Artificial intelligence versus clinicians: systematic review of design, reporting standards, and claims of deep learning studies

Abstract

Objective To systematically examine the design, reporting standards, risk of bias, and claims of studies comparing the performance of diagnostic deep learning algorithms for medical imaging with that of expert clinicians. Design Systematic review. Data sources Medline, Embase, Cochrane Central Register of Controlled Trials, and the World Health Organization trial registry from 2010 to June 2019. Eligibility criteria for selecting studies Randomised trial registrations and non-randomised studies comparing the performance of a deep learning algorithm in medical imaging with a contemporary group of one or more expert clinicians. Medical imaging has seen a growing interest in deep learning research. The main distinguishing feature of convolutional neural networks (CNNs) in deep learning is that when CNNs are fed with raw data, they develop their own representations needed for pattern recognition. The algorithm learns for itself the features of an image that are important for classification rather than being told by humans which features to use. The selected studies aimed to use medical imaging for predicting absolute risk of existing disease or classification into diagnostic groups (eg, disease or non-disease). For example, raw chest radiographs tagged with a label such as pneumothorax or no pneumothorax and the CNN learning which pixel patterns suggest pneumothorax. Review methods Adherence to reporting standards was assessed by using CONSORT (consolidated standards of reporting trials) for randomised studies and TRIPOD (transparent reporting of a multivariable prediction model for individual prognosis or diagnosis) for non-randomised studies. Risk of bias was assessed by using the Cochrane risk of bias tool for randomised studies and PROBAST (prediction model risk of bias assessment tool) for non-randomised studies. Results Only 10 records were found for deep learning randomised clinical trials, two of which have been published (with low risk of bias, except for lack of blinding, and high adherence to reporting standards) and eight are ongoing. Of 81 non-randomised clinical trials identified, only nine were prospective and just six were tested in a real world clinical setting. The median number of experts in the comparator group was only four (interquartile range 2-9). Full access to all datasets and code was severely limited (unavailable in 95% and 93% of studies, respectively). The overall risk of bias was high in 58 of 81 studies and adherence to reporting standards was suboptimal (<50% adherence for 12 of 29 TRIPOD items). 61 of 81 studies stated in their abstract that performance of artificial intelligence was at least comparable to (or better than) that of clinicians. Only 31 of 81 studies (38%) stated that further prospective studies or trials were required. Conclusions Few prospective deep learning studies and randomised trials exist in medical imaging. Most non-randomised trials are not prospective, are at high risk of bias, and deviate from existing reporting standards. Data and code availability are lacking in most studies, and human comparator groups are often small. Future studies should diminish risk of bias, enhance real world clinical relevance, improve reporting and transparency, and appropriately temper conclusions. Study registration PROSPERO CRD42019123605.

Reporting guidelines for clinical trial reports for interventions involving artificial intelligence: the CONSORT-AI Extension

“…It has been recognised that most recent AI studies are inadequately reported and existing reporting guidelines do not fully cover potential sources of bias specific to AI systems 25. The welcome emergence of randomised controlled trials (RCTs) seeking to evaluate newer interventions based on, or including, an AI component (hereafter “AI interventions”)23262728293031 has similarly been met with concerns about the design and reporting 25323334. This has highlighted the need to provide reporting guidance that is “fit-for-purpose” in this domain.…”

Prognostic Models for Knee Osteoarthritis: A Protocol for Systematic Review, Critical Appraisal and Meta-Analysis

“…We will develop a data extraction instrument for study data based on several previous systematic reviews of prediction model [51][52][53]. As the reviewers have different levels of experience and knowledge, the items listed will be reviewed and discussed to ensure that all reviewers had clear knowledge of the procedures.…”

“…However, they include many items that may not available for prediction models built with traditional statistical methods based on clinical characteristics, laboratory examinations, or genetic factors. On the other hand, TROPOD, CHARMS and PROBAST have been already proved suitable for assessing prediction models using arti cial intelligence methods [53]. Thus, we will choose three more widely-adapted and more extensively-accepted tools, to develop our critical appraisal instrument.…”

Guidelines for clinical trial protocols for interventions involving artificial intelligence: the SPIRIT-AI Extension

“…It has been recognised that most recent AI studies are inadequately reported and existing reporting guidelines do not fully cover potential sources of bias specific to AI systems 17. The welcome emergence of randomised controlled trials (RCTs) seeking to evaluate clinical efficacy of newer interventions based on, or including, an AI component (hereafter ‘AI interventions’) 15181920212223 has similarly been met with concerns about design and reporting 17242526. This has highlighted the need to provide reporting guidance that is ‘fit-for-purpose’ in this domain.…”