Segmented Lung CT Image Refinement Through Hybrid Edge-Preserving De-Noising and Variance-Regulated Adaptive Histogram Equalization

Lung diseases are detected effectively by Chest CT Medical im-ages. Chest CT images are 3D, and they are used for detecting abnormalities in the lungs like pneumonia, tuberculosis, tumors, etc. Chest CT images suffer from different types of noise. The presence of various types of noise, such as Gaussian noise, Quantum noise, and salt-and-pepper noise, will reduce the quali-ty of the image. The noise that is present in CT images may not affect radiologist to diagnose the disease, but it will definitely af-fect AI-based systems in predicting the disease. Noise occurs be-cause of reduced radiation, insufficient photon information, er-rors during image reconstruction, and fluctuations in imaging system components. In this work, to improve the quality of Chest CT images, two algorithms are proposed. The first algorithm, Adaptive Edge Preserving Hybrid De-Noising Frame-work (AEPH) for Segmented Lung CT Images, is used to detect and reduce Noise. The second algorithm, Variance Regulated Adaptive Histogram Equalization (VRAHE), is proposed for improving de-noising and contrast enhancement performance in segmented lung CT images. At first, lung regions are segmented from the LIDC-IDRI dataset. The segmented Chest CT images are converted to 2D CT slices. These 2D CT slices are binary lung-segmented images. These are converted to lung mask im-ages. Random noise models are added only in the lung regions to preserve the background information without modification. The proposed work, AEPH, is applied to reduce noise. The main goal of AEPH is to reduce noise without modifying structural details and edge-related details. Experiments are conducted on a work-station computer with NVIDIA RTX GPU support. When com-pared to proposed AEPH to conventional approaches in remov-ing noise from CT images, it is identified that AEPH is more ef-fective than conventional methods. The metrics used for evaluat-ing the proposed noise removal approach AEPH when com-pared to existing approaches are PSNR, SSIM, EPI, RMSE, and entropy, and the values achieved are 32.96 dB, 0.934, 0.93, 9.84, and 7.12, respectively. After de-noising, the second proposed al-gorithm, VRAHE, successfully improved the image quality by enhancing contrast distribution and preserving important local information. The improved performance was evaluated using CNR, AMBE, UIQI, and EME metrics. From the experimental results, the values achieved are 5.96, 3.14, 0.972, and 31.48, re-spectively. The Results proved that the proposed work, AEPH, reduced noise, and VRAHE improved contrast levels in seg-mented lung CT images. The final chest CT images resulting from applying the proposed algorithms maintain brightness and also show the regional lung details more clearly.

@article{Bhuvaneshwari2026,
  title     = {Segmented Lung CT Image Refinement Through Hybrid Edge-Preserving De-Noising and Variance-Regulated Adaptive Histogram Equalization},
  journal   = {International Journal of Advanced Computer Science and Applications},
  volume    = {17},
  number    = {7},
  year      = {2026},
  publisher = {The Science and Information Organization},
  author    = {T. Bhuvaneshwari and Rohini Pinapatruni},
  doi       = {10.14569/IJACSA.2026.0170745},
  url       = {https://doi.org/10.14569/IJACSA.2026.0170745}
}