Orthodontic referrals: why do GDPs get it wrong?

Comparison of deep learning models to detect crossbites on 2D intraoral photographs

BACKGROUND

To support dentists with limited experience, this study trained and compared six convolutional neural networks to detect crossbites and classify non-crossbite, frontal, and lateral crossbites using 2D intraoral photographs.

METHODS

Based on 676 photographs from 311 orthodontic patients, six convolutional neural network models were trained and compared to classify (1) non-crossbite vs. crossbite and (2) non-crossbite vs. lateral crossbite vs. frontal crossbite. The trained models comprised DenseNet, EfficientNet, MobileNet, ResNet18, ResNet50, and Xception.

FINDINGS

Among the models, Xception showed the highest accuracy (98.57%) in the test dataset for classifying non-crossbite vs. crossbite images. When additionally distinguishing between lateral and frontal crossbites, average accuracy decreased with the DenseNet architecture achieving the highest accuracy among the models with 91.43% in the test dataset.

CONCLUSIONS

Convolutional neural networks show high potential in processing clinical photographs and detecting crossbites. This study provides initial insights into how deep learning models can be used for orthodontic diagnosis of malocclusions based on intraoral 2D photographs.

UK dental students' ability and confidence in applying the Index of Orthodontic Treatment Need and determining appropriate orthodontic referral

INTRODUCTION

The aim was to assess the ability and confidence of UK undergraduate dental students in applying the Index of Orthodontic Treatment Need (IOTN) and determining appropriate orthodontic referral.

MATERIALS AND METHODS

This was a cross-sectional survey using a case-based online questionnaire. Fourth and fifth year undergraduate dental students were recruited from UK dental schools through their university and social media. Six cases were presented and participants were asked to provide an IOTN score and judgement about referral. Participants were asked about their confidence, experience and orthodontic teaching.

RESULTS

Sixty-nine responses were returned. A quarter of participants reported having used IOTN before in a clinical setting. Clinical experience with IOTN influenced confidence. Familiarity with making orthodontic referrals was low and only one participant reported having made an orthodontic referral. Correct IOTN scores were given by 68% of participants for a large 14 mm overjet (5a) case, 43% of participants for an impacted canine (5i) case and 26% of participants for an impacted premolar (5i) case. Incorrect IOTN was most common in hypodontia cases with only 19% correctly identifying mild hypodontia (4h) and 28% identifying severe hypodontia (5h). For the majority of cases, incorrect answers about referral were due to confusion between specialist practitioner and orthodontic consultant pathways.

CONCLUSIONS

Dental students' ability and confidence in correctly applying the IOTN Dental Health Component and selecting the appropriate referral pathway was inadequate. Responses suggest a lack of clinical experience in assessing patients, applying the IOTN and making referrals. The low response rate is disappointing and limits the scope for making recommendations.

Awareness of patient’s orthodontic problems and referral practices among general dental practitioners and non-orthodontic specialists

Background/Aim. Dentists and particularly general dental practitioners (GDPs) should be able to identify problems such as malocclusions and refer patients with such problems to orthodontics. The aim of the study was to evaluate the awareness of orthodontic problems and referral practices among GDPs and non-orthodontic specialists practicing in the Kingdom of Saudi Arabia. Methods. A cross-sectional prospective study included GDP and specialists/consultants of both genders, aged between 22?60 years, practicing general dentistry and specialists other than orthodontics who had worked in their field for two or more years. Non-practicing dentists and nonregistered dentists were excluded. Data was analyzed on SPSS version 21.00. Results. Among the respondents, 55.5% were men, and 44% were GDPs. Out of the total number of respondents, 107 (53.5%) GDPs and 66 (33%) non-orthodontic specialists practiced orthodontic referral in their practices. By using the ?2 test, statistical analysis of different variables depending on the participants? gender and specialty was performed. It revealed that variables with significant p-value were: orthodontist referral, the recommendation of orthodontic treatment only after the eruption of all permanent teeth, awareness of tooth extraction for alignment of irregular teeth, and awareness regarding worsening of temporomandibular joint problems by orthodontic treatment. Conclusion. The referral practices of dentists practicing in the Kingdom of Saudi Arabia concerning orthodontic patients were satisfactory, although awareness regarding orthodontic problems requires continuous learning and considerate positive progression toward basic orthodontics.

Does the All Wales Universal Orthodontic Referral Form enable accurate triage of new NHS orthodontic patients? A service evaluation

Introduction This study was designed to investigate the accuracy of clinical information provided by referring general dental practitioners (GDPs) following the introduction of a standardised referral form across Wales (the All Wales Universal Orthodontic Referral Form [AWUORF]) and to see whether the information given could be reliably used to screen the referrals.Aim To evaluate whether priority cases were being readily identified and whether inappropriate referrals could be minimised, thereby potentially reducing waiting lists.Method A service evaluation involving the retrospective study of 200 consecutive referrals to a specialist practice over a three-month period. A descriptive data analysis was undertaken.Results The GDPs had successfully identified the main complaint in 156 (78%) of the referrals. Of the 44 (22%) clinically inaccurate referrals, there was no impact on the patient in terms of referral pathway in 32 (16%) cases, but in the remaining 12 (6%), 5 (2.5%) cases were prioritised unnecessarily and the remaining 7 (3.5%) would have been seen more quickly had the GDP provided the relevant clinical information. The appropriateness of referral in terms of eligibility of the patient to receive NHS-funded orthodontic treatment was high with only 18 (9%) patients failing to meet the criteria.Conclusion The AWUORF successfully guides the GDP to make appropriate referrals and enables accurate triage in the majority of cases.