The increasing availability of multidimensional and annotated digital medical data (so-called big data) and the continuous increase of computational capabilities have in the last five years led to the development of a new generation of data-driven artificial intelligence systems (so-called machine learning or deep learning). These systems, applied to a spectrum of medical tasks, which range from diagnosis to monitoring, including prognosis and treatment support, have shown performances that the medical community has not hesitated to define as relevant (in some cases, such as in pattern matching, even superhuman) and capable of revolutionizing medical practice, as some technological innovations in the diagnostic field made 50 years ago (e.g., CT, MRI). For months, the scientific literature has hosted an exponential increase in the number of studies that report the classification and predictive performance of machine/deep learning systems in almost every medical field, including spine surgery, to automate, or more often to augment, a wide range of business cases: these include the identification of fractures and lesions, the characterization and staging of health conditions, risk and prognostic stratification associated with different types of surgery and surgical planning.
We have therefore reached a point where it is possible to ask: can these systems actually revolutionize medical practice in spine surgery and improve the outcome of treatments in this discipline which is still characterized by wide margins of standardization of care, reduction of post-surgical adverse events and improvement in outcomes?
To answer this question, we share this call for papers for the Special Issue 'Clinical Integration of Artificial Intelligence in Spine Surgery' to collect a series of papers that show how machine/deep learning can have an impact on the clinical practice in the field of Spine Surgery. For this Special Issue, we invite the interested authors to contribute to this special issue, which, in the now broad landscape of similar initiatives, has the ambitious goal of being the first one to primarily accept and publish works in which authors can provide evidence about at least one of the following quality dimensions:
(1) The reliability of the reference ground truth, for example, by applying a suitable reliability metric to the annotations of a panel of experts;
(2) The robustness of the system in out-of-distribution or multicenter external validation contexts;
(3) The measured utility of the system in the field of application.
The contributions received will be evaluated in the light of the Checklist for assessment of medical AI (ChAMAI), to ensure the highest methodological quality and reproducibility in the specialist landscape.
The increasing availability of multidimensional and annotated digital medical data (so-called big data) and the continuous increase of computational capabilities have in the last five years led to the development of a new generation of data-driven artificial intelligence systems (so-called machine learning or deep learning). These systems, applied to a spectrum of medical tasks, which range from diagnosis to monitoring, including prognosis and treatment support, have shown performances that the medical community has not hesitated to define as relevant (in some cases, such as in pattern matching, even superhuman) and capable of revolutionizing medical practice, as some technological innovations in the diagnostic field made 50 years ago (e.g., CT, MRI). For months, the scientific literature has hosted an exponential increase in the number of studies that report the classification and predictive performance of machine/deep learning systems in almost every medical field, including spine surgery, to automate, or more often to augment, a wide range of business cases: these include the identification of fractures and lesions, the characterization and staging of health conditions, risk and prognostic stratification associated with different types of surgery and surgical planning.
We have therefore reached a point where it is possible to ask: can these systems actually revolutionize medical practice in spine surgery and improve the outcome of treatments in this discipline which is still characterized by wide margins of standardization of care, reduction of post-surgical adverse events and improvement in outcomes?
To answer this question, we share this call for papers for the Special Issue 'Clinical Integration of Artificial Intelligence in Spine Surgery' to collect a series of papers that show how machine/deep learning can have an impact on the clinical practice in the field of Spine Surgery. For this Special Issue, we invite the interested authors to contribute to this special issue, which, in the now broad landscape of similar initiatives, has the ambitious goal of being the first one to primarily accept and publish works in which authors can provide evidence about at least one of the following quality dimensions:
(1) The reliability of the reference ground truth, for example, by applying a suitable reliability metric to the annotations of a panel of experts;
(2) The robustness of the system in out-of-distribution or multicenter external validation contexts;
(3) The measured utility of the system in the field of application.
The contributions received will be evaluated in the light of the Checklist for assessment of medical AI (ChAMAI), to ensure the highest methodological quality and reproducibility in the specialist landscape.
