The Classification of VOCs Based on Sensor Images Using a Lightweight Neural Network for Lung Cancer Diagnosis
Abstract
:1. Introduction
- (1)
- A hardware-friendly lightweight neural network model (LTNet) using a depth-separable convolution structure for gas classification is constructed.
- (2)
- To settle the decrease in classification accuracy caused by depthwise separable convolutions, we propose to add squeeze-and-excitation (SE) attention mechanisms and residual connections in the model.
- (3)
- The convolutional and batch normalization (BN) layers are combined together so as to reduce the model parameters, speed up the inference speed and improve the stability of the model.
- (4)
- Compared to the unimproved LTNet (LTNet (Original version)), this validates the effectiveness of the improvements made to LTNet.
2. Experimental Section
2.1. Data Source I: Gas Mixture Dataset
2.2. Data Sources II: UCI Database
2.3. Experimental Environment and Hardware Configuration
3. Data Processing
3.1. Image Conversion Methods
3.2. Lightweight Neural Network Model
3.3. Calculation of Depthwise Separable Convolutions Parameters
4. Results and Discussion
4.1. Data Conversion Comparison Test and Results Discussion
4.2. Model Evaluation and Comparison Experiment
4.3. Classification Results of Own Mixed Gas Dataset
4.4. UCI Database Classification Results
5. Conclusions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Conflicts of Interest
References
- Huang, Z.P.; Cheng, H.L.; Loh, S.Y.; Cheng, K.K.F. Functional Status, Supportive Care Needs, and Health-Related Quality of Life in Advanced Lung Cancer Patients Aged 50 and Older. Asia-Pac. J. Oncol. Nurs. 2020, 7, 151–160. [Google Scholar] [CrossRef]
- Vranas, K.C.; Lapidus, J.A.; Ganzini, L.; Slatore, C.G.; Sullivan, D.R. Association of Palliative Care Use and Setting With Health-care Utilization and Quality of Care at the End of Life Among Patients With Advanced Lung Cancer. Chest 2020, 158, 2667–2674. [Google Scholar] [CrossRef]
- Siegel, R.L.; Miller, K.D.; Wagle, N.S.; Jemal, A. Cancer statistics, 2023. CA-A Cancer J. Clin. 2023, 73, 17–48. [Google Scholar] [CrossRef] [PubMed]
- Marzorati, D.; Mainardi, L.; Sedda, G.; Gasparri, R.; Spaggiari, L.; Cerveri, P. A review of exhaled breath: A key role in lung cancer diagnosis. J. Breath Res. 2019, 13, 20. [Google Scholar] [CrossRef]
- Sun, X.H.; Shao, K.; Wang, T. Detection of volatile organic compounds (VOCs) from exhaled breath as noninvasive methods for cancer diagnosis. Anal. Bioanal. Chem. 2016, 408, 2759–2780. [Google Scholar] [CrossRef]
- Zhu, H.Y.; Liu, C.; Zheng, Y.; Zhao, J.; Li, L. A Hybrid Machine Learning Algorithm for Detection of Simulated Expiratory Markers of Diabetic Patients Based on Gas Sensor Array. IEEE Sens. J. 2023, 23, 2940–2947. [Google Scholar] [CrossRef]
- Buszewski, B.; Ulanowska, A.; Kowalkowski, T.; Cieslinski, K. Investigation of lung cancer biomarkers by hyphenated separation techniques and chemometrics. Clin. Chem. Lab. Med. 2012, 50, 573–581. [Google Scholar] [CrossRef] [PubMed]
- Ulanowska, A.; Kowalkowski, T.; Trawinska, E.; Buszewski, B. The application of statistical methods using VOCs to identify patients with lung cancer. J. Breath Res. 2011, 5, 046008. [Google Scholar] [CrossRef] [PubMed]
- Apolonski, A.; Maiti, K.S. Towards a standard operating procedure for revealing hidden volatile organic compounds in breath: The Fourier-transform IR spectroscopy case. Appl. Opt. 2021, 60, 4217–4224. [Google Scholar] [CrossRef]
- Schulz, E.; Woollam, M.; Grocki, P.; Davis, M.D.; Agarwal, M. Methods to Detect Volatile Organic Compounds for Breath Biopsy Using Solid-Phase Microextraction and Gas Chromatography-Mass Spectrometry. Molecules 2023, 28, 4533. [Google Scholar] [CrossRef]
- Ye, Z.Y.; Wang, J.; Hua, H.; Zhou, X.D.; Li, Q.L. Precise Detection and Quantitative Prediction of Blood Glucose Level With an Electronic Nose System. IEEE Sens. J. 2022, 22, 12452–12459. [Google Scholar] [CrossRef]
- Baldini, C.; Billeci, L.; Sansone, F.; Conte, R.; Domenici, C.; Tonacci, A. Electronic Nose as a Novel Method for Diagnosing Cancer: A Systematic Review. Biosensors 2020, 10, 84. [Google Scholar] [CrossRef] [PubMed]
- Dragonieri, S.; Annema, J.T.; Schot, R.; van der Schee, M.P.C.; Spanevello, A.; Carratú, P.; Resta, O.; Rabe, K.F.; Sterk, P.J. An electronic nose in the discrimination of patients with non-small cell lung cancer and COPD. Lung Cancer 2009, 64, 166–170. [Google Scholar] [CrossRef] [PubMed]
- Dragonieri, S.; Pennazza, G.; Carratu, P.; Resta, O. Electronic Nose Technology in Respiratory Diseases. Lung 2017, 195, 157–165. [Google Scholar] [CrossRef] [PubMed]
- Marzorati, D.; Mainardi, L.; Sedda, G.; Gasparri, R.; Spaggiari, L.; Cerveri, P. MOS Sensors Array for the Discrimination of Lung Cancer and At-Risk Subjects with Exhaled Breath Analysis. Chemosensors 2021, 9, 209. [Google Scholar] [CrossRef]
- Liu, B.; Yu, H.; Zeng, X.; Zhang, D.; Gong, J.; Tian, L.; Qian, J.; Zhao, L.; Zhang, S.; Liu, R. Lung cancer detection via breath by electronic nose enhanced with a sparse group feature selection approach. Sens. Actuators B Chem. 2021, 339, 129896. [Google Scholar] [CrossRef]
- Wang, Z.; Oates, T. Imaging Time-Series to Improve Classification and Imputation. In Proceedings of the 24th International Conference on Artificial Intelligence, Buenos Aires, Argentina, 25–31 July 2015. [Google Scholar]
- Wang, X.; Qian, C.; Zhao, Z.; Li, J.; Jiao, M. A Novel Gas Recognition Algorithm for Gas Sensor Array Combining Savitzky–Golay Smooth and Image Conversion Route. Chemosensors 2023, 11, 96. [Google Scholar] [CrossRef]
- Binson, V.A.; Subramoniam, M.; Sunny, Y.; Mathew, L. Prediction of Pulmonary Diseases With Electronic Nose Using SVM and XGBoost. IEEE Sens. J. 2021, 21, 20886–20895. [Google Scholar] [CrossRef]
- Chen, K.; Liu, L.; Nie, B.; Lu, B.; Fu, L.; He, Z.; Li, W.; Pi, X.; Liu, H. Recognizing lung cancer and stages using a self-developed electronic nose system. Comput. Biol. Med. 2021, 131, 104294. [Google Scholar] [CrossRef]
- Peng, C.; Zheng, Y. Robust gas recognition with mixed interference using a spiking neural network. Meas. Sci. Technol. 2021, 33, 015105. [Google Scholar] [CrossRef]
- Lekha, S.; Suchetha, M. Real-Time Non-Invasive Detection and Classification of Diabetes Using Modified Convolution Neural Network. IEEE J. Biomed. Health Inform. 2018, 22, 1630–1636. [Google Scholar] [CrossRef] [PubMed]
- Wang, T.; Zhang, H.X.; Wu, Y.; Jiang, W.K.; Chen, X.W.; Zeng, M.; Yang, J.H.; Su, Y.J.; Hu, N.T.; Yang, Z. Target discrimination, concentration prediction, and status judgment of electronic nose system based on large-scale measurement and multi-task deep learning. Sens. Actuator B-Chem. 2022, 351, 12. [Google Scholar] [CrossRef]
- Avian, C.; Mahali, M.I.; Putro, N.A.S.; Prakosa, S.W.; Leu, J.-S. Fx-Net and PureNet: Convolutional Neural Network architecture for discrimination of Chronic Obstructive Pulmonary Disease from smokers and healthy subjects through electronic nose signals. Comput. Biol. Med. 2022, 148, 105913. [Google Scholar] [CrossRef] [PubMed]
- Guo, J.; Cheng, Y.; Luo, D.H.; Wong, K.Y.; Hung, K.; Li, X. ODRP: A Deep Learning Framework for Odor Descriptor Rating Prediction Using Electronic Nose. IEEE Sens. J. 2021, 21, 15012–15021. [Google Scholar] [CrossRef]
- Li, Z.; Kang, S.; Feng, N.; Yin, C.; Shi, Y. PSCFormer: A lightweight hybrid network for gas identification in electronic nose system. Pattern Recognit. 2024, 145, 109912. [Google Scholar] [CrossRef]
- Shi, Y.; Wang, B.; Yin, C.; Li, Z.; Yu, Y. Performance improvement: A lightweight gas information classification method combined with an electronic nose system. Sens. Actuators B Chem. 2023, 396, 134551. [Google Scholar] [CrossRef]
- Rodriguez-Lujan, I.; Fonollosa, J.; Vergara, A.; Homer, M.; Huerta, R. On the calibration of sensor arrays for pattern recognition using the minimal number of experiments. Chemom. Intell. Lab. Syst. 2014, 130, 123–134. [Google Scholar] [CrossRef]
- Vergara, A.; Vembu, S.; Ayhan, T.; Ryan, M.A.; Homer, M.L.; Huerta, R. Chemical gas sensor drift compensation using classifier ensembles. Sens. Actuator B-Chem. 2012, 166, 320–329. [Google Scholar] [CrossRef]
- Xiong, L.J.; He, M.; Hu, C.; Hou, Y.X.; Han, S.Y.; Tang, X.Y. Image presentation and effective classification of odor intensity levels using multi-channel electronic nose technology combined with GASF and CNN. Sens. Actuator B-Chem. 2023, 395, 134492. [Google Scholar] [CrossRef]
- Bai, L.; Zhao, Y.M.; Huang, X.M. A CNN Accelerator on FPGA Using Depthwise Separable Convolution. IEEE Trans. Circuits Syst. Ii-Express Briefs 2018, 65, 1415–1419. [Google Scholar] [CrossRef]
- Adams, S.J.; Stone, E.; Baldwin, D.R.; Vliegenthart, R.; Lee, P.; Fintelmann, F.J. Lung cancer screening. Lancet 2023, 401, 390–408. [Google Scholar] [CrossRef] [PubMed]
NO. | Models | Target Gases | Detection Ranges (ppm) | Optimal Operating Currents (mA) |
---|---|---|---|---|
1 | TGS2600 | Ethanol, Hydrogen | 1–30 | 45 |
2 | TGS2602 | Ammonia, Ethanol | Ethanol 1–30 | 50 |
3 | TGS2610 | Organic compounds | 500–10,000 | 55 |
4 | TGS2620 | Ethanol, Organic compounds | Ethanol 50–5000 | 43 |
NO. | Ethanol (ppm) | Acetone (ppm) | Mixed Gas (ppm) |
---|---|---|---|
1 | 0 | 1 | 1 |
2 | 0 | 3 | 3 |
3 | 0 | 5 | 5 |
4 | 0 | 7 | 7 |
5 | 0 | 9 | 9 |
6 | 0 | 11 | 11 |
7 | 0 | 13 | 13 |
8 | 0 | 15 | 15 |
9 | 1 | 0 | 1 |
10 | 3 | 0 | 3 |
11 | 5 | 0 | 5 |
12 | 7 | 0 | 7 |
13 | 9 | 0 | 9 |
14 | 11 | 0 | 11 |
15 | 13 | 0 | 13 |
16 | 15 | 0 | 15 |
17 | 1 | 1 | 2 |
18 | 1 | 5 | 6 |
19 | 1 | 10 | 11 |
20 | 1 | 15 | 16 |
21 | 5 | 1 | 6 |
22 | 5 | 5 | 10 |
23 | 5 | 10 | 15 |
24 | 5 | 15 | 20 |
25 | 10 | 1 | 11 |
26 | 10 | 5 | 15 |
27 | 10 | 10 | 20 |
28 | 10 | 15 | 25 |
29 | 15 | 1 | 16 |
30 | 15 | 5 | 20 |
31 | 15 | 10 | 25 |
32 | 15 | 15 | 30 |
Models | Accuracy | Training Time (S) | GPU RAM (G) |
---|---|---|---|
GADF | 98.79% | 863.49 | 2.3 |
MTF | 92.34% | 929.08 | 2.6 |
STFT | 100% | 1630.92 | 2.6 |
GASF | 99.06% | 844.38 | 2.1 |
Models | Params. | Weight Size (MB) |
---|---|---|
AlexNet | 57,012,034 | 217 |
ResNet50 | 23,514,179 | 89.9 |
VGG16 | 134,268,738 | 512 |
EfficientNet | 4,586,092 | 17.8 |
MobileNetV3_large | 4,208,443 | 16.2 |
LTNet (Original version) | 296,994 | 1.15 |
LTNet (This work) | 32,614 | 0.155 |
Models | Accuracy | GPU RAM (G) | Training Time (S) | Inference Time (S) |
---|---|---|---|---|
AlexNet | 97.71% | 3.1 | 853.27 | 283 |
ResNet50 | 98.39% | 3.8 | 1234.34 | 284 |
VGG16 | 97.98% | 6.9 | 2249.56 | 592 |
EfficientNet | 99.06% | 5.4 | 1373.48 | 170 |
MobileNetV3_large | 98.79% | 3.3 | 877.53 | 91 |
LTNet (Original version) | 98.65% | 2.3 | 1112.03 | 26 |
LTNet (This work) | 99.06% | 2.1 | 844.38 | 23 |
Models | Accuracy | GPU RAM (G) | Training Time (S) | Inference Time (S) |
---|---|---|---|---|
AlexNet | 98.92% | 3.2 | 613.50 | 187 |
ResNet50 | 98.92% | 3.7 | 841.81 | 178 |
VGG16 | 98.71% | 7.1 | 1477.44 | 377 |
EfficientNet | 98.92% | 5.3 | 933.21 | 109 |
MobileNetV3_large | 98.92% | 3.3 | 606.20 | 60 |
LTNet (Original version) | 98.06% | 2.3 | 859.49 | 18 |
LTNet (this work) | 99.14% | 2.1 | 584.67 | 14 |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Zha, C.; Li, L.; Zhu, F.; Zhao, Y. The Classification of VOCs Based on Sensor Images Using a Lightweight Neural Network for Lung Cancer Diagnosis. Sensors 2024, 24, 2818. https://doi.org/10.3390/s24092818
Zha C, Li L, Zhu F, Zhao Y. The Classification of VOCs Based on Sensor Images Using a Lightweight Neural Network for Lung Cancer Diagnosis. Sensors. 2024; 24(9):2818. https://doi.org/10.3390/s24092818
Chicago/Turabian StyleZha, Chengyuan, Lei Li, Fangting Zhu, and Yanzhe Zhao. 2024. "The Classification of VOCs Based on Sensor Images Using a Lightweight Neural Network for Lung Cancer Diagnosis" Sensors 24, no. 9: 2818. https://doi.org/10.3390/s24092818
APA StyleZha, C., Li, L., Zhu, F., & Zhao, Y. (2024). The Classification of VOCs Based on Sensor Images Using a Lightweight Neural Network for Lung Cancer Diagnosis. Sensors, 24(9), 2818. https://doi.org/10.3390/s24092818