Photo Credit: Tippapatt

Findings underscore the proposed approach’s potential for advancing diagnostic and prognostic lung cancer treatment planning and decision-making systems.

Combining the most relevant features from machine learning significantly improves the potential for lung cancer detection, according to data recently published in Heliyon. According to Liangyu Li and colleagues, machine learning offers significant potential for lung cancer detection and enables early diagnosis, potentially improving patient outcomes. However, feature extraction remains a crucial challenge in lung cancer detection. The study combined the Gray-level co-occurrence matrix (GLCM) with Haralick features and autoencoder techniques, which the study team analyzed using supervised machine learning methods. Support Vector Machine (SVM) Radial Base Function (RBF) and SVM Gaussian demonstrated flawless performance, with SVM polynomial achieving 99.89% accuracy using GLCM with autoencoder, Haralick, and autoencoder features. SVM Gaussian achieved 99.56% accuracy, and SVM RBF achieved 99.35% accuracy using GLCM with Haralick features. The researchers suggest that these findings underscore the proposed approach’s potential for advancing diagnostic and prognostic lung cancer treatment planning and decision-making systems.