Artificial intelligence in lung cancer diagnostic imaging: a review of the reporting and conduct of research published 2018-2019.

Objective: This study aimed to describe the methodologies used to develop and evaluate models that use artificial intelligence (AI) to analyse lung images in order to detect, segment (outline borders of), or classify pulmonary nodules as benign or malignant. Methods: In October 2019, we systematically searched the literature for original studies published between 2018 and 2019 that described prediction models using AI to evaluate human pulmonary nodules on diagnostic chest images. Two evaluators independently extracted information from studies, such as study aims, sample size, AI type, patient characteristics, and performance. We summarised data descriptively. Results: The review included 153 studies: 136 (89%) development-only studies, 12 (8%) development and validation, and 5 (3%) validation-only. CT scans were the most common type of image type used (83%), often acquired from public databases (58%). Eight studies (5%) compared model outputs with biopsy results. 41 studies (26.8%) reported patient characteristics. The models were based on different units of analysis, such as patients, images, nodules, or image slices or patches. Conclusion: The methods used to develop and evaluate prediction models using AI to detect, segment, or classify pulmonary nodules in medical imaging vary, are poorly reported, and therefore difficult to evaluate. Transparent and complete reporting of methods, results and code would fill the gaps in information we observed in the study publications. Advances in knowledge: We reviewed the methodology of AI models detecting nodules on lung images and found that the models were poorly reported and had no description of patient characteristics, with just a few comparing models' outputs with biopsies results. When lung biopsy is not available, lung-RADS could help standardise the comparisons between the human radiologist and the machine. The field of radiology should not give up principles from the diagnostic accuracy studies, such as the choice for the correct ground truth, just because AI is used. Clear and complete reporting of the reference standard used would help radiologists trust in the performance that AI models claim to have. This review presents clear recommendations about the essential methodological aspects of diagnostic models that should be incorporated in studies using AI to help detect or segmentate lung nodules. The manuscript also reinforces the need for more complete and transparent reporting, which can be helped using the recommended reporting guidelines.

A review found inadequate reporting of case-control studies of risk factors for pancreatic cancer.

OBJECTIVES: Case-control studies are often used to identify the risk factors for pancreatic cancer. The objective of this study was to evaluate the reporting of case-control studies of the risk factors for pancreatic cancer using the Strengthening The Reporting of OBservational Studies in Epidemiology (STROBE) for case-control studies checklist. STUDY DESIGN AND SETTING: We conducted a comprehensive literature search of the MEDLINE and EMBASE databases to identify reports of case-control studies published between 2016 and 2018. We scored article reporting using a reporting adherence form developed from the STROBE checklist for case-control studies, consisting of 14 STROBE items related to the title, abstract, methods, and results sections. RESULTS: We included reports of 47 case-control studies investigating a variety of risk factors, such as medical conditions and lifestyle factors. Reporting was inconsistent and inadequate. Efforts to address bias and how the study size was arrived at were particularly poorly described. Study cases were described in more detail than study controls. CONCLUSION: Reporting of case-control studies remains inadequate more than 10 years after the STROBE reporting guideline was published. Our findings suggest that authors do not understand the extent to which study methods and findings should be reported to enable studies to be fully understood, and their methods reproduced.

Reporting guidelines for oncology research: helping to maximise the impact of your research.

Many reports of health research omit important information needed to assess their methodological robustness and clinical relevance. Without clear and complete reporting, it is not possible to identify flaws or biases, reproduce successful interventions, or use the findings in systematic reviews or meta-analyses. The EQUATOR Network (http://www.equator-network.org/) promotes responsible reporting and the use of reporting guidelines to improve the accuracy, completeness, and transparency of health research. EQUATOR supports researchers by providing online resources and training. EQUATOR Oncology, a project funded by Cancer Research UK, aims to support cancer researchers reporting their research through the provision of online resources. In this article, our objective is to highlight reporting issues related to oncology research publications and to introduce reporting guidelines that are designed to aid high-quality reporting. We describe generic reporting guidelines for the main study types, and explain how these guidelines should and should not be used. We also describe 37 oncology-specific reporting guidelines, covering different clinical areas (e.g., haematology or urology) and sections of the report (e.g., methods or study characteristics); most of these are little-used. We also provide some background information on EQUATOR Oncology, which focuses on addressing the reporting needs of the oncology research community.