EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

An Adaptive Preprocessing Pipeline for the Enhancement of Camouflaged Lung Tumor Detection in Whole-Body PET-CT Images

N. Shruthi, N. Manju

Engineering, Technology & Applied Science Research · 04.04.2026 · 16 (2) · 34189-34196

DOI: 10.48084/etasr.16854 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Engineering, Technology & Applied Science Research, online erschienen am 04.04.2026. Verfasst von N. Shruthi und 1 weiteren Autor:innen. Die Arbeit nennt 32 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Lung cancer is the leading cause of cancer-related mortality worldwide; therefore, its early detection and accurate delineation are important for effective clinical management. A very helpful tool for the diagnosis of lung cancer and its staging is whole-body Positron Emission Tomography Computed Tomography (PET-CT); however, the interpretation of these images is often challenging due to low Signal to Noise Ratios (SNRs), limited contrast, and physiological uptake patterns. As a result, small or low-uptake tumors may become camouflaged within surrounding tissues, potentially delaying diagnosis and increasing uncertainty in clinical decision-making. This study proposes an adaptive image preprocessing framework for whole-body PET-CT to enhance the visibility of camouflaged lung tumors and improve overall image quality for downstream analysis. This framework consists of three self-tuning modules: i) the Adaptive Non-Local Means Denoising (ANLMD), which suppresses noise while preserving structural edges; ii) the Camouflage Aware Contrast Enhancement (CACE), which selectively amplifies low-contrast tumor regions; and iii) the Selective Background Attenuation (SBA), which suppresses physiologically high uptake in non-tumor organs using CT-guided anatomical masking. Contrary to deep-learning-based preprocessing approaches, the proposed approach is fully interpretable, modular, and does not require training or retraining across scanners. Quantitative evaluation revealed significant enhancements in image quality, including a Peak Signal to Noise Ratio (PSNR) of 37.26 dB, a Structural Similarity Index Measure (SSIM) of 0.956, and a Contrast to Noise Ratio (CNR) of 4.55, validating that the proposed methodology produces PET-CT images assisting radiologists for much more accurate interpretation by reducing the camouflage effects and enhancing diagnostic clarity.

Zitieren

APA

Shruthi, N., & Manju, N. (2026). An Adaptive Preprocessing Pipeline for the Enhancement of Camouflaged Lung Tumor Detection in Whole-Body PET-CT Images. Engineering, Technology & Applied Science Research, 16(2), 34189-34196. https://doi.org/10.48084/etasr.16854

Vancouver

Shruthi N, Manju N. An Adaptive Preprocessing Pipeline for the Enhancement of Camouflaged Lung Tumor Detection in Whole-Body PET-CT Images. Engineering, Technology & Applied Science Research. 2026;16(2):34189-34196. doi:10.48084/etasr.16854

BibTeX
@article{Shruthi2026,
  author = {Shruthi, N. and Manju, N.},
  title = {An Adaptive Preprocessing Pipeline for the Enhancement of Camouflaged Lung Tumor Detection in Whole-Body PET-CT Images},
  journal = {Engineering, Technology & Applied Science Research},
  year = {2026},
  volume = {16},
  number = {2},
  pages = {34189--34196},
  doi = {10.48084/etasr.16854}
}
Kostenlos starten →