Downloads

https://doi.org/10.62836/jcmea.v6i1.0001
Dai , Y. (2026). Medical Biopharmaceutical Image Anomaly Detection under Retinex State Space Duality and Frequency Consensus-Driven Transformer. Journal of Computational Methods in Engineering Applications, 6(1), 0001. https://doi.org/10.62836/jcmea.v6i1.0001
Volume 6, Issue 1, 2026

Medical Biopharmaceutical Image Anomaly Detection under Retinex State Space Duality and Frequency Consensus-Driven Transformer

With the rapid development of medical imaging technologies and biopharmaceutical research, a large amount of high-dimensional and complex medical and biological image data is continuously being generated. Achieving high-precision anomaly detection under conditions of complex backgrounds and low contrast has become an important research problem in the fields of intelligent healthcare and drug development. Traditional anomaly detection methods often struggle to achieve stable and robust detection performance when dealing with issues commonly present in medical images, such as uneven illumination, complex structures, and insufficient utilization of frequency information. To address these challenges, this paper proposes a Transformer-based anomaly detection method for medical and biopharmaceutical images driven by Retinex state-space duality and frequency consensus. By integrating spatial-domain and frequency-domain features, the proposed method enhances the model’s ability to perceive complex abnormal structures. Specifically, the method first employs Retinex state-space duality to decompose medical images into structural and illumination components, thereby strengthening the structural representation of anomalous regions while reducing interference caused by illumination variations. Subsequently, a frequency consensus-driven mechanism is introduced to model feature consistency across different scales in the frequency domain, enabling adaptive enhancement of key anomalous frequency features. On this basis, a global context modeling framework is constructed by incorporating a Vision Transformer, which captures long-range dependencies and potential anomalous patterns in medical images, further improving detection accuracy and feature representation capability. To verify the effectiveness of the proposed method, extensive experiments are conducted on multiple medical and biopharmaceutical image datasets, and comparisons are made with several mainstream anomaly detection models. Experimental results demonstrate that the proposed method outperforms the comparison methods across evaluation metrics, achieving more stable and accurate anomaly detection in complex medical imaging environments. This approach provides a technically promising solution with potential application value for automated image analysis in intelligent pharmaceutical inspection and drug development.

Vision Transformer Retinex state-space duality frequency consensus-driven anomaly detection data analysis

References

  1. Shome D, Sarkar P, Etemad A. Region-Disentangled Diffusion Model for High-Fidelity PPG-to-ECG Translation. In Proceedings of the AAAI Conference on Artificial Intelligence, Vancouver, BC, Canada, 20–27 February 2024; Volume 38, pp. 15009–15019.
  2. Yalcouyé A, Esoh K, Wonkam A. Enhancing Large-Scale Pharmacogenetic Studies in African Populations for Clinical Care and Drug Development. Annual Review of Pharmacology and Toxicology 2026; 66: 171–190.
  3. Lee J, Sulahria A, Luce S, et al. The Clinical Research Assistant. In Clinical Research in Private Practice; Academic Press: Cambridge, MA, USA, 2026; pp. 117–126.
  4. Zhang Y, Mohsin SM, Mujlid H, et al. AI-Driven Blockchain Technology in Smart Healthcare System: Opportunities, Challenges and Future Implications. Computer Science Review 2026; 60: 100909.
  5. Xu D, Chen Y, Chai Z, et al. Knowledge Fusion in Deep Learning-Based Medical Vision-Language Models: A Review. Information Fusion 2026; 125: 103455.
  6. Zhang L, Li Z, Cheng L, et al. DLIENet: A Lightweight Low-Light Image Enhancement Network via Knowledge Distillation. Pattern Recognition 2026; 169: 111777.
  7. Xu M, Wei H, Nie Z, et al. Hybrid Dual-Heterogeneous Knowledge Distillation Network for Anomaly Detection in Retinal OCT Images. IEEE Journal of Biomedical and Health Informatics 2026. https://doi.org/10.1109/JBHI.2025.3650072
  8. Verma RK. AI-Driven Image Enhancement Techniques for Edge Devices Balancing Quality and Performance. In Advances in Image Processing, Reliability, and Artificial Intelligence; Elsevier: Amsterdam, The Netherlands, 2026; pp. 151–168.
  9. Irede EL, Aworinde OR, Lekan OK, et al. Medical Imaging: A Critical Review on X-ray Imaging for the Detection of Infection. Biomedical Materials & Devices 2026; 4(1): 1–45.
  10. Ahn S, Cho TS, Park SH. The Effects of Organizational Status on Market Entry Timing and Performance: The Biopharmaceutical Industry Amid Environmental Disruption. Journal of Business Research 2026; 205: 115893.
  11. Hui RW, Wu TK, Ho KC, et al. Large-Scale Profile Study on Hepatitis B Surface Antigen Levels in Chronic Hepatitis B: Implications for Drug Development Targeting Functional Cure. Gut 2026; 75(1): 119–130.
  12. Cakmak Y, Pacal I. A Comparative Analysis of Transformer Architectures for Automated Lung Cancer Detection in CT Images. Journal of Intelligent Decision Making and Information Science 2026; 3: 528–539.
  13. Sharma K, Hansen J, Susztak K, et al. Spatial Metabolomics and Multiomics Integration for Breakthroughs in Precision Medicine for Kidney Disease. Nature Reviews Nephrology 2026; 22(2): 152–164.
  14. Liu Y, Deng H, Fu J. DCM-Net: A Novel Dual-Branch CNN–Mamba Cross-Layer Feature Fusion Network for Medical Image Segmentation. Biomedical Signal Processing and Control 2026; 114: 109267.
  15. Yang J, Qiu P, Zhang Y, et al. D-Net: Dynamic Large Kernel with Dynamic Feature Fusion for Volumetric Medical Image Segmentation. Biomedical Signal Processing and Control 2026; 113: 108837.
  16. Dalmonte F, Bayar E, Akbas E, et al. Q-Former Autoencoder: A Modern Framework for Medical Anomaly Detection. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision 2026, Tucson, AZ, USA, 6–10 March 2026.
  17. Alrfou K, Zhao T. GC-UNet: Efficient Network for Medical Image Segmentation. Multimedia Tools and Applications 2026; 85(2): 137.
  18. Martínez H, Gómez-Luna J, Palomar R, et al. In-Memory Operators for Medical Image Processing. Future Generation Computer Systems 2026; 174: 107939.
  19. Sachin AS, Karthikeyan J. A Hybrid-Based Deep Learning Framework for Pill Detection, Multi-Attribute Classification and Prediction with Metadata Retrieval. IEEE Access 2026; 14: 17905–17919.
  20. Qomariah DU, Elvira AI, Kurniasari AA, et al. Automatic Pill Counting Using YOLOv8 to Improve Medication Distribution Accuracy. International Journal of Public Health and Epidemiology 2026; 5(2): 28–33.
  21. Usuyama N, Delgado NL, Hall AK, et al. ePillID Dataset: A Low-Shot Fine-Grained Benchmark for Pill Identification. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops 2020, Seattle, WA, USA, 14–19 June 2020.
  22. Peters L, Mortensen J, Nguyen T, et al. Enabling Complex Queries to Drug Information Sources through Functional Composition. Studies in Health Technology and Informatics 2013; 192: 692.
  23. Kim S, Jo Y, Cho J, et al. Spatially Variant Convolutional Autoencoder Based on Patch Division for Pill Defect Detection. IEEE Access 2020; 8: 216781–216792.
  24. Roy Choudhury K, Singh A, Padmini S. Detection of Pharmaceutical Pill Defects Through Deep One-Class Classification. In International Conference on Data Analytics & Management; Springer Nature: Singapore, 2024.
  25. Ou YY, Tsai AC, Zhou XP, et al. Automatic Drug Pills Detection Based on Enhanced Feature Pyramid Network and Convolution Neural Networks. IET Computer Vision 2020; 14(1): 9–17.
  26. Lee Y, Lim H, Jang S, et al. Uniformaly: Towards Task-Agnostic Unified Framework for Visual Anomaly Detection. Pattern Recognition 2026; 169: 111820.
  27. Chen Y, Tao X, Chen B, et al. MPFR: Memory Prompt Feature Reconstruction for Continual Anomaly Detection and Segmentation. Pattern Recognition 2026; 175: 112946.
  28. Liu X, Wu C, Zhang H, et al. A Memory-Tree Driven Network for Multi-View Fusion Anomaly Detection. Pattern Recognition 2026; 170: 112106.
  29. Wang B, Wan J, Zhao J, et al. A Twin-Branch Decoupled Network for Multi-Class Unsupervised Anomaly Detection. Engineering Applications of Artificial Intelligence 2026; 167: 113891.
  30. Deng X, Oda S, Kawano Y. Graphene-Based Midinfrared Photodetector with Bull’s Eye Plasmonic Antenna. Optical Engineering 2023; 62(9): 097102.
  31. Li J, Culver TB, Burgis CR, et al. Validating Nitrogen Removal Models with Field Bioretention Data. Journal of Environmental Engineering 2024; 150(8): 04024037.
  32. Yan H. Real-Time 3D Model Reconstruction Through Energy-Efficient Edge Computing. Optimizations in Applied Machine Learning 2022; 2(1). https://doi.org/10.71070/oaml.v2i1.48
  33. Li J, Culver TB, Persaud PP, et al. Developing Nitrogen Removal Models for Stormwater Bioretention Systems. Water Research 2023; 243: 120381.
  34. Yan H, Shao D. Multimodal Medical Image Analysis: Integrating LLM and RAG Deep Learning Strategies. Journal of Advances in Information Technology 2025; 16(4): 568–581. https://doi.org/10.12720/jait.16.4.568-581.
  35. Lu Y, Shao D, Ni X, et al. Emotion-Style Dual Prediction: A Multi-Task Deep Learning Approach for Artistic Images. Cluster Computing 2026; 29(1): 31.
  36. Deng X, Kawano Y. Surface Plasmon Polariton Graphene Midinfrared Photodetector with Multifrequency Resonance. Journal of Nanophotonics 2018; 12(2): 026017.
  37. Li J. Nitrogen Removal Models for Stormwater Bioretention Systems. Ph.D. Thesis, University of Virginia, Charlottesville, VA, USA, 2023.
  38. Yan H, Shao D. Enhancing Transformer Training Efficiency with Dynamic Dropout. arXiv 2024, arXiv:2411.03236. https://doi.org/10.48550/arXiv.2411.03236.
  39. Luo Z, Yan H, Pan X. Optimizing Transformer Models for Resource-Constrained Environments: A Study on Model Compression Techniques. Journal of Computational Methods in Engineering Applications 2023; 3: 1–12. https://doi.org/10.62836/jcmea.v3i1.030107.
  40. Li J, Culver TB. Review of Process-Based Nitrogen Model for Agricultural Fields with Implications for Nitrogen Simulations in Stormwater BMPs. Environmental Modelling & Software 2022; 151: 105363.

Supporting Agencies

  1. Funding: This research received no external funding.