Deep Electro-Impedance Analytics for Bone Mineral Profiling: A Rough-Fuzzy Neural Attention Model

Aripin Aripin; Mauldy Nawa Ayu  Wulandari; Eunike Laurensya  Agata; Zulhendra Adi  Kusuma; Susilo Susilo; Sari Ayu Wulandari

doi:10.35882/jeeemi.v8i2.1434

Aripin Aripin Departement of Biomedical Engineering, Faculty of Engineering, Universitas Dian Nuswantoro, Semarang, Central Java, Indonesia https://orcid.org/0000-0001-8434-5775
Mauldy Nawa Ayu Wulandari Departement of Biomedical Engineering, Faculty of Engineering, Universitas Dian Nuswantoro, Semarang, Central Java, Indonesia https://orcid.org/0009-0007-9297-437X
Eunike Laurensya Agata Departement of Biomedical Engineering, Faculty of Engineering, Universitas Dian Nuswantoro, Semarang, Central Java, Indonesia https://orcid.org/0009-0007-9533-4474
Zulhendra Adi Kusuma Departement of Biomedical Engineering, Faculty of Engineering, Universitas Dian Nuswantoro, Semarang, Central Java, Indonesia
Susilo Susilo Departement of Biomedical Engineering, Faculty of Engineering, Universitas Dian Nuswantoro, Semarang, Central Java, Indonesia https://orcid.org/0000-0002-5736-5110
Sari Ayu Wulandari Departement of Biomedical Engineering, Faculty of Engineering, Universitas Dian Nuswantoro, Semarang, Central Java, Indonesia https://orcid.org/0000-0002-6287-4403

DOI: https://doi.org/10.35882/jeeemi.v8i2.1434

Keywords: Electrochemical Impedance Spectroscopy, Bone Mineral Density, Non-invasive diagnostics, Deep regression

Abstract

Electrochemical Impedance Spectroscopy (EIS) has emerged as a promising modality for non-invasive biomedical diagnostics, particularly for radiation-free monitoring tasks such as Bone Mineral Density (BMD) assessment. However, the high dimensionality, noise, and non-linear behavior of impedance signals pose significant challenges for accurate and interpretable prediction. This study introduces Hybrid Rough Set-Attention Network (HRSA-Net), a hybrid regression framework that combines Rough Set-based feature selection with a self-attention neural architecture to enable continuous BMD estimation directly from raw EIS data. The proposed framework employs Artificial Neural Network (ANN) and Transformer-based regression models to learn complex impedance-density relationships. Unlike prior studies that are limited to classification tasks or rely on indirect physiological indicators, HRSA-Net is explicitly designed for direct regression of real-valued BMD scores. The model performance is evaluated against reference measurements obtained from Dual-energy X-ray Absorptiometry (DXA), the current clinical gold standard for bone density assessment. Through a comprehensive series of ablation experiments, HRSA-Net achieves an R² of 0.834 using an attention-guided ANN backbone, demonstrating the critical contribution of both Rough Set reduction and attention mechanisms. Performance further improves to an R² of 0.855 when incorporating a Transformer regressor and Huber loss, indicating superior robustness and generalizability under varying signal conditions. Comparative analysis with state-of-the-art EIS-based learning approaches shows that the proposed pipeline consistently outperforms conventional neural models and statistical methods. Overall, HRSA-Net provides an interpretable, accurate, and scalable foundation for future portable EIS-based BMD diagnostic systems, offering a safer alternative to radiological methods such as DXA and enabling feasible deployment in primary or community healthcare settings

Downloads

Download data is not yet available.

References

J. A. Kanis, N. C. Harvey, E. McCloskey, H. Johansson, and C. Oden, “European guidance for the diagnosis and management of osteoporosis,” Osteoporos. Int., vol. 30, pp. 3–44, 2019. doi:10.1007/s00198-018-4704-5.

NIH Consensus Development Panel, “Osteoporosis prevention, diagnosis, and therapy,” JAMA, vol. 285, no. 6, pp. 785–795, 2001. doi:10.1001/jama.285.6.785.

J. A. Kanis, L. J. Melton III, C. Christiansen, C. C. Johnston, and N. Khaltaev, “The diagnosis of osteoporosis,” J. Bone Miner. Res., vol. 9, no. 8, pp. 1137–1141, 1994. doi:10.1002/jbmr.5650090802.

G. M. Blake and I. Fogelman, “The role of DXA bone density scans in the diagnosis and treatment of osteoporosis,” Postgrad. Med. J., vol. 83, pp. 509–517, 2007. doi:10.1136/pgmj.2007.057505.

E. M. Lewiecki, “DXA in clinical practice,” Bone, vol. 104, pp. 15–21, 2017. doi:10.1016/j.bone.2017.01.017.

Armañac-Julián P, Kontaxis S, Rapalis A, Marozas V, Laguna P, Bailón R, Gil E and Lázaro J (2022) Reliability of pulse photoplethysmography sensors: Coverage using different setups and body locations. Front. Electron. 3:906324. doi: 10.3389/felec.2022.906324.

Y. Moon, Z. Dong, S. K. Lee, et al., "Bioelectrical impedance analysis of bone mineral content based on dual-energy X-ray absorptiometry: evaluation of age-stratified optimized models," Sci. Rep., vol. 15, p. 24658, 2025. DOI: 10.1038/s41598-025-08304-8.

K. Ain, A. P. Putra, O. N. Rahma, D. Hikmawati, A. Rahmatillah, and C. A. C. Abdullah, "Electrical Impedance Spectroscopy as a Potential Tool for Detecting Bone Porosity," Sensors and Actuators A: Physical, vol. 370, p. 115252, 2024. DOI: 10.1016/j.sna.2024.115252.

M. Aldobali, K. Pal, and H. Chhabra, "Noninvasive health monitoring using bioelectrical impedance analysis," in Computational Intelligence in Healthcare Applications, R. Agrawal, M.A. Ansari, R.S. Anand, S. Sneha, and R. Mehrotra, Eds. Academic Press, 2022, ch. 14, pp. 209–236. DOI: 10.1016/B978-0-323-99031-8.00008-9.

N. Fukase et al., "Wireless Measurements Using Electrical Impedance Spectroscopy to Monitor Fracture Healing," Sensors, vol. 22, no. 16, p. 6233, Aug. 2022. DOI: 10.3390/s22166233.

Y. M. Liu, S. Q. Li, G. Q. Liu, Z. G. Lyu, and Z. W. Cui, "The Experimental Study of Bone Quality Assessment Based on Cole-Cole Parameter Characterization," in The Proceedings of the 19th Annual Conference of China Electrotechnical Society (ACCES 2024), vol. 1409, Q. Yang, Z. Bie, and X. Yang, Eds. Singapore: Springer, 2025, pp. (nomor halaman jika ada). DOI: 10.1007/978-981-96-4787-3_65.

T. K. Vashishth, V. Sharma, K. K. Sharma, B. Kumar, S. Chaudhary, and R. Panwar, "Enhancing biomedical signal processing with machine learning: A comprehensive review," in Digital Transformation in Healthcare 5.0: Volume 1: IoT, AI and Digital Twin, Berlin, Germany: De Gruyter, 2024, ch. 10, pp. 277–306. DOI: 10.1515/9783111327853.

Z. Shao, J. Wu, Q. Deng, L. Cheng, X. Huang, W. Sun, W. Liang, and H. Li, "Development of a machine learning-based predictive model for osteoporosis risk and its application in clinical decision support," Front. Med., vol. 12, p. 1680731, 2025. DOI: 10.3389/fmed.2025.1680731.

R. Bagaria, S. Wadhwani, and A. K. Wadhwani, "Bone fractures detection using support vector machine and error backpropagation neural network," Optik, vol. 247, p. 168021, 2021. DOI: 10.1016/j.ijleo.2021.168021.

A. A. Ewees, M. E. Abd Elaziz, R. Arafa, and R. Ghoniem, "Improved Approach Based on Fuzzy Rough Set and Sine-Cosine Algorithm: A Case Study on Prediction of Osteoporosis," IEEE Access, vol. 8, pp. 203190–203202, 2020. DOI: 10.1109/ACCESS.2020.3036831.

M. F. Dar and A. Ganivada, "Fuzzy rough set loss for deep learning-based precise medical image segmentation," Comput. Med. Imaging Graph., vol. 128, p. 102716, Feb. 2026. DOI: 10.1016/j.compmedimag.2026.102716.

Y. Sato, N. Yamamoto, N. Inagaki, Y. Iesaki, T. Asamoto, T. Suzuki, and S. Takahara, "Deep Learning for Bone Mineral Density and T-Score Prediction from Chest X-rays: A Multicenter Study," Biomedicines, vol. 10, no. 9, p. 2323, Sep. 2022. DOI: 10.3390/biomedicines10092323.

Z. Lin, Z. Q. Lang, L. Guo, et al., "Deep learning-based electrical impedance spectroscopy analysis for malignant and potentially malignant oral disorder detection," Sci. Rep., vol. 15, p. 19458, 2025. DOI: 10.1038/s41598-025-05116-8

K. Umam, I. K. E. Purnama, R. F. Rachmadi, and S. A. Wulandari, "Detection of Mycobacterium Tuberculosis Using Real-time Detection Transformer," in 2024 International Seminar on Intelligent Technology and Its Applications (ISITIA), Mataram, Indonesia, 2024, pp. 633–638. DOI: 10.1109/ISITIA63062.2024.10667784

S. A. Wulandari, I. K. E. Purnama, E. M. Yuniarno, and M. H. Purnomo, "Multicluster Kernel Intuitionistic Fuzzy C-Means and State Transition Algorithm: Framework for Low Light Segmentation Imaging of Tuberculosis Bacilli Base of Semisupervised Approach," in 2024 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE), Yokohama, Japan, 2024, pp. 1–10. DOI: 10.1109/FUZZ-IEEE60900.2024.10612071

S. A. Wulandari, I. K. E. Purnama, and M. H. Purnomo, "Auto Thresholding Sputum Color Image Segmentation For Tuberculosis Diagnosis Base On Intuitionistic Fuzzy," in 2022 International Conference on Computer Engineering, Network, and Intelligent Multimedia (CENIM), Surabaya, Indonesia, 2022, pp. 151–156. DOI: 10.1109/CENIM56801.2022.10037568.

S. A. Wulandari, E. L. Agata, N. M. C. E, N. Alriski, M. T. Wafa’Ana, and A. Aripin, "Attention-Enhanced CNN-BiLSTM for Bone Mineral Density Estimation from Electrochemical Impedance Spectra," in 2025 International Seminar on Application for Technology of Information and Communication (iSemantic), Semarang, Indonesia, 2025, pp. 260–266. DOI: 10.1109/ISemantic67418.2025.11291862

D. Nurdiyah, E. M. Yuniarno, S. A. Wulandari, Y. K. Surapto, and M. H. Purnomo, "Deep Semantic Feature Extraction to Overcome Overlapping Frequencies for Instrument Recognition in Indonesian Traditional Music Orchestras," IEEE Access, vol. 12, pp. 76936–76954, 2024. DOI: 10.1109/ACCESS.2024.3401699.

Muljono, P. N. Andono, S. A. Wulandari, H. A. Azies, and M. Naufal, "Deep learning for audio signal-based tempo classification scenarios," IAES Int. J. Artif. Intell., vol. 13, no. 2, pp. 1687–1701, Jun. 2024. DOI: 10.11591/ijai.v13.i2.pp1687-1701

Muljono et al., "Breaking Boundaries in Diagnosis: Non-Invasive Anemia Detection Empowered by AI," IEEE Access, vol. 12, pp. 9292–9307, 2024. DOI: 10.1109/ACCESS.2024.3353788.

X. Yao, X. Zhang, E. Yang, and Y. Fan, "Variable precision -fuzzy rough sets and their applications in image processing and fuzzy decision trees," Pattern Recognit., vol. 179, pt. C, p. 113796, Nov. 2026. DOI: 10.1016/j.patcog.2026.113796.

P. Jain and T. Som, "Multigranular rough set model based on robust intuitionistic fuzzy covering with application to feature selection," Int. J. Approx. Reason., vol. 156, pp. 16–37, May 2023. DOI: 10.1016/j.ijar.2023.02.004.

D. Formica and E. Schena, "Smart Sensors for Healthcare and Medical Applications," Sensors, vol. 21, no. 2, p. 543, Jan. 2021. DOI: 10.3390/s21020543.

M. M. Sibhai, A. Alkhateeb, and S. B. Ahmed, "MEDFORMER-UR: Uncertainty-Routed Transformer for Medical Image Classification," arXiv preprint arXiv:2604.08868, Apr. 2026. DOI: 10.48550/arXiv.2604.08868.

M. K. Umair, R. Waheed, M. F. Abrar, et al., "Time series electrocardiography (ECG) data for early prediction of cardiac arrest," Sci. Rep., vol. 16, p. 9761, 2026. DOI: 10.1038/s41598-026-35788-9.

H. A. Tinoco et al., "Bio-structural monitoring of bone mineral alterations through electromechanical impedance measurements of a Piezo-device joined to a tooth," Biomed. Eng. Lett., vol. 10, no. 4, pp. 603–617, Sep. 2020. DOI: 10.1007/s13534-020-00170-9.

A. Banerjee, Y. Tai, N. V. Myung, and J. Nam, "Non-destructive characterization of bone mineral content by machine learning-assisted electrochemical impedance spectroscopy," Front. Bioeng. Biotechnol., vol. 10, p. 961108, Sep. 2022. DOI: 10.3389/fbioe.2022.961108.

M. Han et al., “Deep learning guided electrochemical impedance spectroscopy,” ACS Sensors, vol. 9, pp. 214–224, 2024. doi: 10.1021/acssensors.3c02039.

X. Ma, L. Bifano, and G. Fischerauer, "Evaluation of Electrical Impedance Spectra by Long Short-Term Memory to Estimate Nitrate Concentrations in Soil," Sensors, vol. 23, no. 2, p. 2172, Feb. 2023. DOI: 10.3390/s23042172.

N. V. Minh et al., "Physics-encoded machine learning for performance and emission prediction of nickel ferrite nanocatalyst and hydrogen-enriched biodiesel in diesel engines," RSC Adv., vol. 16, no. 12, pp. 10798-10821, Feb. 2026, doi: 10.1039/d5ra09336h.

E. Oostwal, M. Straat, and M. Biehl, "Hidden unit specialization in layered neural networks: ReLU vs. sigmoidal activation," Phys. A Stat. Mech. Appl., vol. 564, p. 125517, Feb. 2021. DOI: 10.1016/j.physa.2020.125517.

A. Dixit, A. Mandal, S. Sanyal, and S. S. Ganguli, "A Genetic-Evolutionary ADAM (G-ADAM) optimizer for semisupervised convolutional neural network-based acoustic impedance inversion," in SEG Technical Program Expanded Abstracts 2021, 2021, pp. 2208–2212. DOI: 10.1190/segam2021-3594593.1.

X. Ji, W. Duan, J. Peng, and S. Yao, "Fuzzy rough set attribute reduction based on decision ball model," Int. J. Approx. Reason., vol. 179, p. 109364, Apr. 2025. DOI: 10.1016/j.ijar.2025.109364.

V. Diukarev and Y. Starukhin, "Proposed Methods for Preventing Overfitting in Machine Learning and Deep Learning," Asian J. Res. Comput. Sci., vol. 17, no. 10, pp. 85–94, Oct. 2024. DOI: 10.9734/ajrcos/2024/v17i10511

R. White et al., "Robust Optimal Experimental Design Accounting for Sensor Failure," arXiv preprint arXiv:2604.14497, Apr. 2026. DOI: 10.48550/arXiv.2604.14497.

L. Q. Trung, N. Kasai, M. Le, and K. Sekino, "Predicting actual crack size through crack signal obtained by advanced Flexible Eddy Current Sensor using ResNet integrated with CBAM and Huber loss function," NDT & E Int., vol. 149, p. 103249, Oct. 2024. DOI: 10.1016/j.ndteint.2024.103249.

J. A. Ribeiro and P. A. S. Jorge, "Applications of electrochemical impedance spectroscopy in disease diagnosis—A review," Sensors Actuators Rep., vol. 8, p. 100205, Dec. 2024. DOI: 10.1016/j.snr.2024.100205.

L.-F. Mao, S. K. Li, P. Cheng, and L. J. Zhang, "Addressing challenges inverse problem with convolutional neural networks and regulation techniques: Applications in extraction of physical parameters of semiconductors devices," Int. J. Electr. Power Energy Syst., vol. 161, p. 110172, Oct. 2024. DOI: 10.1016/j.ijepes.2024.110172.

J. Dardé, N. Nasr, and L. Weynans, "Immersed boundary method for the complete electrode model in electrical impedance tomography," J. Comput. Phys., vol. 487, p. 112150, Aug. 2023. DOI: 10.1016/j.jcp.2023.112150.

R. E. Meethal, A. Kodakkal, M. Khalil, et al., "Finite element method-enhanced neural network for forward and inverse problems," Adv. Model. and Simul. in Eng. Sci., vol. 10, no. 6, 2023. DOI: 10.1186/s40323-023-00243-1.

R. Alaifari, "Ill-Posed Problems: From Linear to Nonlinear and Beyond," in Harmonic and Applied Analysis, F. De Mari and E. De Vito, Eds. Cham, Switzerland: Birkhäuser, 2021, pp. 101–142. DOI: 10.1007/978-3-030-86664-8_3

Bhardwaj P, Rai DV, Garg ML, Mohanty BP. Potential of electrical impedance spectroscopy to differentiate between healthy and osteopenic bone. Clin Biomech (Bristol). 2018 Aug;57:81-88. doi: 10.1016/j.clinbiomech.2018.05.014. Epub 2018 May 26. PMID: 29960118.

C. Liebich, J. N. Bartsch, I. Schubert, M.-L. von Bruehl, and C. Sander, "Electrical Impedance Spectroscopy Improves Skin Cancer Detection and Reduces the Number of Biopsies," Dermato, vol. 2, no. 2, pp. 21–29, 2022. DOI: 10.3390/dermato2020004.

A. Crowell, J.S. Yakisich, T.N.G. Adams, “Electrical Impedance Spectroscopy to Discriminate Between Normal and Cancerous Mammalian Cells,” Biosensors (Basel), vol. 11, no. 10, p. 401, Oct. 2021, doi:10.3390/bios11100401.

Q. Gong et al., "Non-Invasive and Accurate Blood Glucose Detection Based on an Equivalent Bioimpedance Spectrum," Appl. Sci., vol. 15, no. 3, p. 1266, Jan. 2025. DOI: 10.3390/app15031266.

M. Faisal et al., "Development of Electrical Impedance Spectroscopy (EIS) Technique to Classify Diabetes Mellitus Disease Using Machine Learning with Backpropagation Method," Trends Sci., vol. 22, no. 9, p. 10309, 2025. DOI: 10.48048/tis.2025.10309.

I. Amira et al., "The Degradation Test of IBS (Injectable Bone Substitutes) Paste Scaffold Using EIS (Electrical Impedance Spectroscopy) Methods," Jurnal Teknologi, vol. 88, no. 1, Jan. 2026. DOI: 10.11113/jurnalteknologi.v88.23560.

J. Chen et al., "Machine learning-directed electrical impedance tomography to predict metabolically vulnerable plaques," Bioeng. Transl. Med., vol. 9, no. 1, p. e10616, Oct. 2023. DOI: 10.1002/btm2.10616.