Lung cancer is a prevalent malignancy associated with a high mortality rate, with a 5-year relative survival rate of 23%. Traditional survival analysis methods, reliant on clinician judgment, may lack accuracy due to their subjective nature. Consequently, there is growing interest in leveraging AI-based systems for survival analysis using clinical data and medical imaging. The purpose of this study is to improve survival classification for lung cancer patients by utilizing a 3D-CNN architecture (ResNet-34) applied to CT images from the NSCLC-Radiomics dataset. Through comprehensive ablation studies, we evaluate the effectiveness of different features and methodologies in classification performance. Key contributions include the introduction of a novel feature (GTV1-SliceNum), the proposal of a novel loss function (PEN-BCE) accounting for false negatives and false positives, and the showcasing of their efficacy in classification. Experimental work demonstrates results surpassing those of the existing literature, achieving a classification accuracy of 0.7434 and an ROC-AUC of 0.7768. The conclusions of this research indicate that the AI-driven approach significantly improves survival prediction for lung cancer patients, highlighting its potential for enhancing personalized treatment strategies and prognostic modeling.

To improve battery performance and lifetime, this study highlights the value of prognostics and health management (PHM) in correctly predicting state of health (SoH). The Prognostics Center of Excellence Dataset Repository is used as a benchmark dataset in the study to compare and contrast different machine learning approaches for predicting SoH. The research compares how well different machine learning techniques, including Support Vector Regression, Extreme Gradient Boosting (XGBoost) and Random Forest, predict SoH. The study demonstrates the importance of PHM in improving battery reliability, efficiency, and provides insight into the effectiveness of different machine learning approaches in accurately predicting SoH.

Lung cancer is known as one of the most prevalent and dangerous types of cancer, and its survival rate is lower compared to many other types of cancer. Survival analyses are vital to accurately estimate the expected time to death of lung cancer patients. Since these analyses are generally based on the judgment of clinicians, accuracy is often a matter of debate. This highlights the importance of performing survival analysis using statistical or machine learning systems that use clinical data and/or images obtained from medical imaging devices. In this study, we applied various methods to analyze clinical records from a publicly available lung cancer database. We approached survival analysis as both a survival regression and a survival classification problem, conducting comparative analyses using multiple techniques. We conducted a detailed examination of numerous survival analysis studies related to lung cancer. Our findings revealed that incorporating the number of slices containing the tumor area in patients' CT images (GTV1-SliceNum) as an extra feature significantly enhanced survival analysis performance. Additionally, we observed that the presence of censored observations negatively impacted survival classification performance.