Recent research suggests that dentists can use artificial neural networks to understand better the complex factors affecting postoperative pain and even predict it.
By Dr. Mehmood Asghar BDS, M Phil, Ph.D.
Worldwide, root canal therapy is among the most commonly performed dental procedures. According to the American Association of Endodontists, more than 15 million root canal procedures are performed annually in the U.S. alone. Despite being an excellent option for saving grossly damaged teeth, many patients experience post-endodontic treatment pain. It's estimated that over 25% of endodontic patient experience moderate to severe pain after a root canal procedure [Gao et al., 2021].
While there are various strategies for obtaining post-endodontic treatment pain control, the diagnosis of this type of pain is mainly made objectively by dentists. Specifically, postoperative pain following root canal treatment is multifactorial. These factors include the type of procedure performed, the number of sittings involved, the use of any pain medication during treatment, and patient-related, psychosocial, and demographic factors.
Here comes the problem; with multiple factors involved, dentists often need help precisely predicting or quantifying postoperative pain. This uncertainty can lead to a loss of trust or miscommunication between dentists and patients.
By now, you might think the solution is simple: Give the patient pain medication, and they are good to go! But is it the ultimate solution? Wouldn't it be ideal if we could predict if any patient will experience moderate to severe post-endodontic pain and take appropriate measures in advance? But how?
Dentists have been struggling to find a solution to this problem for many decades. But we finally have an answer, thanks to artificial intelligence and artificial neural networks.
Neural Networks – What's the Hype?
Artificial Neural networks (ANNs) fall under the umbrella of artificial intelligence. This system imitates the function and structure of the brain to analyze various predictors and their complex relations [Han et al., 2013]. In short, neural networks simplify complex interactions and relationships and help us make precise, scientific evidence-based decisions. Recently, ANNs have shown the ability to predict the likelihood and severity of postoperative pain precisely.
ANNs and Pain Prediction
A recent study published in the Community Dentistry and Oral Epidemiology Journal used an ANN-based predictive model for one-week acute and six-month persistent pain in patients who had undergone endodontic treatment and showed interesting results. The study's results [Du et al., 2022] showed that the incidence of pre and intraoperative pain played a vital role in causing postoperative pain. Similarly, clinical factors, such as the type of the procedure, the differences in the dentist's approach to performing the procedure, and the severity of the issue, were more important in causing postoperative pain than non-clinical factors, such as the patient's socioeconomic level, gender, and age.
While ANNs have found applications in virtually all spheres of life, it has a crucial role in medicine dentistry. Studies have shown that in addition to simplifying the decision-making process for clinicians, the excellent accuracy of these models – thanks to machine learning – enables clinicians to make clinically sound decisions. For example, a study by Gao et al.  showed that the accuracy of their model to predict dental pain based on its associated factors was more than 95%.
As clinicians, our success depends on how happy and satisfied our patients are. Unfortunately, patient satisfaction is affected when they experience intra or postoperative pain, making them afraid of seeking dental treatment.
As such, AI-driven solutions could significantly improve patient satisfaction and ensure a more comfortable recovery period. And it's not just dental pain; AI has multiple potential applications in dentistry – from caries detection to treatment planning and diagnosis of pathologies. So perhaps it's time to take the plunge and start leveraging AI in our dental practices to improve the efficiency and accuracy of our procedures, as well as patient satisfaction.
If you're not yet subscribed to receive the Incisor newsletter, filled with cutting-edge dental news sent directly to your inbox twice a month, you can do so here.
Author: Dr. Mehmood Asghar is a dentist by profession and an Assistant Professor of Dental Biomaterials at the National University of Medical Sciences, Pakistan. Dr. Asghar received his undergraduate and postgraduate dental qualifications from the National University of Science and Technology (NUST). He has recently received a Ph.D. in Restorative Dentistry from the University of Malaya, Malaysia. Apart from his hectic clinical and research activities, Dr. Asghar likes to write evidence-based, informative articles for dental professionals and patients. Dr. Asghar has published several articles in international, peer-reviewed journals.
Du M, Haag DG, Lynch JW, Mittinty MN. Application of multilevel models for predicting pain following root canal treatment. Community Dentistry and Oral Epidemiology. DOI: https://doi.org/10.1111/cdoe.12807.
Gao X, Xin X, Li Z, Zhang W. Predicting postoperative pain following root canal treatment using artificial neural network evaluation. Scientific Reports. 2021 2021/08/26;11(1):17243.
Han H-G, Wang L-D, Qiao J-F. Efficient self-organizing multilayer neural network for nonlinear system modeling. Neural Networks. 2013;43:22-32.