Lung cancer is the leading cause of cancer-related mortality globally. Although low-dose computed tomography (LDCT) has shown its value in reducing lung cancer mortality through early detection, the clinical implementation of LDCT screening still faces multiple challenges. These include high false-positive rates, limited access to expert radiological interpretation, and suboptimal risk stratification strategies. In parallel, treatment decision-making in lung cancer has become increasingly complex, with a range of treatment options, such as surgery, chemotherapy, radiotherapy, targeted therapy, and immunotherapy, requiring individualized selection based on patient-specific characteristics.
Artificial intelligence (AI), particularly machine learning, deep learning, large-scale models, and radiomics, offers transformative potential across the continuum of lung cancer care. In the screening setting, AI models can enhance the detection, classification, and risk assessment of pulmonary nodules, thereby improving diagnostic accuracy and reducing unnecessary interventions. In the treatment domain, AI can facilitate the integration of radiological, clinical, pathological, and molecular data to predict treatment response, forecast survival outcomes, and identify patients likely to benefit from specific therapeutic strategies. The convergence of AI with real-world clinical practice holds promise for improving decision-making, enabling timely interventions, and optimizing personalized treatment planning for patients with lung cancer.
This Research Topic aims to highlight innovative studies that apply AI to improve lung cancer screening and treatment. The goal is to foster the development of clinically applicable, interpretable, and robust AI models that can support early diagnosis, guide therapeutic decisions, and enhance outcome prediction. By connecting technological advancement with clinical relevance, this collection seeks to accelerate the translation of AI into meaningful improvements in patient care.
This Research Topic welcomes high-quality contributions focusing on, but not limited to, the following areas:
• AI-driven detection, segmentation, and risk assessment of pulmonary nodules
• Predictive modeling for treatment response, resistance, and toxicity across neoadjuvant, adjuvant, and definitive therapies
• Radiomics and deep learning approaches for non-invasive tumor characterization and biomarker discovery
• Multimodal AI techniques and novel architectures for image analysis
• Applications of large language and vision-language models in lung cancer imaging and decision support
• Clinical validation, robustness, interpretability, and ethical evaluation of AI models in lung cancer care
• AI-enabled epidemiological analysis to inform screening strategies and healthcare planning
Please note: Manuscripts consisting solely of bioinformatics, computational analysis, or predictions of public databases which are not accompanied by validation (independent clinical or patient cohort, or biological validation in vitro or in vivo, which are not based on public databases) are not suitable for publication in this journal.
Lung cancer is the leading cause of cancer-related mortality globally. Although low-dose computed tomography (LDCT) has shown its value in reducing lung cancer mortality through early detection, the clinical implementation of LDCT screening still faces multiple challenges. These include high false-positive rates, limited access to expert radiological interpretation, and suboptimal risk stratification strategies. In parallel, treatment decision-making in lung cancer has become increasingly complex, with a range of treatment options, such as surgery, chemotherapy, radiotherapy, targeted therapy, and immunotherapy, requiring individualized selection based on patient-specific characteristics.
Artificial intelligence (AI), particularly machine learning, deep learning, large-scale models, and radiomics, offers transformative potential across the continuum of lung cancer care. In the screening setting, AI models can enhance the detection, classification, and risk assessment of pulmonary nodules, thereby improving diagnostic accuracy and reducing unnecessary interventions. In the treatment domain, AI can facilitate the integration of radiological, clinical, pathological, and molecular data to predict treatment response, forecast survival outcomes, and identify patients likely to benefit from specific therapeutic strategies. The convergence of AI with real-world clinical practice holds promise for improving decision-making, enabling timely interventions, and optimizing personalized treatment planning for patients with lung cancer.
This Research Topic aims to highlight innovative studies that apply AI to improve lung cancer screening and treatment. The goal is to foster the development of clinically applicable, interpretable, and robust AI models that can support early diagnosis, guide therapeutic decisions, and enhance outcome prediction. By connecting technological advancement with clinical relevance, this collection seeks to accelerate the translation of AI into meaningful improvements in patient care.
This Research Topic welcomes high-quality contributions focusing on, but not limited to, the following areas:
• AI-driven detection, segmentation, and risk assessment of pulmonary nodules
• Predictive modeling for treatment response, resistance, and toxicity across neoadjuvant, adjuvant, and definitive therapies
• Radiomics and deep learning approaches for non-invasive tumor characterization and biomarker discovery
• Multimodal AI techniques and novel architectures for image analysis
• Applications of large language and vision-language models in lung cancer imaging and decision support
• Clinical validation, robustness, interpretability, and ethical evaluation of AI models in lung cancer care
• AI-enabled epidemiological analysis to inform screening strategies and healthcare planning
Please note: Manuscripts consisting solely of bioinformatics, computational analysis, or predictions of public databases which are not accompanied by validation (independent clinical or patient cohort, or biological validation in vitro or in vivo, which are not based on public databases) are not suitable for publication in this journal.