Palabras clave
AI embedded system
inference time
quantization
Cómo citar
Resumen
Objetivo: el creciente uso de la inteligencia artificial (IA) en diversos campos ha aumentado la necesidad de una gran cantidad de datos. Se necesita un dispositivo con la potencia de cálculo adecuada para gestionar los datos y producir un resultado con una alta velocidad de procesamiento y con una precisión satisfactoria. Además, el uso de varios dispositivos de sistemas integrados para redes neuronales (NN) se ve limitado por la escasa capacidad del procesador y de la memoria. Se han desarrollado varios dispositivos de sistemas embebidos con capacidades de procesador mejoradas para el procesamiento de datos de redes neuronales. Materiales y método: en este estudio se analizaron las capacidades de un dispositivo de sistema integrado para redes neuronales en aplicaciones sanitarias; en concreto, se probó la detección de imágenes de rayos X de pacientes con neumonía, mediante una red neuronal convolucional (CNN). Se emplearon arquitecturas CNN bidimensionales con diversos parámetros, como profundidad de color, capas, filtros, núcleos y cuantización. Los resultados se expresaron en términos de precisión, tiempo de inferencia y consumo de RAM y flash. Resultados y discusión: los resultados revelaron una asociación positiva significativa entre todas las métricas de salida y el número de filtros. Sin embargo, en algunas situaciones, la utilización de RAM y flash del sistema embebido superó su capacidad, haciéndolo inutilizable. Este hallazgo indica que el tiempo de inferencia y la memoria se ven influidos por el número de capas, filtros y cuantización. Conclusiones: así pues, el uso de dispositivos con sistemas embebidos para CNN puede realizarse con un ajuste adecuado de los hiperparámetros.
W. S. Lim, “Pneumonia—Overview,” Reference Module in Biomedical Sciences, pp. B978-0-12-801238-3.11636-8, 2020, doi: 10.1016/B978-0-12-801238-3.11636-8.
V. Jain, R. Vashisht, G. Yilmaz, and A. Bhardwaj, “Pneumonia Pathology,” in StatPearls, Treasure Island (FL): StatPearls Publishing, 2022. Accessed: Dec. 20, 2022. [Online]. Available: http://www.ncbi.nlm.nih.gov/books/NBK526116/
E. E. Adaji, W. Ekezie, M. Clifford, and R. Phalkey, “Understanding the effect of indoor air pollution on pneumonia in children under 5 in low- and middle-income countries: a systematic review of evidence,” Environ Sci Pollut Res, vol. 26, no. 4, pp. 3208–3225, Feb. 2019, doi: 10.1007/s11356-018-3769-1.
L. Gattinoni et al., “COVID-19 pneumonia: pathophysiology and management,” European Respiratory Review, vol. 30, no. 162, Dec. 2021, doi: 10.1183/16000617.0138-2021.
A. Arora, “One is Too Many: Ending child deaths from pneumonia and diarrhoea,” UNICEF DATA. Accessed: Jan. 16, 2023. [Online]. Available: https://data.unicef.org/resources/one-many-ending-child-deaths-pneumonia-diarrhoea/
P. Meenakshi., K. Bhavana, and A. K. Nair, “Pneumonia Detection using X-ray Image Analysis with Image Processing Techniques,” in 2022 7th International Conference on Communication and Electronics Systems (ICCES), Jun. 2022, pp. 1657–1662. doi: 10.1109/ICCES54183.2022.9835798.
C. Ortiz-Toro, A. García-Pedrero, M. Lillo-Saavedra, and C. Gonzalo-Martín, “Automatic detection of pneumonia in chest X-ray images using textural features,” Comput Biol Med, vol. 145, p. 105466, Jun. 2022, doi: 10.1016/j.compbiomed.2022.105466.
I. Sirazitdinov, M. Kholiavchenko, T. Mustafaev, Y. Yixuan, R. Kuleev, and B. Ibragimov, “Deep neural network ensemble for pneumonia localization from a large-scale chest x-ray database,” Computers & Electrical Engineering, vol. 78, pp. 388–399, Sep. 2019, doi: 10.1016/j.compeleceng.2019.08.004.
Z. Meng, L. Meng, and H. Tomiyama, “Pneumonia Diagnosis on Chest X-Rays with Machine Learning,” Procedia Computer Science, vol. 187, pp. 42–51, Jan. 2021, doi: 10.1016/j.procs.2021.04.032.
H. Agrawal, “Pneumonia Detection Using Image Processing And Deep Learning,” in 2021 International Conference on Artificial Intelligence and Smart Systems (ICAIS), Mar. 2021, pp. 67–73. doi: 10.1109/ICAIS50930.2021.9395895.
A. H. Alharbi and H. A. Hosni Mahmoud, “Pneumonia Transfer Learning Deep Learning Model from Segmented X-rays,” Healthcare (Basel), vol. 10, no. 6, p. 987, May 2022, doi: 10.3390/healthcare10060987.
R. Kundu, R. Das, Z. W. Geem, G.-T. Han, and R. Sarkar, “Pneumonia detection in chest X-ray images using an ensemble of deep learning models,” PLOS ONE, vol. 16, no. 9, p. e0256630, Sep. 2021, doi: 10.1371/journal.pone.0256630.
T. Rajasenbagam, S. Jeyanthi, and J. A. Pandian, “Detection of pneumonia infection in lungs from chest X-ray images using deep convolutional neural network and content-based image retrieval techniques,” J Ambient Intell Human Comput, Mar. 2021, doi: 10.1007/s12652-021-03075-2.
A. Mabrouk, R. P. Díaz Redondo, A. Dahou, M. Abd Elaziz, and M. Kayed, “Pneumonia Detection on Chest X-ray Images Using Ensemble of Deep Convolutional Neural Networks,” Applied Sciences, vol. 12, no. 13, Art. no. 13, Jan. 2022, doi: 10.3390/app12136448.
A. K. Jaiswal, P. Tiwari, S. Kumar, D. Gupta, A. Khanna, and J. J. P. C. Rodrigues, “Identifying pneumonia in chest X-rays: A deep learning approach,” Measurement, vol. 145, pp. 511–518, Oct. 2019, doi: 10.1016/j.measurement.2019.05.076.
W. Jeon, G. Ko, J. Lee, H. Lee, D. Ha, and W. W. Ro, “Chapter Six - Deep learning with GPUs,” in Advances in Computers, vol. 122, S. Kim and G. C. Deka, Eds., in Hardware Accelerator Systems for Artificial Intelligence and Machine Learning, vol. 122. , Elsevier, 2021, pp. 167–215. doi: 10.1016/bs.adcom.2020.11.003.
M. Pandey et al., “The transformational role of GPU computing and deep learning in drug discovery,” Nat Mach Intell, vol. 4, no. 3, Art. no. 3, Mar. 2022, doi: 10.1038/s42256-022-00463-x.
F. Sakr, F. Bellotti, R. Berta, and A. De Gloria, “Machine Learning on Mainstream Microcontrollers,” Sensors, vol. 20, no. 9, Art. no. 9, Jan. 2020, doi: 10.3390/s20092638.
J. Pardo, F. Zamora-Martínez, and P. Botella-Rocamora, “Online Learning Algorithm for Time Series Forecasting Suitable for Low Cost Wireless Sensor Networks Nodes,” Sensors, vol. 15, no. 4, Art. no. 4, Apr. 2015, doi: 10.3390/s150409277.
J. Gorospe, R. Mulero, O. Arbelaitz, J. Muguerza, and M. Á. Antón, “A Generalization Performance Study Using Deep Learning Networks in Embedded Systems,” Sensors, vol. 21, no. 4, Art. no. 4, Jan. 2021, doi: 10.3390/s21041031.
S. S. Saha, S. S. Sandha, and M. Srivastava, “Machine Learning for Microcontroller-Class Hardware: A Review,” IEEE Sensors Journal, vol. 22, no. 22, pp. 21362–21390, Nov. 2022, doi: 10.1109/JSEN.2022.3210773.
L. Saad Saoud and A. Khellaf, “A neural network based on an inexpensive eight-bit microcontroller,” Neural Comput & Applic, vol. 20, no. 3, pp. 329–334, Apr. 2011, doi: 10.1007/s00521-010-0377-5.
J. Lin, W.-M. Chen, H. Cai, C. Gan, and S. Han, “Memory-efficient Patch-based Inference for Tiny Deep Learning,” presented at the Annual Conference on Neural Information Processing Systems, Dec. 2021. Accessed: Dec. 20, 2022. [Online]. Available: https://research.ibm.com/publications/memory-efficient-patch-based-inference-for-tiny-deep-learning
B. Almaslukh, “A Lightweight Deep Learning-Based Pneumonia Detection Approach for Energy-Efficient Medical Systems,” Wireless Communications and Mobile Computing, vol. 2021, p. e5556635, Apr. 2021, doi: 10.1155/2021/5556635.
T. Barakbayeva and F. M. Demirci, “Fully automatic CNN design with inception and ResNet blocks,” Neural Comput & Applic, Sep. 2022, doi: 10.1007/s00521-022-07700-9.
P. Kalgaonkar and M. El-Sharkawy, “CondenseNeXt: An Ultra-Efficient Deep Neural Network for Embedded Systems,” in 2021 IEEE 11th Annual Computing and Communication Workshop and Conference (CCWC), Jan. 2021, pp. 0524–0528. doi: 10.1109/CCWC51732.2021.9375950.
D. Menaka and S. G. Vaidyanathan, “Chromenet: a CNN architecture with comparison of optimizers for classification of human chromosome images,” Multidim Syst Sign Process, vol. 33, no. 3, pp. 747–768, Sep. 2022, doi: 10.1007/s11045-022-00819-x.
Y. Sun, B. Xue, M. Zhang, and G. G. Yen, “Completely Automated CNN Architecture Design Based on Blocks,” IEEE Transactions on Neural Networks and Learning Systems, vol. 31, no. 4, pp. 1242–1254, Apr. 2020, doi: 10.1109/TNNLS.2019.2919608.
M. Garifulla et al., “A Case Study of Quantizing Convolutional Neural Networks for Fast Disease Diagnosis on Portable Medical Devices,” Sensors, vol. 22, no. 1, Art. no. 1, Jan. 2022, doi: 10.3390/s22010219.
H. Liu, Q. Luo, M. Shao, J.-S. Pan, and J. Li, “Joint Channel Pruning and Quantization-Based CNN Network Learning with Mobile Computing-Based Image Recognition,” Wireless Communications and Mobile Computing, vol. 2021, p. e9310558, Oct. 2021, doi: 10.1155/2021/9310558.
J. Pérez, J. Díaz, J. Berrocal, R. López-Viana, and Á. González-Prieto, “Edge computing,” Computing, vol. 104, no. 12, pp. 2711–2747, Dec. 2022, doi: 10.1007/s00607-022-01104-2.
D. S. Kermany et al., “Identifying Medical Diagnoses and Treatable Diseases by Image-Based Deep Learning,” Cell, vol. 172, no. 5, pp. 1122-1131.e9, Feb. 2018, doi: 10.1016/j.cell.2018.02.010.
L. J. Muhammad, A. A. Haruna, U. S. Sharif, and M. B. Mohammed, “CNN-LSTM deep learning based forecasting model for COVID-19 infection cases in Nigeria, South Africa and Botswana,” Health Technol., vol. 12, no. 6, pp. 1259–1276, Nov. 2022, doi: 10.1007/s12553-022-00711-5.
C. Alippi, S. Disabato, and M. Roveri, “Moving Convolutional Neural Networks to Embedded Systems: The AlexNet and VGG-16 Case,” in 2018 17th ACM/IEEE International Conference on Information Processing in Sensor Networks (IPSN), Apr. 2018, pp. 212–223. doi: 10.1109/IPSN.2018.00049.
Esta obra está bajo una licencia internacional Creative Commons Atribución 4.0.
Derechos de autor 2025 Yasser Abd Djawad, Dary Mochamad Rifqie, Ridwansyah, Supriadi, Saharuddin Ronge Sokku, Andi Baso Kaswar, Ma’ruf Idris