BMC Pulmonary Medicine is calling for submissions to our Collection on Artificial intelligence and machine learning: applications in pulmonary medicine. Advancements in artificial intelligence (AI) and machine learning (ML) have the potential to revolutionize pulmonary medicine by enabling innovative approaches to diagnosis, treatment, and prevention of pulmonary disorders. In this era of rapid technological advancements, AI can assist in early detection, risk assessment, and prognostic evaluation by analyzing large datasets, thus leading to improved patient outcomes and better management strategies.
BMC Pulmonary Medicine is launching this collection in alignment with the United Nations' Sustainable Development Goals (SDGs) 3: Good health and well-being and SDG 10: Reduced inequalities. The aim of this collection is to consolidate both fundamental and clinical research to advance our comprehension of pulmonary disorders.
BMC Pulmonary Medicine welcomes original research on the design, implementation, optimization, and clinical impact of AI applications in the field of pulmonary medicine. Topics of interest include, but are not limited to, the following:
• Machine learning (ML) algorithms for early detection of pulmonary diseases
• AI applications for diagnostic accuracy studies
• AI systems as an intervention in live clinical settings
• Predictive modeling using AI for personalized risk assessment of pulmonary disorders
• Application of AI in pulmonary imaging analysis
• Utilizing natural language processing and AI for analyzing electronic health records in pulmonary care
• Exploring the potential of AI in optimizing pulmonary surgical procedures
• Wearable devices and AI algorithms for continuous monitoring of pulmonary health
• AI-enabled precision medicine approaches for personalized treatment
• AI-powered automated risk scoring systems for exacerbations of pulmonary diseases
• Ethical considerations and challenges in the implementation of AI in pulmonary medicine
We encourage the use of standardized reporting guidelines for research with AI/ ML components to encourage authors to provide information to allow their work to be evaluated appropriately. Reporting guidelines and checklists have been developed for a broad range of study design and research types with AI/ML components. Those that have been developed, adapted, or are planned to be adapted for research using AI/ML can be found summarized in the table below:
Reporting guideline | AI-guideline | Study design | AI- guideline description |
SPIRIT, 2013 | SPIRIT-AI, 2020 | Randomized controlled trials (protocols) | Used to report the protocols of randomized controlled trials evaluating AI systems as interventions. |
CONSORT, 2010 | CONSORT-AI, 2020 | Randomized controlled trials | Used to report randomized controlled trials evaluating AI systems as interventions (large-scale, summative evaluation), independently of the AI system modality (diagnostic, prognostic, therapeutic). Focuses on effectiveness and safety. |
TRIPOD, 2015 | TRIPOD-AI, 2024 | Prediction model evaluation | Used to report prediction models (diagnostic or prognostic) development, validation and updates. |
STARD, 2015 | STARD-AI | Diagnostic accuracy studies | Used to report diagnostic accuracy studies, either at development stage or as an offline validation in clinical settings. |
N/A | CLAIM , 2020 | Diagnostic accuracy studies | Used to report a wide spectrum of AI applications using medical images. Contains elements of the STARD 2015 guideline. Lists information such as descriptions of ground truth, data partitions, model description, and training and evaluation steps. |
N/A | DECIDE-AI, 2022 | Various (e.g. prospective cohort studies and non-randomized controlled trials) with additional features, such as modification of intervention, analysis of pre-specified subgroups or learning curve analysis. | Used to report the early evaluation of AI systems as an intervention in live clinical settings (small-scale, formative evaluation), independently of the study design and AI system modality (diagnostic, prognostic, therapeutic). Focuses on clinical utility, safety and human factors. |
Image credit: © Shuo / Stock.adobe.com