Iranian Journal of Medical Physics

Iranian Journal of Medical Physics

AI-Driven Identification of Microplastics via STIM-Based Proton Transmission: A Computational Proof-of-Concept

Document Type : Original Paper

Authors
Marjan Atghaei 1
Mohamad Reza Rezaie Rayeni Nejad 2
Hossien Afarideh 2
Ali Negarestani 1
1 Faculty of Sciences and Modern Technologies, Graduate University of Advanced Technology
2 Department of Nuclear Engineering, Faculty of Physics and Energy Engineering, Amirkabir University of Technology, Tehran, Iran
10.22038/ijmp.2026.95156.2685
Abstract
Introduction: This simulation-driven study investigated the potential of a non-destructive detection system concept for microplastic (MP) pollution using a computational framework. Traditional detection methods are often time-consuming or lack the sensitivity required for real-time monitoring. We proposed an integrated approach combining proton Scanning Transmission Ion Microscopy (STIM) with Metal-Organic Framework (MOF) nanofilters and deep learning-based analysis.
Material and Methods: Computational modeling was conducted using the Monte Carlo N-Particle extended (MCNPX) code to simulate proton beam interactions (at 30 and 60 MeV) with MPs. Polyimide was utilized as a simulation proxy for the MIL-101(Cr) MOF nanofilters to represent the structural density and elemental composition. The physical changes in the proton Bragg peak and energy loss spectra were analyzed. To automate the detection process, a sequential-attention architecture integrating Bidirectional Long Short-Term Memory (BiLSTM) and Transformer models was developed and trained on the simulated spectral data.
Results: The simulation-derived results suggested that the integration of MOF nanofilters could enhance the sensitivity of STIM to minor density variations caused by microplastics within a controlled computational environment. The BiLSTM-Transformer model achieved high predictive accuracy on the simulated dataset, with an R² value of 0.982 and a Mean Absolute Error (MAE) of 0.015. Detailed statistical uncertainties (standard deviations) for the MCNPX outputs are reported in the Results section.
Conclusion: This simulation study demonstrates the theoretical feasibility of using AI-enhanced proton STIM for the identification of microplastics. While the current findings are based on deterministic Monte Carlo outputs, the high accuracy of the BiLSTM-Transformer model provides a computational proof-of-concept. Future experimental work is required to bridge the gap between simulation and real-world cyclotron-based monitoring systems, particularly regarding electronic noise and detector efficiency.
Keywords
Microplastic Scanning Ion Transmission Microscopy (STIM) MIL-101(Cr) Nanofilter MCNPX Simulation Bi LSTM&ndash
Transformer Neural Network
Subjects
1.        Jaikumar IM, et al. A comprehensive review of microplastic pollution in freshwater and marine environments. Green Analytical Chemistry. 2024:100202.
2.        Chaudhary HD, Shah G, Bhatt U, Singh H, Soni V. Microplastics and plant health: a comprehensive review of sources, distribution, toxicity, and remediation. npj Emerging Contaminants. 2025;1(1):8.
3.        Egbueri JC, et al. Microplastic contamination in Nigerian treated waters and packaged (sachet, bottled) sources: trends, regional disparities, and policy implications for sustainable practices. Analytical Letters. 2025;58(5):769-805.
4.        Hossain MI, Zhang Y, Haque ANMA, Naebe M. Fibrous microplastics release from textile production phases: a brief review of current challenges and applied research directions. Materials. 2025;18(11):2513.
5.        Austen B. Global efforts to combat marine plastic pollution: addressing single-use plastics and microbeads through policy and innovation. SSRN Electronic Journal. 2025; doi: 10.2139/ssrn.5118234.
6.        Choudhary A, et al. Exploring microplastic pollution from origin to environmental impact and remediation approaches. Discover Environment. 2025;3(1):1-29.
7.        Aib H, Parvez MS, Czédli HM. Pharmaceuticals and microplastics in aquatic environments: a comprehensive review of pathways and distribution, toxicological and ecological effects. International Journal of Environmental Research and Public Health. 2025;22(5):799.
8.        Lane T, Wardani I, Koelmans AA. Exposure scenarios for human health risk assessment of nano-and microplastic particles. Microplastics and Nanoplastics. 2025;5(1):28.
9.        Yang T, et al. Mechanistic insights into heavy metal adsorption and co-transport by UV-aged microplastics: a combined experimental and DFT study. Chemical Engineering Journal. 2025:171838.
10.     Thakur S, Chandra A, Kumar V, Bharti S. Environmental pollutants: endocrine disruptors/pesticides/reactive dyes and inorganic toxic compounds metals, radionuclides, and metalloids and their impact on the ecosystem. In: Biotechnology for environmental sustainability. Springer; 2025. p. 55-100.
11.     Barik D, Rakhi Mol K, Anand G, Nandamol P, Das D, Porel M. Environmental pollutants such as endocrine disruptors/pesticides/reactive dyes and inorganic toxic compounds metals, radionuclides, and metalloids and their impact on the ecosystem. In: Biotechnology for Environmental Sustainability. Springer; 2025. p. 391-442.
12.     Tarapoulouzi M, Ioannidis I, Pashalidis I. Applications, trends, and challenges in the non-destructive assessment of microplastics in fish. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy. 2025:126462.
13.     Nene A, et al. Recent advances and future technologies in nano-microplastics detection. Environmental Sciences Europe. 2025;37(1):7.
14.     Santos FA, Andre RS, Alvarenga AD, Alves ALM, Correa DS. Micro-and nanoplastics in the environment: a comprehensive review on detection techniques. Environmental Science: Nano. 2025;12(7):3442-3467.
15.     Wang J, Ma M, Wang L, Yu Y, Zhang L. Emerging techniques for environmental nanoplastic pollutants detecting. Nano Research. 2026;19(2):1120-1145.
16.     Song Y, Ma S. Pore engineering in metal–organic frameworks and covalent organic frameworks: strategies and applications. Chemical Science. 2025;16(4):892-915.
17.     Sun N, Shah SSA, Lin Z, Zheng YZ, Jiao L, Jiang HL. MOF-based electrocatalysts: an overview from the perspective of structural design. Chemical Reviews. 2025;125(5):2703-2792.
18.     Chen J, Tang M, Nie S, Xiao P, Zhao T, Chen Y. Synthesis methods, performance optimization, and application progress of metal–organic framework material MIL-101 (Cr). Chemistry. 2025;7(3):78.
19.     Feng J, Dong Y, Li H, Tu J, Chen Y. Engineered magnetic metal-organic frameworks for efficient and broad-spectrum adsorption of micro/nanoplastics in beverages. Journal of Hazardous Materials. 2025;495:139040.
20.     [20] Singh P, Kumar S, Asati V. Metal-organic framework-driven advances in solid phase microextraction for comprehensive environmental monitoring. Current Green Chemistry. 2025;12(1):45-68.
21.     Schwertner M, Sakellariou A, Reinert T, Butz T. Scanning transmission ion micro-tomography (STIM-T) of biological specimens. Ultramicroscopy. 2006;106(7):574-581.
22.     Paramasamy J. Impact of variations in patient and beam parameters on proton therapy using pencil beam scanning technique [Master’s Thesis]. University of Texas Health Science Center; 2024.
23.     Parashar M, Sharma M, Saini DK, Byers TA, Rout B. Rutherford backscattering spectrometry: a multimodal ion beam analysis technique for evaluating elemental depth profiles and radiation hardness for space photovoltaics applications. APL Energy. 2025;3(3).
24.     Whitlow HJ. On quantitative off-axis scanning transmission ion microscopy (STIM). Nuclear Instruments and Methods in Physics Research Section B: Beam Interactions with Materials and Atoms. 2023;540:51-61.
25.     Jing S, et al. Radiation shielding performance of CeF3-doped Li2O-PbO-GdF3-SiO2 glasses: Monte Carlo (MCNPX) and PHY-X/PSD study. Radiation Physics and Chemistry. 2025;228:112035.
26.     Cahill J, et al. Modeling and benchmarking XRF analysis using MCNP for applications in accident tolerant fuel and cladding. Progress in Nuclear Energy. 2025;178:105487.
27.     Lv H, Hao B. Convolutional neural network and long short-term memory hybrid model-based gait prediction for spacesuit intelligent assistive device in low gravity environment. Acta Astronautica. 2025;236:199-212.
28.     Riaz S, Saghir A, Khan MJ, Khan H, Khan HS, Khan MJ. TransLSTM: a hybrid LSTM-transformer model for fine-grained suggestion mining. Natural Language Processing Journal. 2024;8:100089.
29.     Buckton C. A hybrid deep learning based algorithm for gamma spectroscopy analysis [Thesis]. Stellenbosch: Stellenbosch University; 2025.
30.     Andayani F, Theng LB, Tsun MT, Chua C. Hybrid LSTM-transformer model for emotion recognition from speech audio files. IEEE Access. 2022;10:36018-36027.
 
Iranian Journal of Medical Physics
Volume 23, Issue 2 - Serial Number 2
March and April 2026
Pages 76-85

XML
PDF 1.42 M
Supplementary File
2685.pdf

  • Receive Date 16 April 2026
  • Revise Date 07 June 2026
  • Accept Date 04 July 2026