Clinical applications of AI in Renal Diseases
Artificial intelligence (AI) has been applied widely in almost every area of our daily lives, due to the growth of computing power, advances in methods and techniques, and the explosion of data, it also plays a critical role in academic disciplines, medicine is not an exception. AI can augment the intelligence of clinicians in diagnosis, prognosis, and treatment decisions.
Kidney disease causes great economic burden worldwide, with both acute kidney injury and chronic kidney disease bringing about high morbidity and mortality. Outstanding challenges in nephrology may be addressed by leveraging big data and AI.
In this review, Lijing Yao and colleagues summarized advances in machine learning (ML), artificial neural network (ANN), convolution neural network (CNN) and deep learning (DL), with a special focus on acute kidney injury (AKI), chronic kidney disease (CKD), end-stage renal disease (ESRD), dialysis, kidney transplantation and nephropathology.
First, the applications and studies of AI in nephropathies with high-morbidity and mortality, such as DKD, IgAN, ESRD are of high priority. Moreover, in coming years, AI is most likely to achieve the most success in nephropathology. Renal biopsy images will undoubtedly be valuable for use in the application of AI in kidney diseases. With the development of EMRs, the resulting large dialysis datasets will be ideal for AI research. Interdisciplinary integration is necessary and of great importance.
However, there are some issues concerning data quality and quantity, patient privacy, information safety, medical responsibility, AI-clinician relationships and physician-patient relationships. Thus, ethical regulations and regulations on AI in healthcare should be established to overcome these challenges. Overall, AI will be an excellent assistant in clinical and research settings to improve patients’ outcomes.
AI may not be anticipated to replace the nephrologists’ medical decision-making for now, but instead assisting them in providing optimal personalized therapy for patients.