Search

Abstract Current research in nephrology is increasingly focused on elucidating the complexity inherent in tightly interwoven molecular systems and their correlation with pathology and related therapeutics, including dialysis and renal transplantation. Rapid advances in the omics sciences, medical device sensorization, and networked digital medical devices have made such research increasingly data centered. Data-centric science requires the support of computationally powerful and sophisticated tools able to handle the overflow of novel biomarkers and therapeutic targets. This is a context in which artificial intelligence (AI) and, more specifically, machine learning (ML) can provide a clear analytical advantage, given the rapid advances in their ability to harness multimodal data, from genomic information to signal, image and even heterogeneous electronic health records (EHR). However, paradoxically, only a small fraction of ML-based medical decision support systems undergo validation and demonstrate clinical usefulness. To effectively translate all this new knowledge into clinical practice, the development of clinically compliant support systems based on interpretable and explainable ML-based methods and clear analytical strategies for personalized medicine are imperative. Intelligent nephrology, that is, the design and development of AI-based strategies for a data-centric approach to nephrology, is just taking its first steps and is by no means yet close to its coming of age. These first steps are not even homogeneously taken, as a digital divide in access to technology has become evident between developed and developing countries, also affecting underrepresented minorities. With all this in mind, this editorial aim to provide a selective overview of the current use of AI technologies in nephrology and heralds the “Artificial Intelligence in Nephrology” special issue launched by BMC Nephrology.

IntroductionCardiac surgery-associated acute kidney injury (CSA-AKI) is a well-known complication that increases morbidity and mortality rates. The objective of this study was to reduce CSA-AKI through nephrologist intervention in patients awaiting cardiac surgery.MethodsWe performed a single center, open-label, randomized clinical trial including 380 patients who underwent scheduled cardiac surgery at the Hospital de Bellvitge between July 2015 and October 2019. A total of 184 patients were evaluated by the same Nephrologist one month before the surgery to minimize the risk factors for AKI. In addition to assessments at the outpatient clinic, we also collected clinical data during hospitalization and during the first year.ResultsDespite the intervention, no differences were observed between the groups in the incidence of CSA-AKI (intervention group 26.37% vs. standard of care 25.13%, p=0.874), mortality (3.91% vs. 3.59%, p=0.999), length of Intensive Care Unit (ICU) stay (10 days [7.00;15.0] for both groups, p=0.347), or renal function after one year of follow-up (estimated glomerular filtration rate (eGFR) by CKD-EPI: 74.5 ml/min (standard deviation 20.6) vs 76.7 (20.8) ml/min, respectively, p=0.364). A reduction in the need for blood transfusion was observed in the intervention group, although the difference was not statistically significant (37.22% vs. 45.03%, p =0.155).ConclusionIn this clinical trial, nephrologist intervention in the entire population on the cardiac surgery waiting list did not show a nephroprotective benefit.Clinical trial registrationClinicalTrials.gov, identifier (NCT02643745).

Artificial intelligence (AI) generative models driven by the integration of AI and natural language processing technologies, such as OpenAI’s chatbot generative pre-trained transformer large language model (LLM), are receiving much public attention and have the potential to transform personalized medicine. Dialysis patients are highly dependent on technology and their treatment generates a challenging large volume of data that has to be analyzed for knowledge extraction. We argue that, by integrating the data acquired from hemodialysis treatments with the powerful conversational capabilities of LLMs, nephrologists could personalize treatments adapted to patients’ lifestyles and preferences. We also argue that this new conversational AI integrated with a personalized patient-computer interface will enhance patients’ engagement and self-care by providing them with a more personalized experience. However, generative AI models require continuous and accurate updates of data, and expert supervision and must address potential biases and limitations. Dialysis patients can also benefit from other new emerging technologies such as Digital Twins with which patients’ care can also be addressed from a personalized medicine perspective. In this paper, we will revise LLMs potential strengths in terms of their contribution to personalized medicine, and, in particular, their potential impact, and limitations in nephrology. Nephrologists’ collaboration with AI academia and companies, to develop algorithms and models that are more transparent, understandable, and trustworthy, will be crucial for the next generation of dialysis patients. The combination of technology, patient-specific data, and AI should contribute to create a more personalized and interactive dialysis process, improving patients’ quality of life.