Dual-Sensitivity Peak Detection for Shoulder Peaks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current peak detection methods in chromatography and mass spectrometry struggle to accurately identify shoulder peaks due to overlapping peaks and varying signal/noise ratios, leading to either missed detections or false noise identification.
Innovation Solution
A dual-step peak detection method using different sensitivities to differentiate between true shoulder peaks and noise, where the first step detects peaks with a higher sensitivity and the second step adjusts sensitivity to confirm or exclude candidates, allowing users to set detection thresholds.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the threshold value of the S/N ratio is decreased to detect shoulder peaks, then the detection sensitivity is improved, but noise is erroneously detected as peaks
Solution Approach 1:
The peak detection process is divided into two distinct stages: first detection with a first S/N threshold, and second detection with a second S/N threshold. This segmentation allows the system to identify potential shoulder peaks in the first stage while filtering out noise in the second stage, resolving the contradiction between detection sensitivity and false detection rate
Solution Approach 2:
The first peak detection with a lower S/N threshold performs a preliminary identification of potential peaks including shoulder peaks. This preliminary action creates a candidate list that is then refined in the second detection stage, allowing the system to capture weak signals without committing to false positives yet
2Reliability
If the threshold value of the S/N ratio is increased to exclude noise, then the false detection rate is reduced, but shoulder peaks are not detected
Solution Approach 1:
The detection process is segmented into two passes with different threshold criteria. The first pass uses a permissive threshold to ensure no shoulder peaks are missed, while the second pass uses a strict threshold to verify true peaks. This segmentation resolves the contradiction by applying different reliability standards at different stages of the detection process
Solution Approach 2:
The first detection stage intentionally uses an excessive action by lowering the S/N threshold below what would normally be required for reliable detection. This ensures that even weak shoulder peaks are captured as candidates, with the understanding that further verification will be performed in the second stage
3Ease of operation
If a single threshold value is used for peak detection, then the detection process is simple, but both shoulder peaks and noise cannot be properly distinguished
Solution Approach 1:
The single threshold approach is segmented into two thresholds applied in sequence. The first threshold identifies candidates, and the second threshold confirms true peaks. This segmentation maintains operational simplicity by using a straightforward two-step process while dramatically improving peak identification accuracy through the differential threshold application
Solution Approach 2:
The detection system dynamically adjusts the S/N threshold based on the detection stage. Rather than using a fixed threshold, the system transitions from a lower threshold in the first stage to a higher threshold in the second stage, allowing the detection criteria to adapt to the needs of each detection phase while maintaining an overall simple operational framework
Data Source
AI summary
A peak detecting method according to the present invention includes a first peak detecting step (S1) of detecting, with a first detection sensitivity, one or more peaks in object data whose peak is to be detected; and a second peak detecting step (S5) of detecting one or more peaks in the object data with a second detection sensitivity that is different from the first detection sensitivity. With this method, since two kinds of detection are performed at the different detection sensitivities, a shoulder peak (or a non-shoulder peak) candidate can be chosen from peaks that are detected with one of the detection sensitivities but are not detected with the other detection sensitivity, and thus, a shoulder peak can be appropriately detected with ease.