enhancing lung cancer diagnosis accuracy through autoencoder-based reconstruction of computed tomography (ct) lung images

Lung cancer is a major global cause of cancer-related deaths, emphasizing the importance of early detection through chest imaging. accurate reconstruction of computed tomography (ct) lung images plays a crucial role in the diagnosis and treatment planning of lung cancer patients. however, noise present in ct images poses a significant challenge, hindering the precise interpretation of internal tissue structures. low-dose ct, with reduced radiation risks compared to conventional-dose ct, has gained popularity. nonetheless, the noise inherent in these images compromises their quality, potentially impacting diagnostic performance. denoising autoencoder models, and unsupervised deep learning algorithms, offer a promising solution. by reconstructing clean inputs from corrupted ones, the hidden layers of the autoencoder capture robust features. in this study, a dataset of ct images from patients suspected of lung cancer was categorized into four disease groups, aiming to evaluate and compare different autoencoder models in terms of noise reduction and other evaluation criteria. the results demonstrated that all the designed autoencoder models effectively reduced noise in ct images, improving overall image quality. notably, semi-supervised autoencoder models exhibited superior performance, preserving fine details and enhancing diagnostic information. this research highlights the potential of autoencoder models in improving the accuracy of lung cancer diagnosis by reconstructing ct lung images, emphasizing the importance of noise reduction techniques in enhancing image quality and diagnostic performance, with the semi-supervised approach showing particular promise in preserving critical details.