RESUMO
Although still in its infancy, artificial intelligence (AI) analysis of kidney biopsy images is anticipated to become an integral aspect of renal histopathology. As these systems are developed, the focus will understandably be on developing ever more accurate models, but successful translation to the clinic will also depend upon other characteristics of the system.In the extreme, deployment of highly performant but "black box" AI is fraught with risk, and high-profile errors could damage future trust in the technology. Furthermore, a major factor determining whether new systems are adopted in clinical settings is whether they are "trusted" by clinicians. Key to unlocking trust will be designing platforms optimized for intuitive human-AI interactions and ensuring that, where judgment is required to resolve ambiguous areas of assessment, the workings of the AI image classifier are understandable to the human observer. Therefore, determining the optimal design for AI systems depends on factors beyond performance, with considerations of goals, interpretability, and safety constraining many design and engineering choices.In this article, we explore challenges that arise in the application of AI to renal histopathology, and consider areas where choices around model architecture, training strategy, and workflow design may be influenced by factors beyond the final performance metrics of the system.
Assuntos
Inteligência Artificial , Confiança , Humanos , RimRESUMO
Quality assessment in kidney transplantation involves inspection to identify negative markers of organ quality. However, there is a paucity of evidence guiding surgical appraisal, and currently there is no evidence to differentiate important features from those that can be safely ignored. We propose a method to standardize surgical assessment and derived a simple rule to rapidly identify kidneys suitable for transplantation. Donor and recipient data were recorded alongside clinical outcomes in a prospectively maintained database. We developed a proforma (Cambridge Kidney Assessment Tool, CKAT) and used it to assess deceased donor kidney transplants. Factors predictive of utilization were identified by multivariate and univariate logistic regression analysis of CKAT-assessment scores, and test performance was evaluated using standard 2 × 2 contingency tables. Ninety-seven kidneys were included at a single center (2013-2014), and 184 CKAT assessments were performed. A CKAT threshold of "Carrell + Perfusion >3" was highly specific (99%) and performed favorably to consultant opinion (specificity 95%). 96% of the kidneys implanted in accordance with the rule survived to 1 year (mean eGFR 45.3 mL/min/1.73 m2 ). To our knowledge, this is the first attempt to objectively define macroscopic features that are relevant to kidney utilization. Common language could support training in organ assessment and ultimately help address unnecessary discard of donor kidneys.
