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.

To systematically review studies evaluating the influence of surgical experience on individual performance. Experience, measured in case volume or years of practice, is recognized as a key driver of individual surgical performance, giving rise to a learning curve. However, this topic has not been reviewed at the cross-specialty level. MEDLINE, EMBASE, PsycINFO, AMED, and the Cochrane Database of Systematic Reviews were searched (from inception to February 2013). Two reviewers independently reviewed citations using predetermined inclusion and exclusion criteria. Ninety-one data points per study were extracted. The search strategy yielded 6950 citations. Fifty-seven studies were eligible, including 1,061,913 cases and 35 procedure types, performed by 17,912 surgeons. Forty-five studies monitored case volume, and 6 studies measured experience as both case volume and years of practice. Of these 51 studies, 44 found that increased case volume was associated with significantly improved health outcomes. Several studies noted a plateau phase or maturation in the surgical learning curve. Acquisition of this phase was procedure specific and outcome specific, ranging from 25 to 750 procedures. Twelve studies assessed the impact of years of surgical practice, 11 of which found that increased years of experience was associated with significantly improved health outcomes. Two studies noted a plateau phase, where increases in years of experience were no longer associated with improvements in operative outcomes. Three studies identified performance deterioration after the plateau phase. Increasing surgical case volume and years of practice are associated with improved performance, in a procedure-specific manner. Performance may deteriorate toward the end of a surgeon's career.

ABO blood group incompatible renal transplantation, using desensitization procedures, is an effective strategy. Efforts have been made to reduce desensitization: these are usually applied to all patients indiscriminately. The Guy's Hospital ABO blood group incompatible desensitization regimen uses a tiered approach, tailoring strategy according to initial antibody titres. Sixty-two ABO blood group incompatible living donor transplant recipients were compared with 167 recipients of blood group compatible living donor renal transplants. There were no statistically significant differences in allograft survival rates at 1 or 3 years post-transplant, rejection in the first year post-transplant or renal function in the first 3 years post-transplant. There was a higher rate of death in ABO blood group incompatible transplant recipients - this could be associated with differences in age and HLA mismatch between the two groups. Four ABO blood group incompatible patients experienced antibody-mediated rejection (no episode was associated with a rise in ABO blood group antibodies). Of the patients who received no desensitization, or rituximab alone, none has experienced antibody mediated rejection or experienced allograft loss. Tailoring the use of desensitization in ABO blood group incompatible renal transplantation according to initial ABO blood group antibody titres led to comparable results to blood group compatible transplantation.