No abstract

Substantial interest and investment in clinical artificial intelligence (AI) research has not resulted in widespread translation to deployed AI solutions. Current attention has focused on bias and explainability in AI algorithm development, external validity and model generalisability, and lack of equity and representation in existing data. While of great importance, these considerations also reflect a model-centric approach seen in published clinical AI research, which focuses on optimising architecture and performance of an AI model on best available datasets. However, even robustly built models using state-of-the-art algorithms may fail once tested in realistic environments due to unpredictability of real-world conditions, out-of-dataset scenarios, characteristics of deployment infrastructure, and lack of added value to clinical workflows relative to cost and potential clinical risks. In this perspective, we define a vertically integrated approach to AI development that incorporates early, cross-disciplinary, consideration of impact evaluation, data lifecycles, and AI production, and explore its implementation in two contrasting AI development pipelines: a scalable “AI factory” (Mayo Clinic, Rochester, United States), and an end-to-end cervical cancer screening platform for resource poor settings (Paps AI, Mbarara, Uganda). We provide practical recommendations for implementers, and discuss future challenges and novel approaches (including a decentralised federated architecture being developed in the NHS (AI4VBH, London, UK)). Growth in global clinical AI research continues unabated, and introduction of vertically integrated teams and development practices can increase the translational potential of future clinical AI projects.

The statistics are not encouraging: 1 out of every 4 individuals worldwide will experience a mental or neurological disorder at some point in their lifetime (1). In 2017 alone, approximately 17.3 million U.S. adults experienced at least one major depressive episode (2). Globally, more than 300 million persons suffer from depression (3). Addressing the human suffering that these statistics represent requires a professional workforce that does not currently exist. There are too few mental health professionals to meet the needs of this patient population. By one estimate, 35.5% to 50.3% of patients with serious mental disorders in developed countries did not receive treatment within the previous 12 months. Among patients in less developed countries, 76.3% to 85.4% received no treatment (4). New solutions that can increase access to mental health care and close the treatment gap are sorely needed. To fill this gap, many clinicians, patients, and consumers are now looking to digital solutions, including mobile apps and chatbots. While the potential is high, current research on the effectiveness of these digital tools, however, is mixed. This paper offers a selective review of the digital mental health landscape with the goal of informing patients and clinicians about the best available evidence.

We are at a pivotal moment in the development of healthcare artificial intelligence (AI), a point at which enthusiasm for machine learning has not caught up with the scientific evidence to support the equity and accuracy of diagnostic and therapeutic algorithms. This proposal examines algorithmic biases, including those related to race, gender and socioeconomic status, and accuracy, including the paucity of prospective studies and lack of multisite validation. We then suggest solutions to these problems. We describe the Mayo Clinic, Duke University, Change Healthcare project that is evaluating 35.1 billion healthcare records for bias. And we propose ‘Ingredients’ style labels and an AI evaluation/testing system to help clinicians judge the merits of products and services that include algorithms. Said testing would include input data sources and types, dataset population composition, algorithm validation techniques, bias assessment evaluation and performance metrics.