The use of AI, and in particular ML, is expanding in every area, including healthcare. Many results are now available about the excellent predictive capabilities of these new tools in medicine. Thanks to their use, intelligent tools useful to support healthcare professionals in daily practice will be increasingly available. Their possible exploitation in the health organization is no exception. Liu et al. showed as ML is superior to logistic regression for risk estimation, in the context of hospital performance assessment. Furthermore, similar applications can be found in the context of forecasting healthcare costs, risk of readmission, and hospitalization.[14–17]
The study by Luo et al [18] is also very interesting; ML models are applied with the intent of estimating the risk of cancellation of an operating session, with the negative impact that this entails both in terms of costs and on waiting lists and therefore also translates into delayed access of the patient to surgery [17, 18] Moreover, a novel approach has been presented by Abbou et al.[35]. The authors used data from the electronic hospital register (EHR) from December 2009 to May 2020 for a total of 297,480 interventions from two public hospitals in Israel in this study. They use pre-operative data to predict the duration of the surgery, including: patient clinical data, experience of surgeons, patient nationality, results of analyzes carried out before the operation, etc. They compare the predictions between a naive model and a ML model (Xgboost), comparing various metrics. The Authors deduced that the use of Big Data can certainly be useful for predicting the duration of interventions in the operating room and that ML model performs better than the their naive model.
However, in order to be able to apply these models, it is essential to obtain data with quality and precision. Currently, recording times and patient movements within the surgical block are often made manually by various operators involved and subsequently reported in computer systems. However, this detection method is often partial and most not in real-time. Having the possibility of a direct recording, with the minimum human interference, could allow instead to increase the quality of the dataset and therefore obtain more precise results.[19–21] Furthermore, having a system equipped with the ability to independently record the patient's movements, could be able, in addition to reducing the error rate, to lighten the workload of the operators themselves and indirectly reduce the changeover times between the different patients. [17,21−25] Nevertheless, building a tracking system inside a hospital is not a simple task. Not all hospitals are equipped with the highest technology available. Some are small in size and therefore do not have sufficient resources available to make organizational systems high-tech. Others, despite being larger in size, have old structures in which it is not always easy to implement new technologies. Therefore, we decided to look for a solution that suits our needs, and that could be economical, did not require an expensive infrastructure to rely on, and that could be achievable in most situations.
To understand which technologies were the best for our use case, we decided to analyze several options and constraints. The main constraints encountered were:
Considering these major constraints, we analyzed and considered several technologies. The first one we analyzed was RFID, unfortunately, this technology provided for cumbersome structures for tracking and the range of action was limited, which could have brought discomfort to the members of the medical staff. Moreover, the RFID devices' battery life was not enough for our case study. After that, we analyzed UWB, discarded due to the low autonomy that the devices reach. Finally, we decided to adopt BLE tracking devices.
BLE fitted our use case. The detectors are small and easy to install, the devices’ range of action is very wide and the tracking devices’ battery life can last even a few months. Furthermore, our use case does not involve tracking of healthcare personnel, but only patients.
Building a tracking system using BLE also has economic advantages; it does not burden the hospital budgets and is cost-effective. For all the reasons mentioned above, it was decided to use this technology excluding others. The architecture presented in this way can trace the movements of patients within the OR. Data obtained will form a dataset that can be used to perform ML tasks. By combining the tracking data with those of clinical assessment, it will be possible to create an algorithm capable of predicting the duration of a specific surgical intervention.
The IoT architecture developed in this study will be useful to collect real data about the patients flow in the surgical department. But having solid data and predictions about OR occupancy is still not enough. It is then necessary to create organizational systems capable of using that information. Not being able to transform these results into intelligent tools that can be used in daily practice has been recognized in fact as one of the major limitations that currently hinder the use of these technologies in medicine[26]. To overcome these limits, a simulation model will be developed for reproducing the patients’ flow. This model, feed with high quality data, will be used to determine the best scheduling logic, in order to optimize the usage of resources in the ORs. A preliminary testing phase was performed using randomly generated patients’ data, in the attempt to check the model effectiveness in reproducing the system, as well as to embody alternative scheduling logics and performance indexes. Results, both in terms of the model capability to reproduce the current system and in evaluating its performance, are promising and, at the same time, highlight potential for improvement in the efficiency of the surgery department, which is the ultimate aim of the project. In a more advanced phase of the research, the simulation and scheduling logics will be supported by the availability of accurate predictions, enabled by a ML model. In general terms, it is expected that an accurate evaluation of the patient’s flow (based on real data) and a reliable prediction of the surgery cases (based on the ML model) will make it possible to define an optimized planning of surgical procedures, decreasing, consequently, the unused times of the ORs and hopefully, increasing the number of surgical interventions in a day. This is an innovative aspect of the whole project; indeed, although ML, simulation and scheduling have been applied in healthcare environment, their combined usage opens new ways for process improvement. This approach, thanks to its consistent predictive performance over various forecast intervals, can positively influence the choices of healthcare personnel in the short term and obviously long-term strategic planning[27].