovation Solution
A computer-implemented method that extracts feature items from chromatograph mass spectrometry data, generates training data, and uses a machine learning model to predict and identify the importance of these features, assisting operators in identifying key factors influencing physical properties.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual identification of peaks is used in chromatograph mass spectrometry data, then the operator can identify feature amounts, but the operator requires extraordinary effort and time
Solution Approach 1:
The system performs automatic peak identification and feature amount extraction through machine learning models, eliminating the need for operator intervention in routine identification tasks. The model independently processes chromatograph mass spectrometry data to identify peaks and extract feature amounts, serving itself rather than relying on manual operator analysis.
Solution Approach 2:
The manual mechanical process of peak identification by operators is replaced with an automated machine learning system. The mechanical action of visually inspecting and identifying peaks is substituted with computational algorithms that automatically detect and analyze peaks in the chromatograph data.
2Ease of operation
If automatic extraction of markers is implemented, then operator effort is reduced, but the ability to handle complex relations between analysis results and physical properties is limited
Solution Approach 1:
The system provides dynamic adaptability by allowing operators to adjust the degree of automation and intervention based on the complexity of the analysis. The machine learning model can handle complex relations when needed, while also providing automated extraction for simpler cases, making the system flexible rather than fixed in its approach.
Solution Approach 2:
The machine learning model acts as an intermediary between fully automated extraction and fully manual analysis. It provides automated feature amount extraction for routine cases while enabling operator intervention and customization for complex scenarios, serving as a bridge that combines the benefits of both approaches.
3Measurement precision
If multiple feature items are extracted with different combinations of retention time and mass-to-charge ratio, then comprehensive analysis is achieved, but data processing complexity increases
Solution Approach 1:
The comprehensive analysis is achieved by segmenting the data processing into distinct feature items, each corresponding to specific combinations of retention time and mass-to-charge ratio ranges. This segmentation allows the system to systematically process different aspects of the chromatograph data separately, making the complex processing manageable through structured division.
Solution Approach 2:
The system handles multiple feature items by adding dimensional organization to the data processing. Feature items are structured according to their retention time and mass-to-charge ratio characteristics, creating a multi-dimensional framework that organizes complex data in a systematic manner that simplifies processing rather than increasing complexity.
Data Source
AI summary
In the present disclosure, values of a plurality of feature items are extracted from chromatograph mass spectrometry data of a sample. Each of the plurality of feature items corresponds to a combination of a range of retention time and a mass-to-charge ratio. A machine learning model for predicting physical property information from a value of at least one of the plurality of feature items is generated. For each of one or more feature items of the at least one of the plurality of feature items, importance in the machine learning model is identified and output.