Non-Cochrane · DTA · PM35399834
Uses of Different Machine Learning Algorithms for Diagnosis of Dental Caries.
Talpur Sarena, Azim Fahad, Rashid Munaf, et al. · Journal of healthcare engineering · 2022 · PubMed · doi:10.1155/2022/5032435
Full textEvery question was extracted from the full text, figures and supplements by two independent LLM extractors, disagreements adjudicated. Rows cite where in the review the value comes from.
Questions
21
current version
GRADE rated
0
No studies
0
Included studies
17
linked below
No studies 0 · 0%Ungraded 21 · 100%Very low 0 · 0%Low 0 · 0%Moderate 0 · 0%High 0 · 0%
Ungraded means an effect was reported without a GRADE rating; it is not a gap. No studies means the review found nothing to answer the question. The six categories are a display order, not a score.
Review question
| Element | Planned (methods) | Found (results) |
|---|---|---|
| Population | Dental X-ray images of human beings | Human dental radiographic/intraoral image datasets across 12 studies |
| Intervention | Analysis of different machine-learning algorithms used to detect dental caries | Machine-learning algorithms, deep learning dominant: CNN/FCNN, ANN, backpropagation neural network, SVM, ADA-NN, Mask R-CNN, KNN, Naive Bayes, random forest, logistic regression (n=12) |
| Comparison | Different machine-learning algorithms to predict caries | Head-to-head comparisons among ML algorithms (e.g., ADA-NN vs SVM/KNN/NB/LM-NN; SVM vs random forest/k-NN/logistic regression) (n=12) |
| Outcomes | Accuracy of algorithms | Diagnostic accuracy 69%-99% (n=12); maximum 99% for neural-network backpropagation (Javed et al.) |
| Study designs | Original studies (human, English) applying machine learning to detect any type of dental caries; IEEE conference papers eligible; reviews/meta-analyses, theses, editorials, letters, case reports, and cross-sectional studies excluded | 12 original diagnostic/machine-learning studies on image datasets (2008-2022) |
| Timing | Studies published 2008-2022 | 2008-2022 |
Search date: Not specified. Databases: PubMed; IEEE Xplore; Science Direct; Google Scholar
Questions and results
Machine-learning algorithms for dental caries detection vs Different machine-learning algorithms for caries prediction
Population: Dental images of human beings
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Caries detection accuracy in unrestored molars and premolars Duong et al., 2021; support vector machine The review reported 92.37% accuracy for automated caries detection in Duong et al. (Ungraded). Source: Table 2 · same question in other reviews | Accuracy 92.37 | 1 study · 620 participants | Ungraded |
Occlusal caries detection accuracy Moutselos et al., 2019; MASK R-CNN; worst class The review reported 66.7% accuracy for the worst class in Moutselos et al. (Ungraded). Source: Table 2 · same question in other reviews | Accuracy 66.7 | 1 study · 88 participants | Ungraded |
Proximal caries detection accuracy Devito et al., 2008; hidden-layer perceptron with backpropagation The review reported 88.4% accuracy for proximal dental caries in Devito et al. (Ungraded). Source: Table 2 · same question in other reviews | Accuracy 88.4 | 1 study · 160 participants | Ungraded |
Dental caries detection accuracy Zanella-Calzada et al., 2018; ANN The review reported 69% accuracy for dental caries detection in Zanella-Calzada et al. (Ungraded). Source: Table 2 · same question in other reviews | Accuracy 69 | 1 study · 189 participants | Ungraded |
Root caries prediction accuracy Hung et al., 2019; support vector machine The review reported 97.1% accuracy for root caries prediction in Hung et al. (Ungraded). Source: Table 2 · same question in other reviews | Accuracy 97.1 | 1 study · 5135 participants | Ungraded |
Caries detection accuracy on intraoral images Kühnisch et al., 2022; convolutional neural networks The review reported 93.3% accuracy for caries detection on intraoral images in Kühnisch et al. (Ungraded). Source: Table 2 · same question in other reviews | Accuracy 93.3 | 1 study · 2417 participants | Ungraded |
Proximal caries detection accuracy Casalegno et al., 2018; CNN The review reported 85.6% accuracy for proximal caries in Casalegno et al. (Ungraded). Source: Table 2 · same question in other reviews | Accuracy 85.6 | 1 study · 217 participants | Ungraded |
Premolar caries detection accuracy Lee et al., 2018; CNN The review reported 88% accuracy for premolar caries in Lee et al. (Ungraded). Source: Table 2 · same question in other reviews | Accuracy 88 | 1 study · 3000 participants | Ungraded |
Diagnostic accuracy for root caries detection Root caries Population: Dental radiographic/intraoral images of human teeth (X-ray and photographic) One study reported machine-learning accuracy for root caries detection (97.1%, support vector machine); not pooled and not graded. Source: Results 3.3 (Hung et al.); Discussion; Table 2 · same question in other reviews | Accuracy 97.1 | 1 study | Ungraded |
Dental caries detection accuracy Mayank et al., 2017; F-CNN The review reported 70% accuracy for dental caries detection in Mayank et al. (Ungraded). Source: Table 2 · same question in other reviews | Accuracy 70 | 1 study · 3000 participants | Ungraded |
Molar and premolar caries detection accuracy Lee et al., 2018; CNN The review reported 82% accuracy for combined molar and premolar caries in Lee et al. (Ungraded). Source: Table 2 · same question in other reviews | Accuracy 82 | 1 study · 3000 participants | Ungraded |
Post-Streptococcus mutans prediction accuracy Javed et al., 2019; feedforward backpropagation ANN The review reported 99% accuracy for post-Streptococcus mutans prediction in Javed et al. (Ungraded). Source: Table 2 · same question in other reviews | Accuracy 99 | 1 study · 45 participants | Ungraded |
Dental caries detection accuracy Geetha et al., 2020; backpropagation The review reported 97.1% accuracy for dental caries diagnosis in Geetha et al. (Ungraded). Source: Table 2 · same question in other reviews | Accuracy 97.1 | 1 study · 105 participants | Ungraded |
Molar caries detection accuracy Lee et al., 2018; CNN The review reported 89% accuracy for molar caries in Lee et al. (Ungraded). Source: Table 2 · same question in other reviews | Accuracy 89 | 1 study · 3000 participants | Ungraded |
Occlusal caries detection accuracy Moutselos et al., 2019; MASK R-CNN; center pixel class The review reported 77.8% accuracy for the center pixel class in Moutselos et al. (Ungraded). Source: Table 2 · same question in other reviews | Accuracy 77.8 | 1 study · 88 participants | Ungraded |
Occlusal caries detection accuracy Casalegno et al., 2018; CNN The review reported 83.6% accuracy for occlusal caries in Casalegno et al. (Ungraded). Source: Table 2 · same question in other reviews | Accuracy 83.6 | 1 study · 217 participants | Ungraded |
Occlusal caries detection accuracy Moutselos et al., 2019; MASK R-CNN; most common class The review reported 88.9% accuracy for the most common class in Moutselos et al. (Ungraded). Source: Table 2 · same question in other reviews | Accuracy 88.9 | 1 study · 88 participants | Ungraded |
Adaptive dragonfly algorithm with neural-network classifier vs K-nearest neighbors; support vector machine; Naive Bayes; LM-NN
Population: Dental images of human beings
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Accuracy advantage in test case 1 Patil et al., 2019 The review reported a 5.55% accuracy advantage in Patil test case 1 (Ungraded). Source: Table 2 · same question in other reviews | Accuracy advantage 5.55 | 1 study · 40 participants | Ungraded |
Adaptive dragonfly algorithm with neural-network classifier vs KNN or SVM (not disaggregated in source)
Population: Dental images of human beings
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Accuracy advantage in test case 2 Patil et al., 2019 The review reported an 11.76% accuracy advantage in Patil test case 2 (Ungraded). Source: Table 2 · same question in other reviews | Accuracy advantage 11.76 | 1 study · 40 participants | Ungraded |
Accuracy advantage in test case 2 Patil et al., 2019 The review reported a 52% accuracy advantage in Patil test case 2 (Ungraded). Source: Table 2 · same question in other reviews | Accuracy advantage 52 | 1 study · 40 participants | Ungraded |
Adaptive dragonfly algorithm with neural-network classifier vs Support vector machine and Naive Bayes
Population: Dental images of human beings
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Accuracy advantage in test case 3 Patil et al., 2019 The review reported a 6.30% accuracy advantage in Patil test case 3 (Ungraded). Source: Table 2 · same question in other reviews | Accuracy advantage 6.30 | 1 study · 40 participants | Ungraded |
Included primary studies 17
Open as a trial list| Study | Also included by | Outcomes it informs here | Trial PICO | Results |
|---|---|---|---|---|
| Geetha 2020 Dental caries diagnosis in digital radiographs using back-propagation neural network. PMID:31949895 | 1 other review | not stated | not extracted yet | not extracted yet |
| Kuhnisch 2022 Caries Detection on Intraoral Images Using Artificial Intelligence. PMID:34416824 | 1 other review | not stated | not extracted yet | not extracted yet |
| Lee 2018 Detection and diagnosis of dental caries using a deep learning-based convolutional neural network algorithm. PMID:30056118 | 1 other review | not stated | not extracted yet | not extracted yet |
| Casalegno 2018 Caries Detection with Near-Infrared Transillumination Using Deep Learning. PMID:31449759 | this review only | not stated | not extracted yet | not extracted yet |
| Devito 2008 An artificial multilayer perceptron neural network for diagnosis of proximal dental caries DOI:10.1016/j.tripleo.2008.03.002 | this review only | not stated | not extracted yet | not extracted yet |
| Devito 2008 An artificial multilayer perceptron neural network for diagnosis of proximal dental caries. PMID:18718785 | this review only | not stated | not extracted yet | not extracted yet |
| Duong 2021 Automated caries detection with smartphone color photography using machine learning DOI:10.1177/14604582211007530 | this review only | not stated | not extracted yet | not extracted yet |
| Duong 2021 Automated caries detection with smartphone color photography using machine learning. PMID:33863251 | this review only | not stated | not extracted yet | not extracted yet |
| Hung 2019 Application of machine learning for diagnostic prediction of root caries. PMID:31274221 | this review only | not stated | not extracted yet | not extracted yet |
| Javed 2019 Development of artificial neural network model for prediction of post-streptococcus mutans in dental caries DOI:10.1016/j.cmpb.2019.105198 | this review only | not stated | not extracted yet | not extracted yet |
| Javed 2019 Development of artificial neural network model for prediction of post-streptococcus mutans in dental caries. PMID:31760304 | this review only | not stated | not extracted yet | not extracted yet |
| Lee 2018 Detection and diagnosis of dental caries using a deep learning-based convolutional neural network algorithm DOI:10.1016/j.jdent.2018.07.015 | this review only | not stated | not extracted yet | not extracted yet |
| Mayank 2017 matched by author and year (probable) | this review only | not stated | not extracted yet | not extracted yet |
| Moutselos 2019 Recognizing Occlusal Caries in Dental Intraoral Images Using Deep Learning DOI:10.1109/embc.2019.8856553 | this review only | not stated | not extracted yet | not extracted yet |
| Moutselos 2019 Recognizing Occlusal Caries in Dental Intraoral Images Using Deep Learning. PMID:31946206 | this review only | not stated | not extracted yet | not extracted yet |
| Patil 2019 Algorithmic analysis for dental caries detection using an adaptive neural network architecture. PMID:31080904 | this review only | not stated | not extracted yet | not extracted yet |
| Zanella-Calzada 2018 Deep Artificial Neural Networks for the Diagnostic of Caries Using Socioeconomic and Nutritional Features as Determinants: Data from… PMID:29912173 | this review only | not stated | not extracted yet | not extracted yet |
Overlap with other reviews
Reviews that include the same primary studies are not independent evidence. Shared counts match studies by PMID, DOI or NCT, and otherwise by author and year (a probable match).