Shivani Mallishery1, Pavan Chhatpar2, K S Banga3, Trusha Shah3, Pankaj Gupta3. 1. Nair Hospital Dental College, Mumbai, 400008, India. sshivani96@gmail.com. 2. Northeastern University, Boston, USA. 3. Department of Conservative Dentistry and Endodontics, Nair Hospital Dental College, Mumbai, India.
Abstract
OBJECTIVES: Endodontic treatment works as a successful treatment modality in several cases. However, it may fail due to some reasons unforeseeable by the dentist. Many failures can be prevented by carefully assessing the difficulty level of the case before initiating treatment or by referral to a specialist. This study presents an approach using machine learning to generate an algorithm which can help predict the difficulty level of the case and decide about a referral, with the help of the standard American Association of Endodontists (AAE) Endodontic Case Difficulty Assessment Form. MATERIALS AND METHODS: Using the AAE Endodontic Case Difficulty Form after obtaining the patients' consent, 500 potential root canal patients were diagnosed. The filled forms were assessed by two pre-calibrated endodontists, and, in cases of conflicting opinion, a third endodontist's opinion was taken. Artificial neural network was used for generating the algorithm. RESULTS: Using 500 filled AAE forms, a sensitivity of 94.96% was achieved by the machine learning algorithm. CONCLUSION: This study provides an option for automation to the conventional method of predicting the difficulty level of a case, thus increasing the speed of decision-making and referrals if necessary. CLINICAL RELEVANCE: An AAE Endodontic Case Difficulty Assessment Form when utilized along with machine learning can assist general dentists in rapid assessment of the case difficulty. This is a helpful tool in developing countries, where endodontic treatment and referral guidelines are often neglected. It also helps to make difficulty level assessments easier for novice practitioners, when they are in doubt about the same.
OBJECTIVES: Endodontic treatment works as a successful treatment modality in several cases. However, it may fail due to some reasons unforeseeable by the dentist. Many failures can be prevented by carefully assessing the difficulty level of the case before initiating treatment or by referral to a specialist. This study presents an approach using machine learning to generate an algorithm which can help predict the difficulty level of the case and decide about a referral, with the help of the standard American Association of Endodontists (AAE) Endodontic Case Difficulty Assessment Form. MATERIALS AND METHODS: Using the AAE Endodontic Case Difficulty Form after obtaining the patients' consent, 500 potential root canal patients were diagnosed. The filled forms were assessed by two pre-calibrated endodontists, and, in cases of conflicting opinion, a third endodontist's opinion was taken. Artificial neural network was used for generating the algorithm. RESULTS: Using 500 filled AAE forms, a sensitivity of 94.96% was achieved by the machine learning algorithm. CONCLUSION: This study provides an option for automation to the conventional method of predicting the difficulty level of a case, thus increasing the speed of decision-making and referrals if necessary. CLINICAL RELEVANCE: An AAE Endodontic Case Difficulty Assessment Form when utilized along with machine learning can assist general dentists in rapid assessment of the case difficulty. This is a helpful tool in developing countries, where endodontic treatment and referral guidelines are often neglected. It also helps to make difficulty level assessments easier for novice practitioners, when they are in doubt about the same.
Entities:
Keywords:
Artificial intelligence; Case difficulty; Machine learning; Referral; Treatment planning