Deep Learning Peak Detection for Co-Eluted Gas Chromatography Signals
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional gas chromatography (GC) peak detection methods, particularly for co-eluted peaks, are inconsistent and unreliable due to sensitivity to noise and the inability to handle complex GC signals with multiple volatile organic compounds (VOCs) present in human breath.
Innovation Solution
A deep learning approach using a trained peak identification model, combined with a post-processing algorithm, is employed to analyze chromatographic data and identify peaks, including co-eluted peaks, by generating peak identification probabilities and applying smoothing techniques to enhance accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional peak detection methods are used, then the system is simple, but the reliability and measurement precision deteriorate due to noise sensitivity and inability to handle co-eluted peaks
Solution Approach 1:
The patent replaces conventional mechanical/mathematical peak detection algorithms with a deep learning-based system. The deep learning model processes chromatographic data through neural networks to identify peaks, including co-eluted peaks, thereby improving reliability while accepting increased system complexity through software-based intelligence rather than algorithmic complexity
Solution Approach 2:
The system changes the detection parameters by using deep learning models that can adapt to complex signal patterns. The model learns from training data and adjusts its internal parameters to accurately identify peaks even in noisy conditions and when peaks are co-eluted, transforming fixed threshold-based detection into adaptive, learning-based detection
2Measurement precision
If conventional peak detection methods are used, then the processing is fast, but the measurement precision deteriorates due to noise sensitivity
Solution Approach 1:
The deep learning model is pre-trained on extensive training data representing various chromatographic patterns, including noisy signals and co-eluted peaks. This preliminary training allows the model to quickly and accurately identify peaks in new, unseen data without requiring time-consuming real-time analysis, thus improving precision while maintaining acceptable processing speeds
3Adaptability or versatility
If conventional peak detection methods are used, then the system is simple, but the adaptability deteriorates due to inability to handle complex GC signals with multiple VOCs
Solution Approach 1:
The deep learning model is designed to be universal, handling diverse chromatographic signals including simple peaks, noisy signals, and complex co-eluted peaks with multiple VOCs. The model's architecture and training data are designed to cover a broad range of signal types, making it adaptable to various GC analysis scenarios without requiring separate specialized algorithms for each case
Data Source
AI summary
Techniques for identifying gas chromatography peaks are disclosed herein. An example method includes receiving chromatographic data of a user that includes data representing at least one volatile organic compound (VOC). The example method further includes analyzing the chromatographic data using a trained peak identification model to output a set of peak identification probabilities. The trained peak identification model is trained using a plurality of training chromatographic data to output a plurality of training sets of peak identification probabilities. The example method further includes generating a set of identified peaks within the chromatographic data by applying a post-processing algorithm to the set of peak identification probabilities and causing the set of identified peaks to be displayed to the user.