Machine Learning-Driven Image Fusion to Improve Diagnostic Accuracy in Medical Imaging

Muhammad Shafiq; Waeal J. Obidallah; Qun Wang; Tahir Kamal; Anas Bilal

doi:10.5566/ias.3781

Authors

Muhammad Shafiq School of Computer Science, Shandong Xiehe University, Jinan, China.
Waeal J. Obidallah College of Computer and Information Sciences, Imam Mohammad Ibn Saud Islamic University (IMSIU), Riyadh, Saudi Arabia.
Qun Wang School of Computer Science, Shandong Xiehe University, Jinan, China.
Tahir Kamal School of Information and Software Engineering, University of Electronic Science and Technology of China, Chengdu, China.
Anas Bilal College of Information Science & Technology, Hainan Normal University, Haikou, China.

DOI:

https://doi.org/10.5566/ias.3781

Keywords:

Medical Image Processing, Image Fusion Model, Disease Prediction, MRI, Dual-Channel Pulse-Coupled Neural Network, Machine Learning

Abstract

Image-based disease diagnosis and treatment have long been used to refine human lives. Recent advancements in imaging and image processing technologies have attracted significant research in this discipline, enhancing image modalities to better represent features and help healthcare practitioners make more precise findings and treatments. Various medical image modalities are in use, including X-rays, Positron Emission Tomography (PET), Computer Tomography (CT), and Magnetic Resonance Imaging (MRI). Each modality has its strengths; for instance, MRI provides detailed anatomical perceptions, while PET reveals functional and metabolic information. However, using these modalities separately limits diagnostic probability, as they cannot propose a comprehensive view of structure and function. The Image Fusion Model (IFM) helps to overcome this limitation by combining features from both images. Existing IFMs have limitations, such as loss of high-frequency details, underprivileged retention of structural data, and insufficient preservation of functional data. This recommended model combines the Fast Discrete Cosine Transform (FDCT), the HSV color model, and a Dual-Channel Pulse-Coupled Neural Network (DC-PCNN) to address these drawbacks. The model was evaluated using eight Harvard Medical School Image Database images for each modality, achieving better spatial frequency values and other metrics than previous models.

References

Abhisheka B, Biswas SK, Purkayastha B (2023). A comprehensive review on breast cancer detection, classification and segmentation using deep learning. Arch Comput Methods Eng 30:5023-52.

Abhisheka B, Biswas SK, Purkayastha B, Das D, Escargueil A (2024). Recent trend in medical imaging modalities and their applications in disease diagnosis: a review. Multimed Tools Appl 83:43035-70.

Amiri E, Rahmanian M, Amiri S, Yazdani Praee H (2021). Medical images fusion using two-stage combined model DWT and DCT. Int Adv Res Eng J 5:344-51.

Arabi H, Saberi Manesh A, Zaidi H (2024). Innovations in dedicated PET instrumentation: from the operating room to specimen imaging. Phys Med Biol 69:11TR03.

Ashwanth B, Swamy KV (2020). Medical image fusion using transform techniques. In: Proc. 2020 5th Int Conf Devices, Circuits and Systems (ICDCS), 2020 Mar 5-6; Coimbatore, India. Piscataway (NJ): IEEE, 303-6.

Bilal A, Imran A, Baig TI, Liu X, Long H, Alzahrani A, Shafiq M (2024a). DeepSVDNet: a deep learning-based approach for detecting and classifying vision-threatening diabetic retinopathy in retinal fundus images. Comput Syst Sci Eng 48:511-28.

Bilal A, Imran A, Liu X, Liu X, Ahmad Z, Shafiq M, El-Sherbeeny AM, Long H (2024b). BC-QNet: a quantum-infused ELM model for breast cancer diagnosis. Comput Biol Med 175:108483.

Bilal A, Liu X, Baig TI, Long H, Shafiq M (2023). EdgeSVDNet: 5G-enabled detection and classification of vision-threatening diabetic retinopathy in retinal fundus images. Electronics 12:4094.

Bilal A, Liu X, Shafiq M, Ahmed Z, Long H (2024c). NIMEQ-SACNet: a novel self-attention precision medicine model for vision-threatening diabetic retinopathy using image data. Comput Biol Med 171:108099.

Chen C-I (2017). Fusion of PET and MR brain images based on IHS and log-Gabor transforms. IEEE Sens J 17:6995-7010.

Eckhorn R, Reitboeck HJ, Arndt M, Dicke P (1989). A neural network for feature linking via synchronous activity: results from cat visual cortex and from simulations. In: Cotterill RMJ, ed. Models of brain function. Cambridge: Cambridge University Press, 255-72.

Furtado FS, Wu MZ, Esfahani SA, Ferrone CR, Blaszkowsky LS, Clark JW, Ryan DP, Goyal L, Franses JW, Wo JY, Hong TS, Qadan M, Tanabe KK, Weekes CD, Cusack JC, Crafa F, Mahmood U, Anderson MA, Mojtahed A, Hahn PF, Caravan P, Kilcoyne A, Vangel M, Striar RM, Rosen BR, Catalano OA (2023). Positron emission tomography/magnetic resonance imaging (PET/MRI) versus the standard of care imaging in the diagnosis of peritoneal carcinomatosis. Ann Surg 277:e893-9.

Haddadpour M, Daneshvar S, Seyedarabi H (2017). PET and MRI image fusion based on combination of 2-D Hilbert transform and IHS method. Biomed J 40:219-25.

Haribabu M, Guruviah V, Yogarajah P (2023). Recent advancements in multimodal medical image fusion techniques for better diagnosis: an overview. Curr Med Imaging 19:673-94.

Islam SMS, Nasim MAA, Hossain I, Ullah DMA, Gupta DKD, Bhuiyan MMH (2023). Introduction of medical imaging modalities. In: Zheng B, Andrei S, Sarker MK, Gupta KD, eds. Data-driven approaches on medical imaging. Cham: Springer, 1-25.

Johnson JL, Padgett ML (1999). PCNN models and applications. IEEE Trans Neural Netw 10:480-98.

Johnson JL, Ranganath H, Kuntimad G, Caulfield H (1998). Pulse-coupled neural networks. In: Omidvar O, Dayhoff JE, eds. Neural networks and pattern recognition. Academic Press, 1-56.

Johnson JL, Ritter D (1993). Observation of periodic waves in a pulse-coupled neural network. Opt Lett 18:1253-5.

Karim S, Tong G, Li J, Qadir A, Farooq U, Yu Y (2023). Current advances and future perspectives of image fusion: a comprehensive review. Inf Fusion 90:185-217.

Li H, Manjunath B, Mitra SK (1995). Multisensor image fusion using the wavelet transform. Graph Models Image Process 57:235-45.

Murtaza G, Shuib L, Abdul Wahab AW, Mujtaba G, Mujtaba G, Nweke HF, Al-garadi MA, Zulfiqar F, Raza G, Azmi NA (2020). Deep learning-based breast cancer classification through medical imaging modalities: state of the art and research challenges. Artif Intell Rev 53:1655-720.

Nobariyan B, Amini N, Daneshvar S, Abbasi A (2018). A novel architecture of medical image fusion based on YCbCr-DWT transform. Int Arab J Inf Technol 15:850-6.

Perez RC, Kim D, Maxwell AWP, Camacho JC (2023). Functional imaging of hypoxia: PET and MRI. Cancers 15:3336.

Preethi S, Aishwarya P (2021). An efficient wavelet-based image fusion for brain tumor detection and segmentation over PET and MRI image. Multimed Tools Appl 80:14789-806.

Rana DH, Degadwala SD (2014). Medical image fusion using combined multi-resolution and multi-scaling transform. Int J Comput Appl 107:26-9.

Ranganath H, Kuntimad G, Johnson JL (1995). Pulse coupled neural networks for image processing. In: Proc. IEEE Southeastcon '95. Visualize the Future, 1995 Mar 26-29; Raleigh, NC. Piscataway (NJ): IEEE, 37-43.

Sayadi M, Ghassemian H, Naimi R (2020). A new composite multimodality image fusion method based on shearlet transform and retina inspired model. In: Proc. 2020 Int Conf Machine Vision and Image Processing (MVIP), 2020 Feb 18-20; Qom, Iran. Piscataway (NJ): IEEE.

Singh P, Bhandari AK (2024). Laplacian and gaussian pyramid based multiscale fusion for nighttime image enhancement. Multimed Tools Appl:1-25.

Singh S, Singh H, Bueno G, Deniz O, Singh S, Monga H, Hrisheekesha PN, Pedraza A (2023). A review of image fusion: methods, applications and performance metrics. Digit Signal Process 137:104020.

Wang Z, Ma Y (2007). Dual-channel PCNN and its application in the field of image fusion. In: Proc. 3rd Int Conf Natural Computation (ICNC 2007), 2007 Aug 24-27; Haikou, China. Los Alamitos (CA): IEEE Computer Society, 755-59.