Study protocol
This study assessed the effect of feature selection technique (FST) on DLM performance in classifying ERR lesions. In the first stage of this study, image pre-processing was performed using Contrast-Limited Adaptive Histogram Equalization (CLAHE) filter. Then, image classification analysis were conducted using two pretrained deep convolutional neural networks (CNN), namely, EfficiennetB4 [27] and VGG16. These two deep CNNs were ensemble with two machine learning classifiers; Random Forest (RF) and Support Vector Machine (SVM) to perform ERR classification. As a result, four DLMs were developed; RF with VGG16 (RF+VGG), RF with EfficientnetB4 (RF+EFNET), SVM with VGG16 (SVM+VGG) and SVM with EfficientnetB4 (SVM+EFNET) in the first stage. In the second stage, a feature selection algorithm (Boruta and RFE) was employed to generate four new optimized DLMs (FS+RF+VGG, FS+RF+EFNET, FS+SVM+VGG and FS+SVM+EFNET) [27]. The Institutional Review Board of the Medical Ethics Committee Faculty of Dentistry University of Malaya (DF RD2030/0139 (L)) has reviewed and approved this study protocol.
2.1 Dataset
A total of 88 extracted premolars were collected from the Faculty of Dentistry, University of Malaya. The inclusion criteria set for this study were absence of root destruction, complete root formation, absence of caries or abrasions in the cervical region, and no endodontic treatment. [28] Tungsten burrs of various sizes (0.5 mm, 1.0 mm, and 2.0 mm) were used to simulate different depths ERR on each tooth at three locations (cervical, mid-root and apical). All teeth were scanned with a CBCT machine (CS 9000 CBCT, Carestream Dental, Atlanta, GA). The acquisition settings were 65 kVp, 5 mA, 10.8 S 5x3 cm F.O.V., 0.076mm isotropic voxel size. In total, 2125 2D slices of CBCT images were obtained. All CBCT datasets were converted to Digital Imaging and Communication in Medicine (DICOM) format.
The sample size was calculated based on a previous comparable study [29, 30] by a priori power analysis in G*power 3.1.9.7, assuming a paired, two-sided t-test dataset with a power of 80% and significance of 5%.
2.2 Ground truth labelling
Data analysis for ERR detection and labelling were performed by an oral and maxillofacial radiologist with five years of experience analyzing CBCT images and was considered as the ground truth. Each annotation was further classified into four groups of depths. All CBCT data was visualized on a Dell laptop (1920 × 1080 pixels, Dell Latitude E7450; Dell, Austin, TX). The ground truths dataset was prepared by segmenting the CBCT images (DICOM format) using a third-party A.I. tool (Makesense.AI.) [31]. Teeth were grouped as 0 (ERR depth = 0.5 mm), 1 (ERR depth = 1.0 mm), 2 (ERR depth= 2.0 mm), 3 (no ERR).
2.3 AI network architecture and training
2.3.1 Image pre-processing
In Phase 1, the extraction of region of interest (ROI) and image enhancement was performed (Figure 1). A bounding box of 320 × 160 pixels was assigned to all 2D slices, with tooth centered in the box and converted into Portable Network Graphic format. All sagittal slices were used to train and test (ROI) from these bounding boxes. The ROI obtained from a single tooth ranged from 17 to 80 slices resulting in a total of 2125 number of ROI extracted from the CBCT volumes (training and validation 1700, test 425). Then, image intensity was adjusted, and a CLAHE filter was applied before the pre-processing procedure.
2.3.2 Image classification
In Phase 2, four main DLMs (RF+VGG, RF+EFNET, SVM+VGG, and SVM+EFNET) were implemented to classify ERR lesions (Figure 1). Subsequently, all four models were optimized using FST to produce four new enhanced DLMs (FS+RF+VGG, FS+RF+EFNET, FS+SVM+VGG and FS+SVM+EFNET). Two-dimensional CBCT images of ERR were entered into deep CNN models. These images were randomly distributed into training (70%), validation (10%) and test (20%) dataset. Subsequently, the ERR lesions observed in the images were classified as 0, 1,2 or 3 as the output, according to the depth of ERR in the images. In VGG16 and EfficientnetB4 systems, 555,328 and 18,764,579 parameters were utilized (Tables 1 and 2) [27]. Multiclass classification was performed by all models using Tensorflow and Keras phyton deep learning library.
Table 1. VGG16 parameter
Layer (Type)
|
Output shape
|
Parameter
|
Block1_conv1(Conv2D)
|
(None, 320, 160, 64)
|
1792
|
Block1_conv2(Conv2D)
|
(None, 320, 160, 64)
|
36928
|
Block1_pool (MaxPooling2D)
|
(None, 160, 80, 64)
|
0
|
Block1_conv1(Conv2D)
|
(None, 160, 80, 128)
|
73856
|
Block1_conv2(Conv2D)
|
(None, 160, 80, 128)
|
147584
|
Block2_pool (MaxPooling2D)
|
(None, 80, 40, 128)
|
0
|
Block3_conv1 (Conv2D)
|
(None, 80, 40, 256)
|
295168
|
Total parameters: 555328
Trainable parameters: 555328
Non-trainable parameters: 0
Table 2. EfficientnetB4 parameter
Layers (Type)
|
Output Shape
|
Parameters
|
EfficientnetB4 (Functional)
|
(None, 1792)
|
17673823
|
Module_wrapper_4
|
(None, 1792)
|
0
|
Module_wrapper_5
|
(None, 512)
|
918016
|
Module_wrapper_6
|
(None, 256)
|
131328
|
Module_wrapper_7
|
(None, 128)
|
32896
|
Module_wrapper_8
|
(None, 64)
|
8256
|
Module_wrapper_9
|
(None, 4)
|
260
|
Total parameters: 18,764,579
Trainable parameters: 1,090,756
Non-trainable parameters: 17,673,823
2.3.2.1 Performance evaluation
The model's performance was evaluated based on the calculation of accuracy. A confusion matrix summarized the prediction results on a classification task [32]. Five metrics were used to demonstrate the classification models performance: classification accuracy, F1-score, precision, specificity, and error rate [33]. One-way ANOVA was performed to assess the difference between the accuracy of all DLMs. An independent t-test was conducted to assess any significant difference between the results obtained with and without FST. The metric evaluation was performed according to the following formula using confusion matrix in Table 3:
![](data:image/png;base64,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)
Precision= number of correctly detected ERR / (number of correctly detected ERR+ number of falsely detected ERR)
F1-score = 2 × (recall precision)/(recall+presicion)
Error rate = (false positive + false negative)/total population
Table 3. Confusion matrix for binary classification
Data class
|
Classified as Positive
|
Classified as Negative
|
positive
|
true positive (T.P.)
|
false negative (F.N.)
|
negative
|
false positive (F.P.)
|
true negative (TN)
|