Machine Learning Applied for Spectra Classification

Sun, Yue; Brockhauser, Sándor; Hegedűs, Péter

doi:10.1007/978-3-030-87013-3_5

Cited by 3 publications

(6 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The collected AC-mode sensor data was analyzed with machine learning (ML) classification in an attempt to classify NPY concentrations based on observed currents. Machine learning has revolutionized the field of data analysis and has allowed for the interpretation and classification of complex data without human intervention and has been applied in various fields, including the analysis of sensor data and spectra − and human performance. , The results of classification with only three categories are shown in the SI with a maximum average accuracy of 74.38 ± 7.14% seen for an Ada boost classifier. This result could be useful for NPY screening and eventually screening for human performance.…”

Section: Resultsmentioning

confidence: 99%

Label-Free Neuropeptide Detection beyond the Debye Length Limit

Sarker,

Shrestha,

Singh

et al. 2023

ACS Nano

View full text Add to dashboard Cite

Biosensors with high selectivity, high sensitivity, and real-time detection capabilities are of significant interest for diagnostic applications as well as human health and performance monitoring. Graphene field-effect transistor (GFET) based biosensors are suitable for integration into wearable sensor technology and can potentially demonstrate the sensitivity and selectivity necessary for real-time detection and monitoring of biomarkers. Previously reported DC-mode GFET biosensors showed a high sensitivity for sensing biomarkers in solutions with a low salt concentration. However, due to Debye length screening, the sensitivity of the DC-mode GFET biosensors decreases significantly during operation in a physiological fluid such as sweat or interstitial fluid. To overcome the Debye screening length limitation, we report here alternating current (AC) mode heterodyne-based GFET biosensors for sensing neuropeptide-Y (NPY), a key stress biomarker, in artificial sweat at physiologically relevant ionic concentrations. Our ACmode GFET biosensors show a record ultralow detection limit of 2 × 10 −18 M with an extensive dynamic range of 10 orders of magnitude in sensor response to target NPY concentration. The sensors were characterized for various carrier frequencies (ranging from 30 kHz to 2 MHz) of the applied AC voltages and various salt concentrations (10, 50, and 100 mM). Contrary to DC-mode sensing, the AC-mode sensor response increases with an increase in salt concentration in the electrolyte. The sensor response can be further enhanced by tuning the carrier frequency of the applied AC voltage. The optimum response frequency of our sensor is approximately 400−600 kHz for salt concentrations of 50 and 100 mM, respectively. The saltconcentration-and frequency-dependent sensor response can be explained by an electrolyte-gated capacitance model.

show abstract

Section: Resultsmentioning

confidence: 99%

Label-Free Neuropeptide Detection beyond the Debye Length Limit

Sarker,

Shrestha,

Singh

et al. 2023

ACS Nano

View full text Add to dashboard Cite

show abstract

“…Specifically, the 16 spectral curves marked in red belong to class label 0 (before phase transition) and the 12 ones marked in blue belong to class label 1 (after phase transition). It can be clearly seen from this figure, the amplitude of spectral peaks changes (increases, decreases, vanishes) at certain locations, and peaks also shift in 2θ-angle dimension, or split, or start to broaden [106]. These changes correspond to modifications of the crystal lattice (e.g.…”

Section: Dataset Descriptionmentioning

confidence: 89%

“…In this chapter, we present an unsupervised Spectral Clustering-based method for the application of diffraction spectra data classification [106]. The description of the spectral dataset is given in Section 3.1.…”

Section: Spectral Clustering Methodsmentioning

confidence: 99%

“…In particular, here we will also analyze the correlations between the original variables and the principal components (PCs) from PCA, and their contribution to the PCs. Then, in Section 3.4 a combination of PCA and Spectral Clustering approach is presented to cluster the spectra and detect phase transition [106]. Finally, the implementation details and the result of the model are presented in Section 3.5.…”

Section: Spectral Clustering Methodsmentioning

confidence: 99%

“…Experiment studies and discussion of results are shown in Section 5.3. In addition, an open-source interactive software program [109] on these classification models is shown in Section 5.4. Finally, the conclusion and main contributions are given in Section 5.5.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Application of Machine Learning to the Classification of Spectra-like Data

Yue

View full text Add to dashboard Cite

show abstract

Comparing End-to-End Machine Learning Methods for Spectra Classification

2021

View full text Add to dashboard Cite

In scientific research, spectroscopy and diffraction experimental techniques are widely used and produce huge amounts of spectral data. Learning patterns from spectra is critical during these experiments. This provides immediate feedback on the actual status of the experiment (e.g., time-resolved status of the sample), which helps guide the experiment. The two major spectral changes what we aim to capture are either the change in intensity distribution (e.g., drop or appearance) of peaks at certain locations, or the shift of those on the spectrum. This study aims to develop deep learning (DL) classification frameworks for one-dimensional (1D) spectral time series. In this work, we deal with the spectra classification problem from two different perspectives, one is a general two-dimensional (2D) space segmentation problem, and the other is a common 1D time series classification problem. We focused on the two proposed classification models under these two settings, the namely the end-to-end binned Fully Connected Neural Network (FCNN) with the automatically capturing weighting factors model and the convolutional SCT attention model. Under the setting of 1D time series classification, several other end-to-end structures based on FCNN, Convolutional Neural Network (CNN), ResNets, Long Short-Term Memory (LSTM), and Transformer were explored. Finally, we evaluated and compared the performance of these classification models based on the High Energy Density (HED) spectra dataset from multiple perspectives, and further performed the feature importance analysis to explore their interpretability. The results show that all the applied models can achieve 100% classification confidence, but the models applied under the 1D time series classification setting are superior. Among them, Transformer-based methods consume the least training time (0.449 s). Our proposed convolutional Spatial-Channel-Temporal (SCT) attention model uses 1.269 s, but its self-attention mechanism performed across spatial, channel, and temporal dimensions can suppress indistinguishable features better than others, and selectively focus on obvious features with high separability.

show abstract

Machine Learning Applied for Spectra Classification

Cited by 3 publications

References 33 publications

Label-Free Neuropeptide Detection beyond the Debye Length Limit

Label-Free Neuropeptide Detection beyond the Debye Length Limit

Application of Machine Learning to the Classification of Spectra-like Data

Comparing End-to-End Machine Learning Methods for Spectra Classification

Contact Info

Product

Resources

About