Metal Artifact Reduction in Dental CBCT Images Using Direct Sinogram Correction Combined with Metal Path-Length Weighting

Dual-encoder architecture for metal artifact reduction for kV-cone-beam CT images in head and neck cancer radiotherapy

During a radiotherapy (RT) course, geometrical variations of target volumes, organs at risk, weight changes (loss/gain), tumor regression and/or progression can significantly affect the treatment outcome. Adaptive RT has become the effective methods along with technical advancements in imaging modalities including cone-beam computed tomography (CBCT). Planning CT (pCT) can be modified via deformable image registration (DIR), which is applied to the pair of pCT and CBCT. However, the artifact existed in both pCT and CBCT is a vulnerable factor in DIR. The dose calculation on CBCT is also suggested. Missing information due to the artifacts hinders the accurate dose calculation on CBCT. In this study, we aim to develop a deep learning-based metal artifact reduction (MAR) model to reduce the metal artifacts in CBCT for head and neck cancer RT. To train the proposed MAR model, we synthesized the kV-CBCT images including metallic implants, with and without metal artifacts (simulated image data pairs) through sinogram image handling process. We propose the deep learning architecture which focuses on both artifact removal and reconstruction of anatomic structure using a dual-encoder architecture. We designed four single-encoder models and three dual-encoder models based on UNet (for an artifact removal) and FusionNet (for a tissue restoration). Each single-encoder model contains either UNet or FusionNet, while the dual-encoder models have both UNet and FusionNet architectures. In the dual-encoder models, we implemented different feature fusion methods, including simple addition, spatial attention, and spatial/channel wise attention. Among the models, a dual-encoder model with spatial/channel wise attention showed the highest scores in terms of peak signal-to-noise ratio, mean squared error, structural similarity index, and Pearson correlation coefficient. CBCT images from 34 head and neck cancer patients were used to test the developed models. The dual-encoder model with spatial/channel wise attention showed the best results in terms of artifact index. By using the proposed model to CBCT, one can achieve more accurate synthetic pCT for head and neck patients as well as better tissue recognition and structure delineation for CBCT image itself.

Exploratory study on a novel minimally invasive tunnel like flap approach for orthodontic movement of dental implants

OBJECTIVES

To propose a new flap design that utilize a minimally invasive approach to move implants. We also propose a 3D method to track changes in implant position without the need for CBCT.

MATERIALS AND METHODS

Implants were inserted in two mongrel dogs then a minimally invasive tunnel like flap was made. Bone cutting around implants was done using piezotome device and orthodontic force was applied on the implants. For tracking positional changes of implants, an impression was taken then scanned with an identical implant copy inside. The pre- and post-treatment 3D images were then superimposed over each other to detect changes in implant position.

RESULTS

Healing of the flap was rapid with minimal swelling and edema. The position of the interdental papilla was preserved maintaining optimal esthetic outcome. Implant pocket depth was maintained. Our 3D imaging method was able to accurately detect the movement that occurred in the dental implants after their movement with significantly higher resolution than CBCT (6 μm for this method versus 76-300 μm for the CBCT).

CONCLUSIONS

This flap can preserve the esthetic soft tissue contour of implants during orthodontic movement allowing rapid healing with minimum swelling and edema. Moreover, our proposed 3D method has many advantages over CBCT in tracking the movement of implants by being non harmful and cheap thus allowing several follow up points.

CLINICAL RELEVANCE

Orthodontic movement of implants can be required in cases of improperly placed implants or to intentionally place the implant in a certain position then altering it to avoid the need for bone grafting.

Performance of different cone-beam computed tomography scan modes with and without metal artifact reduction in detection of recurrent dental caries under various restorative materials

Purpose

We aimed to compare the diagnostic performance of different cone-beam computed tomography (CBCT) scan modes with and without the application of a metal artifact reduction (MAR) option under 5 different restorative materials.

Material and methods

Our research was an in vitro study with 150 caries-free premolars and molars. The teeth were randomly divided into experimental (with artificially induced caries, n = 75) and control (without caries, n = 75) groups and were prepared based on 5 types of restorative materials, including conventional composites (Filtek Z250, Gradia), flow composite, glass ionomer, and amalgam. The teeth were examined under 2 CBCT scan modes (high-resolution [HIRes] and standard) with and without MAR application. Finally, the diagnostic accuracy index values (area under the receiver operating characteristic curve [AUC], sensitivity, and specificity) were calculated.

Results

The AUC of standard scan mode with the MAR option was significantly lower than that of HIRes with MAR (p = 0.018) and without MAR option (p = 0.011) in detecting recurrent caries. Also, without MAR option, the diagnostic accuracy (AUC) of the standard mode was significantly lower than that of the HIRes (p = 0.020). Similar findings were observed for sensitivity and specificity. Moreover, diagnostic performance of standard and HIRes scan modes with and without MAR in the amalgam group was lower than that in other restorative material groups.

Conclusions

Diagnostic performance of HIRes CBCT mode was higher than that of standard mode for recurrent caries and remained unaffected by MAR application. However, the accuracy in detecting recurrent caries was lower in the amalgam group compared with other restorative material groups.

Do metal artifact reduction algorithms influence the detection of implant-related injuries to the inferior alveolar canal in CBCT images?

BACKGROUND

The routine application of dental implants for replacing missing teeth has revolutionized restorative and prosthetic dentistry. However, cone beam computed tomography (CBCT) evaluations of structures adjacent to the implants are limited by metal artifacts. There are several methods for reducing metal artifacts, but this remains a challenging task. This study aimed to examine the effectiveness of metal artifact reduction (MAR) algorithms in identifying injuries of implants to the inferior alveolar canal in CBCT images.

METHOD

In this in vitro study, mono-cortical bone windows were created and the inferior alveolar canal was revealed. Using 36 implants, pilot drill and penetration damage of the implant tip into the canal was simulated and compared to the control implants with distance from the canal. CBCT images were evaluated by four experienced observers with and without the MAR algorithm and compared to direct vision as the gold standard. The values of accuracy, sensitivity, and specificity were obtained and compared by receiver operating characteristic (ROC) curve (α = 0.05).

RESULT

The area under the ROC curve values for detection of pilot drill injuries varied between 0.840-0.917 and 0.639-0.854 in the active and inactive MAR conditions, respectively. The increase in ROC area was only significant for one of the observers (P = 0.010). For diagnosing penetrative injuries, the area under the ROC curve values was between 0.990-1.000 and 0.722-1.000 in the active and inactive MAR conditions, respectively. The improvement of ROC curve values in active MAR mode was only significant for one of the observers (P = 0.006).

CONCLUSION

Activation of MAR improved the diagnostic values of CBCT images in detecting both types of implant-related injuries to the inferior alveolar canal. However, for most observers, this increase was not statistically significant.