Automatic Segmentation of Mandible from Conventional Methods to Deep Learning-A Review.

Medical imaging techniques, such as (cone beam) computed tomography and magnetic resonance imaging, have proven to be a valuable component for oral and maxillofacial surgery (OMFS). Accurate segmentation of the mandible from head and neck (H&N) scans is an important step in order to build a personalized 3D digital mandible model for 3D printing and treatment planning of OMFS. Segmented mandible structures are used to effectively visualize the mandible volumes and to evaluate particular mandible properties quantitatively. However, mandible segmentation is always challenging for both clinicians and researchers, due to complex structures and higher attenuation materials, such as teeth (filling) or metal implants that easily lead to high noise and strong artifacts during scanning. Moreover, the size and shape of the mandible vary to a large extent between individuals. Therefore, mandible segmentation is a tedious and time-consuming task and requires adequate training to be performed properly. With the advancement of computer vision approaches, researchers have developed several algorithms to automatically segment the mandible during the last two decades. The objective of this review was to present the available fully (semi)automatic segmentation methods of the mandible published in different scientific articles. This review provides a vivid description of the scientific advancements to clinicians and researchers in this field to help develop novel automatic methods for clinical applications.

Mandible Segmentation of Dental CBCT Scans Affected by Metal Artifacts Using Coarse-to-Fine Learning Model.

Accurate segmentation of the mandible from cone-beam computed tomography (CBCT) scans is an important step for building a personalized 3D digital mandible model for maxillofacial surgery and orthodontic treatment planning because of the low radiation dose and short scanning duration. CBCT images, however, exhibit lower contrast and higher levels of noise and artifacts due to extremely low radiation in comparison with the conventional computed tomography (CT), which makes automatic mandible segmentation from CBCT data challenging. In this work, we propose a novel coarse-to-fine segmentation framework based on 3D convolutional neural network and recurrent SegUnet for mandible segmentation in CBCT scans. Specifically, the mandible segmentation is decomposed into two stages: localization of the mandible-like region by rough segmentation and further accurate segmentation of the mandible details. The method was evaluated using a dental CBCT dataset. In addition, we evaluated the proposed method and compared it with state-of-the-art methods in two CT datasets. The experiments indicate that the proposed algorithm can provide more accurate and robust segmentation results for different imaging techniques in comparison with the state-of-the-art models with respect to these three datasets.

Recurrent Convolutional Neural Networks for 3D Mandible Segmentation in Computed Tomography.

PURPOSE: Classic encoder-decoder-based convolutional neural network (EDCNN) approaches cannot accurately segment detailed anatomical structures of the mandible in computed tomography (CT), for instance, condyles and coronoids of the mandible, which are often affected by noise and metal artifacts. The main reason is that EDCNN approaches ignore the anatomical connectivity of the organs. In this paper, we propose a novel CNN-based 3D mandible segmentation approach that has the ability to accurately segment detailed anatomical structures. METHODS: Different from the classic EDCNNs that need to slice or crop the whole CT scan into 2D slices or 3D patches during the segmentation process, our proposed approach can perform mandible segmentation on complete 3D CT scans. The proposed method, namely, RCNNSeg, adopts the structure of the recurrent neural networks to form a directed acyclic graph in order to enable recurrent connections between adjacent nodes to retain their connectivity. Each node then functions as a classic EDCNN to segment a single slice in the CT scan. Our proposed approach can perform 3D mandible segmentation on sequential data of any varied lengths and does not require a large computation cost. The proposed RCNNSeg was evaluated on 109 head and neck CT scans from a local dataset and 40 scans from the PDDCA public dataset. The final accuracy of the proposed RCNNSeg was evaluated by calculating the Dice similarity coefficient (DSC), average symmetric surface distance (ASD), and 95% Hausdorff distance (95HD) between the reference standard and the automated segmentation. RESULTS: The proposed RCNNSeg outperforms the EDCNN-based approaches on both datasets and yields superior quantitative and qualitative performances when compared to the state-of-the-art approaches on the PDDCA dataset. The proposed RCNNSeg generated the most accurate segmentations with an average DSC of 97.48%, ASD of 0.2170 mm, and 95HD of 2.6562 mm on 109 CT scans, and an average DSC of 95.10%, ASD of 0.1367 mm, and 95HD of 1.3560 mm on the PDDCA dataset. CONCLUSIONS: The proposed RCNNSeg method generated more accurate automated segmentations than those of the other classic EDCNN segmentation techniques in terms of quantitative and qualitative evaluation. The proposed RCNNSeg has potential for automatic mandible segmentation by learning spatially structured information.

Robust and Accurate Mandible Segmentation on Dental CBCT Scans Affected by Metal Artifacts Using a Prior Shape Model.

Accurate mandible segmentation is significant in the field of maxillofacial surgery to guide clinical diagnosis and treatment and develop appropriate surgical plans. In particular, cone-beam computed tomography (CBCT) images with metal parts, such as those used in oral and maxillofacial surgery (OMFS), often have susceptibilities when metal artifacts are present such as weak and blurred boundaries caused by a high-attenuation material and a low radiation dose in image acquisition. To overcome this problem, this paper proposes a novel deep learning-based approach (SASeg) for automated mandible segmentation that perceives overall mandible anatomical knowledge. SASeg utilizes a prior shape feature extractor (PSFE) module based on a mean mandible shape, and recurrent connections maintain the continuity structure of the mandible. The effectiveness of the proposed network is substantiated on a dental CBCT dataset from orthodontic treatment containing 59 patients. The experiments show that the proposed SASeg can be easily used to improve the prediction accuracy in a dental CBCT dataset corrupted by metal artifacts. In addition, the experimental results on the PDDCA dataset demonstrate that, compared with the state-of-the-art mandible segmentation models, our proposed SASeg can achieve better segmentation performance.

A Standardized Framework for Fluorescence-Guided Margin Assessment for Head and Neck Cancer Using a Tumor Acidosis Sensitive Optical Imaging Agent.

PURPOSE: Intra-operative management of the surgical margin in patients diagnosed with head and neck squamous cell carcinoma (HNSCC) remains challenging as surgeons still have to rely on visual and tactile information. Fluorescence-guided surgery using tumor-specific imaging agents can assist in clinical decision-making. However, a standardized imaging methodology is lacking. In this study, we determined whether a standardized, specimen-driven, fluorescence imaging framework using ONM-100 could assist in clinical decision-making during surgery. PROCEDURES: Thirteen patients with histologically proven HNSCC were included in this clinical study and received ONM-100 24 ± 8 h before surgery. Fluorescence images of the excised surgical specimen and of the surgical cavity were analyzed. A fluorescent lesion with a tumor-to-background ratio (TBR) > 1.5 was considered fluorescence-positive and correlated to standard of care (SOC) histopathology. RESULTS: All six tumor-positive surgical margins were detected immediately after excision using fluorescence-guided intra-operative imaging. Postoperative analysis showed a median TBR (±IQR) of the fluorescent lesions on the resection margin of 3.36 ± 1.62. Three fluorescence-positive lesions in the surgical cavity were biopsied and showed occult carcinoma and severe dysplasia, and a false-positive fluorescence lesion. CONCLUSION: Our specimen-driven fluorescence framework using a novel, pH-activatable, fluorescent imaging agent could assist in reliable and real-time adequate clinical decision-making showing that a fluorescent lesion on the surgical specimen with a TBR of 1.5 is correlated to a tumor-positive resection margin. The binary mechanism of ONM-100 allows for a sharp tumor delineation in all patients, giving the surgeon a clinical tool for real-time margin assessment, with a high sensitivity.

Improving oral cavity cancer diagnosis and treatment with fluorescence molecular imaging.

Early diagnosis and radical surgical excision of oral squamous cell carcinomas are essential for achieving optimal treatment outcomes. To date, diagnostic tools that rely on anatomical anomalies provide limited information and resolution in clinical practice. As a result, oral cancer is often detected in an advanced stage. Also, no reliable real-time intraoperative tools are readily available for the evaluation of surgical resection margins. Fluorescence imaging visualises biological processes that occur in early carcinogenesis and could, therefore, enable detection of small tumours in early stages. Furthermore, due to the high sensitivity and spatial resolution, fluorescence imaging could assist in resection margin assessment during surgery. In this review, we discuss several techniques that employ fluorescence for early diagnosis and surgical guidance in oral squamous cell carcinoma and present future perspectives on the potential of fluorescence imaging in oral cancer in the near future.

Fluorescence-guided imaging for resection margin evaluation in head and neck cancer patients using cetuximab-800CW: A quantitative dose-escalation study.

Tumor-positive resection margins are present in up to 23% of head and neck cancer (HNC) surgeries, as intraoperative techniques for real-time evaluation of the resection margins are lacking. In this study, we investigated the safety and potential clinical value of fluorescence-guided imaging (FGI) for resection margin evaluation in HNC patients. We determined the optimal cetuximab-800CW dose by quantification of intrinsic fluorescence values using multi-diameter single-fiber reflectance, single-fiber fluorescence (MDSFR/SFF) spectroscopy. Methods: Five cohorts of three HNC patients received cetuximab-800CW systemically: three single dose cohorts (10, 25, 50 mg) and two cohorts pre-dosed with 75 mg unlabeled cetuximab (15 or 25 mg). Fluorescence visualization and MDSFR/SFF spectroscopy quantification was performed and were correlated to histopathology. Results: There were no study-related adverse events higher than Common Terminology Criteria for Adverse Events grade-II. Quantification of intrinsic fluorescence values showed a dose-dependent increase in background fluorescence in the single dose cohorts (p<0.001, p<0.001), which remained consistently low in the pre-dosed cohorts (p=0.6808). Resection margin status was evaluated with a sensitivity of 100% (4/4 tumor-positive margins) and specificity of 91% (10/11 tumor-negative margins). Conclusion: A pre-dose of 75 mg unlabeled cetuximab followed by 15 mg cetuximab-800CW was considered the optimal dose based on safety, fluorescence visualization and quantification of intrinsic fluorescence values. We were able to use a lower dose cetuximab-800CW than previously described, while remaining a high sensitivity for tumor detection due to application of equipment optimized for IRDye800CW detection, which was validated by quantification of intrinsic fluorescence values.