Influence of handling-relevant factors on the behaviour of a novel calculus-detection device

Artificial Intelligence Techniques: Analysis, Application, and Outcome in Dentistry-A Systematic Review

OBJECTIVE

The objective of this systematic review was to investigate the quality and outcome of studies into artificial intelligence techniques, analysis, and effect in dentistry.

MATERIALS AND METHODS

Using the MeSH keywords: artificial intelligence (AI), dentistry, AI in dentistry, neural networks and dentistry, machine learning, AI dental imaging, and AI treatment recommendations and dentistry. Two investigators performed an electronic search in 5 databases: PubMed/MEDLINE (National Library of Medicine), Scopus (Elsevier), ScienceDirect databases (Elsevier), Web of Science (Clarivate Analytics), and the Cochrane Collaboration (Wiley). The English language articles reporting on AI in different dental specialties were screened for eligibility. Thirty-two full-text articles were selected and systematically analyzed according to a predefined inclusion criterion. These articles were analyzed as per a specific research question, and the relevant data based on article general characteristics, study and control groups, assessment methods, outcomes, and quality assessment were extracted.

RESULTS

The initial search identified 175 articles related to AI in dentistry based on the title and abstracts. The full text of 38 articles was assessed for eligibility to exclude studies not fulfilling the inclusion criteria. Six articles not related to AI in dentistry were excluded. Thirty-two articles were included in the systematic review. It was revealed that AI provides accurate patient management, dental diagnosis, prediction, and decision making. Artificial intelligence appeared as a reliable modality to enhance future implications in the various fields of dentistry, i.e., diagnostic dentistry, patient management, head and neck cancer, restorative dentistry, prosthetic dental sciences, orthodontics, radiology, and periodontics.

CONCLUSION

The included studies describe that AI is a reliable tool to make dental care smooth, better, time-saving, and economical for practitioners. AI benefits them in fulfilling patient demand and expectations. The dentists can use AI to ensure quality treatment, better oral health care outcome, and achieve precision. AI can help to predict failures in clinical scenarios and depict reliable solutions. However, AI is increasing the scope of state-of-the-art models in dentistry but is still under development. Further studies are required to assess the clinical performance of AI techniques in dentistry.

Machine Learning and Intelligent Diagnostics in Dental and Orofacial Pain Management: A Systematic Review

Purpose

The study explored the clinical influence, effectiveness, limitations, and human comparison outcomes of machine learning in diagnosing (1) dental diseases, (2) periodontal diseases, (3) trauma and neuralgias, (4) cysts and tumors, (5) glandular disorders, and (6) bone and temporomandibular joint as possible causes of dental and orofacial pain.

Method

Scopus, PubMed, and Web of Science (all databases) were searched by 2 reviewers until 29^th October 2020. Articles were screened and narratively synthesized according to PRISMA-DTA guidelines based on predefined eligibility criteria. Articles that made direct reference test comparisons to human clinicians were evaluated using the MI-CLAIM checklist. The risk of bias was assessed by JBI-DTA critical appraisal, and certainty of the evidence was evaluated using the GRADE approach. Information regarding the quantification method of dental pain and disease, the conditional characteristics of both training and test data cohort in the machine learning, diagnostic outcomes, and diagnostic test comparisons with clinicians, where applicable, were extracted.

Results

34 eligible articles were found for data synthesis, of which 8 articles made direct reference comparisons to human clinicians. 7 papers scored over 13 (out of the evaluated 15 points) in the MI-CLAIM approach with all papers scoring 5+ (out of 7) in JBI-DTA appraisals. GRADE approach revealed serious risks of bias and inconsistencies with most studies containing more positive cases than their true prevalence in order to facilitate machine learning. Patient-perceived symptoms and clinical history were generally found to be less reliable than radiographs or histology for training accurate machine learning models. A low agreement level between clinicians training the models was suggested to have a negative impact on the prediction accuracy. Reference comparisons found nonspecialized clinicians with less than 3 years of experience to be disadvantaged against trained models.

Conclusion

Machine learning in dental and orofacial healthcare has shown respectable results in diagnosing diseases with symptomatic pain and with improved future iterations and can be used as a diagnostic aid in the clinics. The current review did not internally analyze the machine learning models and their respective algorithms, nor consider the confounding variables and factors responsible for shaping the orofacial disorders responsible for eliciting pain.

Influence of motivation and a new digitized training program on undergraduate dental students during preclinical scaling training

BACKGROUND

The current study evaluated whether a new digitized scaling training program (DTP: n = 30; supervisor-student-ratio 1:10) improves the performance of undergraduate dental student during a preclinical course in regard to two different instruments [sonic scalers (AIR) and Gracey curettes (GRA)] compared to a conventional training program (CTP: n = 19; supervisor-student-ratio 1:4).

METHODS

All the participants received a two-hour lecture on both instruments, followed by a 12-week period with a weekly training program lasting 45 min (10 sessions); one group was supported by DTP. At the end of the training phase, all the participants performed the subgingival scaling of six equivalent test teeth using GRA and AIR. Treatment time, proportion of removed simulated biofilm (relative cleaning efficacy, RCE-b) and hard deposits (RCE-d) were recorded. By using a pseudonymized questionnaire with a 5-point Likert scale, self-assessment of scaling effort, handling, root surface roughness/destruction and effectiveness were evaluated. In addition, personal data such as age, gender, handedness, regularity of playing computer games/consoles and previous dental/technical or medical education were elevated and correlated with cleaning efficacy.

RESULTS

The DTP participants showed higher effectiveness in RCE-b compared to those who used the CTP with GRA (71.54% vs. 67.23%, p = 0.004) and AIR (71.75% vs. 62.63%, p ≤ 0.001), and the DTP students were faster with both instruments (p ≤ 0.001). For RCE-d, there was no significant difference between the DTP and CTP groups (GRA p = 0.471; AIR p = 0.158), whereas DTP showed better RCE-d results with GRA versus AIR (84.68% vs. 77.85%, p < 0.001). According to the questionnaire, no significant differences were detected between the training groups in terms of self-assessment, handling, treatment time, root surface roughness/destruction or effectiveness of the instruments. The CTP group favored AIR compared to GRA regarding the fatigue effect. The CTP and playing computer games/consoles regularly was correlated with lower RCE-b, whereas previous education in medicine/dentistry was correlated with higher RCE-b values.

CONCLUSIONS

Within the limitations of the study, the DTP with a reduced supervision effort compared to the CTP resulted in higher effectiveness and lower instrumentation time for removing simulated biofilms.

How to train periodontal endoscopy? Results of a pilot study removing simulated hard deposits in vitro

OBJECTIVES

There is uncertainty regarding the benefits of periodontal endoscopy (PE) for subgingival instrumentation. Moreover, the influence of operators' experience and training with such a device on treatment results is unclear. Therefore, we compared in an in vitro study the use of PE for scaling and root planing (SRP) with the conventional non-surgical therapy, hypothesizing that using a PE allows to remove more simulated hard deposits than without (nPE), influenced by the operators' experience and training.

MATERIAL AND METHODS

A sonic device and Gracey curettes were utilized by 11 operators (six dentists, five dental hygienists) in periodontitis manikins heads. The time required for treatment and the proportion of removed simulated hard deposits by SRP were measured.

RESULTS

Using the PE led to a significant difference in removal of simulated hard deposits (%) (mean ± SD) irrespective of operators' experience (PE 90.78 ± 12.10% (range 58.80-100%); nPE 79.98 ± 22.15% (range 38.10-100%, p < 0.001)), sub-analyses for different tooth types demonstrated a significant difference in favor of PE for front teeth (p < 0.001) and in the upper jaw independent of the tooth type (p < 0.001). Comparison of treatment times for two quadrants with and without PE showed a significantly longer treatment time with PE (∆22.27 ± 17.98 mins, p <0.001).

CONCLUSIONS

Within the present pilot study, the use of PE led to more removal of simulated hard deposits but was concomitantly related to more time and financial effort. Using PE was most beneficial in the front area.

CLINICAL RELEVANCE

PE may provide additional benefits for the removal of hard deposits compared to traditional SRP. The beneficial effect of PE seems not to be influenced by operators experience nor by pocket probing depth.

Evaluation of calculus imaging on root surfaces by spectral-domain optical coherence tomography

OBJECTIVE

The aim of this in vitro study was to evaluate the ability of optical coherence tomography (OCT) to display calculus on root surfaces.

MATERIAL AND METHODS

Ten teeth with calculus on the root surface were embedded in resin, omitting the root surface. A region of interest (ROI) was marked by small drill holes coronally and apically of the calculus and imaged by spectral-domain optical coherence tomography ([SD OCT], Telesto SP5, centre wavelength 1310 nm) and light microscopy (LM). To evaluate the impact of different fluids on calculus visualisation, using OCT, root surfaces were covered by a layer of NaCl and blood and displayed by OCT. Subsequently, teeth were completely covered with resin and sectioned for histological evaluation. Within the ROI, lengths of root surface and calculus were measured by LM and OCT, and the ratio [%] was calculated. In addition, at three sites of each ROI, agreement of presence and length of calculus was evaluated. Both methods were compared using Pearson's correlation.

RESULTS

Regarding the presence of calculus, agreement between LM and OCT was strong (κ_i = 0.783, p = 0.033), and measurements regarding the length of the calculus were strongly correlated (r_i >0.906; p_i <0.001). However, the values differed for dry (p = 0.023) and NaCl-covered root surfaces (p = 0.035).

CONCLUSION

Calculus on the root surface can be displayed by SD-OCT, which therefore may be suited as imaging technology for subgingival calculus in periodontal pockets.

The effect of a 3% hydroxyapatite paste prophylaxis after different root-scaling procedures in periodontics

Abstract Introduction Root-scaling procedures on the root surface show to increase root surface roughness during instrumentation. The use of a prophylactic paste reduces the roughness. Objetive This study evaluated the influence of prophylaxis paste on the surface roughness of root dentin after hand instruments (HI), ultrasonic scalers (US) and rotational instruments (RI) were applied to root surfaces. Material and method Fifty bovine roots were prepared and randomly divided into 10 groups (n=5) in which instrumentation with HI, US, or RI was carried out with or without prophylaxis using either conventional or 3% hydroxyapatite paste; additionally, a control group without treatment was included. The Ra and Rz roughness values were measured using a perfilometer. The morphological aspects of the root surfaces were analysed qualitatively by scanning electron microscopy (SEM). Data were analysed using two-way ANOVA followed by Tukey’s test (α = 0.05). Result The scaling method and the prophylaxis procedure significantly influenced (p < 0.05) root surface roughness. Treatment with US and prophylaxis with 3% hydroxyapatite paste resulted in lower values of Ra (0.14 μm ± 0.02 μm) and Rz (0.89 μm ± 0.18 μm) compared to the other groups. SEM analysis showed smoother and more homogeneous surfaces in the groups with prophylaxis compared to the groups without prophylaxis. There were no substantial differences between the results of applying different pastes. Conclusion The prophylaxis procedure performed with the experimental paste containing 3% hydroxyapatite reduced the roughness of root surfaces treated with HI, US and RI at the same level as the conventional prophylactic paste.

Reliability of recordings of subgingival calculus detected using an ultrasonic device

OBJECTIVES

To assess the intra-examiner reliability of recordings of subgingival calculus detected using an ultrasonic device, and to investigate the influence of subject-, tooth- and site-level factors on the reliability of these subgingival calculus recordings.

MATERIALS AND METHODS

On two occasions, within a 1-week interval, 147 adult periodontitis patients received a full-mouth clinical periodontal examination by a single trained examiner. Duplicate subgingival calculus recordings, in six sites per tooth, were obtained using an ultrasonic device for calculus detection and removal.

RESULTS

Agreement was observed in 65 % of the 22,584 duplicate subgingival calculus recordings, ranging 45 % to 83 % according to subject. Using hierarchical modeling, disagreements in the subgingival calculus duplicate recordings were more likely in all other sites than the mid-buccal, and in sites harboring supragingival calculus. Disagreements were less likely in sites with PD ≥  4 mm and with furcation involvement  ≥  degree 2. Bleeding on probing or suppuration did not influence the reliability of subgingival calculus. At the subject-level, disagreements were less likely in patients presenting with the highest and lowest extent categories of the covariate subgingival calculus.

CONCLUSIONS AND CLINICAL RELEVANCE

The reliability of subgingival calculus recordings using the ultrasound technology is reasonable. The results of the present study suggest that the reliability of subgingival calculus recordings is not influenced by the presence of inflammation. Moreover, subgingival calculus can be more reliably detected using the ultrasound device at sites with higher need for periodontal therapy, i.e., sites presenting with deep pockets and premolars and molars with furcation involvement.

Application of ultrasound in periodontics: Part II

Ultrasound offers great potential in development of a noninvasive periodontal assessment tool that would offer great yield real time information, regarding clinical features such as pocket depth, attachment level, tissue thickness, histological change, calculus, bone morphology, as well as evaluation of tooth structure for fracture cracks. In therapeutics, ultrasonic instrumentation is proven effective and efficient in treating periodontal disease. When used properly, ultrasound-based instrument is kind to the soft tissues, require less healing time, and are less tiring for the operator. Microultrasonic instruments have been developed with the aim of improving root-surface debridement. The dye/paper method of mapping ultrasound fields demonstrated cavitational activity in an ultrasonic cleaning bath. Piezosurgery resulted in more favorable osseous repair and remodeling in comparison with carbide and diamond burs. The effect of ultrasound is not limited to fracture healing, but that bone healing after osteotomy or osteodistraction could be stimulated as well.

In vitro calculus detection with a moved smart ultrasonic device

STUDY OBJECTIVE

The objective of subgingival instrumentation of periodontally diseased root surfaces is to remove the adhering microbial biofilm and calcified deposits. Recently, we have described an automated calculus detection system under static conditions. Clinically however, the tip of the system has to be moved over tooth surfaces. It was thus necessary to study the entire system in motion.

METHODS

The detection device is based on a conventional dental piezoelectric ultrasonic handpiece with a conventional scaler insert. The impulse response of the mechanical oscillation system is analysed by a fuzzy logic-based computerized algorithm, which classifies various surfaces. The present study investigates dental surface recognition properties of the new system with the tip being moved over teeth surfaces in vitro. Following a training set of 7977 measurements (3960 calculus, 4017 cement) on 200 extracted teeth, 1363 measurements were conducted on 34 teeth unknown to the system.

RESULTS

The surfaces cementum and calculus were correctly classified in 78% within the training set and in 81% within the set unknown, with a kappa value of 0.68.

CONCLUSION

It was shown that this method of automatic recognition of tooth surfaces is able to distinguish between different tooth surfaces in vitro independently from tip movements.

A new system to detect residual subgingival calculus: in vitro detection limits

AIM

We recently introduced an experimental surface detection system based on a conventional dental ultrasonic scaler. This device automatically discriminates cementum and dental calculus, which is the prerequisite for complete and thorough calculus removal. In the present study, the detection limits of this device were tested in vitro.

MATERIAL AND METHODS

From 50 extracted teeth, subgingival calculus was gradually removed using a Gracey curette. During this stepwise procedure, detection properties of the surface detection system were continuously monitored and systematically verified until the system stopped discriminating calculus from the root surface. By measuring the diameter, circumference and area of the smallest, yet recognizable deposit, and of the no longer recognizable deposit, the cut-off point of the discriminative capability of the detection device was determined.

RESULTS

The cut-off points for the correct classification of residual deposits averaged on a diameter of 219 microm, an area of 21,600 microm2, and a circumference of 748 microm. This means a sensitivity of 73% and a specificity of 80% in this critical area.

CONCLUSIONS

This calculus detection system was able to detect small deposits. In clinical practice, this device may support dentists in deciding whether to stop or to continue the debridement.

A method for the validation of a new calculus detection system

BACKGROUND

Recently, pilot studies from our laboratory have shown that dental surfaces may be discriminated by the analysis of tip oscillations of an ultrasonic instrument, which possesses computerized calculus-detection features. For the evaluation of this smart detection system, its surface recognition qualities are of crucial importance. For in vivo studies, however, it proved to be difficult to verify the subgingival detection results. Therefore, it was necessary to develop a method, which allowed a reliable validation of surface recognition results of this new device. This evaluation method is described here.

MATERIALS AND METHODS

Thirty extracted human teeth with subgingival calculus were embedded with plaster in a tray. To simulate subgingival pockets, dissected mucoperiostal porcine gingiva was sutured on the teeth. The thus-constructed dentition was mounted into a phantom head. A CCD-cam was attached with an intra-oral X-ray mount to the teeth. The dentist scanned the pockets with the ultrasonic instrument, simultaneously videotaping the scanning path of the supragingival portion of the insert. At the same time, the signals of the modified ultrasound scaler were recorded. After the tooth was removed from the phantom head, the tip of the ultrasound scaler could be repositioned using the video sequences. The actual insert location on calculus or cementum was assessed and compared with the computer signals. The whole procedure was repeated a second time and the reproducibility of the evaluation method was estimated.

RESULTS

A kappa value of 0.95 was attained for the evaluation method.

CONCLUSION

The present experimental design allows the in vitro repositioning of an automated dental instrument for the detection of subgingival surfaces on the tooth following an in vitro phantom-head video recording of its intra-oral scanning movements. This method will be used for the verification of in vivo results of a new ultrasound-based surface detection system.