Applications of Artificial Intelligence in Neonatology

Abstract

The development of artificial intelligence methods has impacted therapeutics, personalized diagnostics, drug discovery, and medical imaging. Although, in many situations, AI clinical decision-support tools may seem superior to rule-based tools, their use may result in additional challenges. Examples include the paucity of large datasets and the presence of unbalanced data (i.e., due to the low occurrence of adverse outcomes), as often seen in neonatal medicine. The most recent and impactful applications of AI in neonatal medicine are discussed in this review, highlighting future research directions relating to the neonatal population. Current AI applications tested in neonatology include tools for vital signs monitoring, disease prediction (respiratory distress syndrome, bronchopulmonary dysplasia, apnea of prematurity) and risk stratification (retinopathy of prematurity, intestinal perforation, jaundice), neurological diagnostic and prognostic support (electroencephalograms, sleep stage classification, neuroimaging), and novel image recognition technologies, which are particularly useful for prompt recognition of infections. To have these kinds of tools helping neonatologists in daily clinical practice could be something extremely revolutionary in the next future. On the other hand, it is important to recognize the limitations of AI to ensure the proper use of this technology.

Potential of use of modern information technology solutions in the work of hospital infection control team, including antibiotic therapy optimisation, reduction of alert pathogen infections and vaccination popularisation

“…Healthcare information technologies need to be chosen wisely [80], as miology leads to the conclusion that ai-based solutions -using machine learning (Ml)-based technologies that are capable of detecting patterns based on large amounts of data without being pre-programmed to do so [83] in order to be capable of linking, in a way available only for artificial neural networks (ann), a very wide variety of often incomplete input data to singular, concrete and practically applicable outputs, like diagnoses [99] -can be a valuable asset in supporting and monitoring the infection control guidelines implementation in healthcare facilities by the infection control teams (iCt). Medical expert systems and ai-based solutions can, as it has already been demonstrated in different medical disciplines [26][27][28], significantly improve and speed up consulting cases of people suspected of infection or infectious disease and those diagnosed with an infection or infectious disease [29,30]. agent-based systems (abS) and multi-agent systems (MaS) [33] can be of help in collecting the required large amounts of data and documenting the actions taken by the epidemiologists.…”

“…For over a decade now, artificial neural networks are being prepared to replace [24], or at least support, medical staff in the clinical diagnostic process [25], in fields ranging from microscopic cytology [26] and intestinal capsule image analysis [27] to coronary heart disease [28]. infectious diseases and epidemiology applications include tuberculosis diagnosis [29] and prompt recognition of infections in closely monitored patients [30].…”

Nutritional Strategies for Preterm Neonates and Preterm Neonates Undergoing Surgery: New Insights for Practice and Wrong Beliefs to Uproot

“…Artificial intelligence (AI) has the potential to significantly increase human understanding of illness and treatment effectiveness, even in neonatology [ 125 ].…”

“…AI can represent an early diagnostic tool for identifying neonates with IP. Machine learning (ML) models can be useful for meticulously assessing these difficult circumstances, starting with massive datasets of parameters and variables [ 125 , 128 ].…”

Editorial: Identification of novel biomarkers for retinopathy of prematurity in preterm infants by use of innovative technologies and artificial intelligence

“…Innovative technologies, including machine learning and artificial intelligence (AI) tools and imaging methods using deep learning (DL) algorithms, have been used to better define the fundus features of the International Classification of ROP (ICROP), which has been the gold standard ( 15 – 17 ). These methods have been proposed to reach the level of human decision-making and to detect retinal characteristics that would otherwise go unnoticed during clinical examination ( 18 – 20 ).…”