Deep Semi-Supervised Learning for Automatic Segmentation of Inferior Alveolar Nerve Using a Convolutional Neural Network (Preprint)

2021 ◽  
Author(s):  
Ho-Kyung Lim ◽  
Seok-Ki Jung ◽  
Seung-Hyun Kim ◽  
Yongwon Cho ◽  
In-Seok Song
Keyword(s):  
Active Learning ◽  
Automatic Segmentation ◽  
Inferior Alveolar Nerve ◽  
Dice Similarity Coefficient ◽  
Clinical Tool ◽  
Lower Lip ◽  
Precise Position ◽  
Visual Scoring ◽  
Mandibular Teeth ◽  
The Times

BACKGROUND The inferior alveolar nerve (IAN) innervates and regulates the sensation of the mandibular teeth and lower lip. The position of the IAN should be monitored during surgery to prevent damage. Therefore, a study using artificial intelligence (AI) was planned to image and track the position of the IAN automatically for a quicker and safer surgery. OBJECTIVE In this study, we segmented the precise position of the IAN using AI. The accuracy of this technique was evaluated by comparing the position with the position manually specified by a specialist, and segmentation accuracy and annotation efficiency were found to be improved with learning. METHODS A total of 138 cone-beam computed tomography datasets (Internal: 98, External: 40) collected from multiple centers (three hospitals) were used in the study. A customized 3D nnU-Net was used for image segmentation. Active learning, which consists of three steps, was carried out in iterations for 83 datasets with cumulative additions after each step. Subsequently, the accuracy of the model for IAN segmentation was evaluated using the residual dataset. We compared the accuracy by deriving the dice similarity coefficient (DSC) value and the segmentation time for each learning step. In addition, visual scoring was considered to comparatively evaluate the manual and automatic segmentation. RESULTS After learning, the DSC gradually increased to 0.48 ± 0.11 to 0.50 ± 0.11, and 0.58 ± 0.08. The DSC for the external dataset was 0.49 ± 0.12. The times required for segmentation were 124.8, 143.4, and 86.4 s, showing a large decrease at the final stage. In visual scoring, the accuracy of manual segmentation was found to be higher than that of automatic segmentation. CONCLUSIONS The deep active learning framework can serve as a fast, accurate, and robust clinical tool for demarcating IAN location.

scholarly journals Deep semi-supervised learning for automatic segmentation of inferior alveolar nerve using a convolutional neural network

2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Ho-Kyung Lim ◽  
Seok-Ki Jung ◽  
Seung-Hyun Kim ◽  
Yongwon Cho ◽  
In-Seok Song
Keyword(s):  
Active Learning ◽  
Automatic Segmentation ◽  
Inferior Alveolar Nerve ◽  
Dice Similarity Coefficient ◽  
Clinical Tool ◽  
Lower Lip ◽  
Learning Framework ◽  
Visual Scoring ◽  
Mandibular Teeth ◽  
The Times

Abstract Background The inferior alveolar nerve (IAN) innervates and regulates the sensation of the mandibular teeth and lower lip. The position of the IAN should be monitored prior to surgery. Therefore, a study using artificial intelligence (AI) was planned to image and track the position of the IAN automatically for a quicker and safer surgery. Methods A total of 138 cone-beam computed tomography datasets (Internal: 98, External: 40) collected from multiple centers (three hospitals) were used in the study. A customized 3D nnU-Net was used for image segmentation. Active learning, which consists of three steps, was carried out in iterations for 83 datasets with cumulative additions after each step. Subsequently, the accuracy of the model for IAN segmentation was evaluated using the 50 datasets. The accuracy by deriving the dice similarity coefficient (DSC) value and the segmentation time for each learning step were compared. In addition, visual scoring was considered to comparatively evaluate the manual and automatic segmentation. Results After learning, the DSC gradually increased to 0.48 ± 0.11 to 0.50 ± 0.11, and 0.58 ± 0.08. The DSC for the external dataset was 0.49 ± 0.12. The times required for segmentation were 124.8, 143.4, and 86.4 s, showing a large decrease at the final stage. In visual scoring, the accuracy of manual segmentation was found to be higher than that of automatic segmentation. Conclusions The deep active learning framework can serve as a fast, accurate, and robust clinical tool for demarcating IAN location.

scholarly journals Coronary artery segmentation under class imbalance using a U-Net based architecture on computed tomography angiography images

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Li-Syuan Pan ◽  
Chia-Wei Li ◽  
Shun-Feng Su ◽  
Shee-Yen Tay ◽  
Quoc-Viet Tran ◽  
...  
Keyword(s):  
Computed Tomography ◽  
Coronary Artery ◽  
Network Architecture ◽  
High Performance ◽  
Promising Result ◽  
Automatic Segmentation ◽  
Class Imbalance ◽  
Dice Similarity Coefficient ◽  
Clinical Tool ◽  
Main Trunk

AbstractCoronary artery disease is caused primarily by vessel narrowing. Extraction of the coronary artery area from images is the preferred procedure for diagnosing coronary diseases. In this study, a U-Net-based network architecture, 3D Dense-U-Net, was adopted to perform fully automatic segmentation of the coronary artery. The network was applied to 474 coronary computed tomography (CT) angiography scans performed at Wanfang Hospital, Taiwan. Of these, 10% were used for testing. The CT scans were divided into patches of 16 original high-resolution slices. The slices were overlapped between patches to take advantage of surrounding imaging information. However, an imbalance between the foreground and background presents a challenge in smaller-object segmentation such as with coronary arteries. The network was optimized and achieved a promising result when the focal loss concept was adopted. To evaluate the accuracy of the automatic segmentation approach, the dice similarity coefficient (DSC) was calculated, and an existing clinical tool was used. The subjective ratings of three experienced radiologists were used to compare the two ratings. The results show that the proposed approach can achieve a DSC of 0.9691, which is significantly higher than other studies using a deep learning approach. In the main trunk, the results of automatic segmentation agree with those of the clinical tool; they were significantly better in some small branches. In our study, automatic segmentation tool shows high-performance detection in coronary lumen vessels, thereby providing potential power in assisting clinical diagnosis.

scholarly journals Anesthetic Efficacy of a Combination of 0.9 M Mannitol Plus 68.8 mg of Lidocaine With 50 μg Epinephrine in Inferior Alveolar Nerve Blocks: A Prospective Randomized, Single Blind Study

2013 ◽  
Vol 60 (4) ◽  
pp. 145-152 ◽  
Author(s):  
Howard Cohen ◽  
Al Reader ◽  
Melissa Drum ◽  
John Nusstein ◽  
Mike Beck
Keyword(s):  
Inferior Alveolar Nerve ◽  
Blind Study ◽  
Maximum Output ◽  
Nerve Blocks ◽  
Posterior Teeth ◽  
The Subject ◽  
Mandibular Teeth ◽  
The Times ◽  
Pulpal Anesthesia ◽  
Single Blind

Abstract The purpose of this prospective randomized, single blind study was to determine the anesthetic efficacy of 68.8 mg of lidocaine with 50 μg epinephrine compared to 68.8 mg lidocaine with 50 μg epinephrine plus 0.9 M mannitol in inferior alveolar nerve (IAN) blocks. Forty subjects randomly received 2 IAN blocks consisting of a 1.72-mL formulation of 68.8 mg lidocaine with 50 μg epinephrine and a 5-mL formulation of 68.8 mg lidocaine with 50 μg epinephrine (1.72 mL) plus 0.9 M mannitol (3.28 mL) in 2 separate appointments spaced at least 1 week apart. Mandibular anterior and posterior teeth were blindly electric pulp tested at 4-minute cycles for 60 minutes postinjection. No response from the subject to the maximum output (80 reading) of the pulp tester was used as the criterion for pulpal anesthesia. Total percent pulpal anesthesia was defined as the total of all the times of pulpal anesthesia (80 readings), for each tooth, over the 60 minutes. One hundred percent of the subjects had profound lip numbness with both inferior alveolar nerve blocks. The results demonstrated that the 5 mL-formulation of 68.8 mg lidocaine with 50 μg epinephrine plus 0.9 M mannitol was significantly better than the 1.72-mL formulation of 68.8 mg lidocaine with 50 μg epinephrine for all teeth, except the lateral incisor. We concluded that adding 0.9 M mannitol to a lidocaine with epinephrine formulation was significantly more effective in achieving a greater percentage of total pulpal anesthesia (as defined in this study) than a lidocaine formulation without mannitol. However, the 0.9 M mannitol/lidocaine formulation would not provide 100% pulpal anesthesia for all the mandibular teeth.

scholarly journals Using spatial-temporal ensembles of convolutional neural networks for lumen segmentation in ureteroscopy

2021 ◽  
Author(s):  
Jorge F. Lazo ◽  
Aldo Marzullo ◽  
Sara Moccia ◽  
Michele Catellani ◽  
Benoit Rosa ◽  
...  
Keyword(s):  
Neural Networks ◽  
Convolutional Neural Networks ◽  
State Of The Art ◽  
Automatic Segmentation ◽  
Temporal Information ◽  
Invasive Technique ◽  
Dice Similarity Coefficient ◽  
Specular Reflections ◽  
Lumen Segmentation ◽  
Previous State

Abstract Purpose Ureteroscopy is an efficient endoscopic minimally invasive technique for the diagnosis and treatment of upper tract urothelial carcinoma. During ureteroscopy, the automatic segmentation of the hollow lumen is of primary importance, since it indicates the path that the endoscope should follow. In order to obtain an accurate segmentation of the hollow lumen, this paper presents an automatic method based on convolutional neural networks (CNNs). Methods The proposed method is based on an ensemble of 4 parallel CNNs to simultaneously process single and multi-frame information. Of these, two architectures are taken as core-models, namely U-Net based in residual blocks ($$m_1$$ m 1 ) and Mask-RCNN ($$m_2$$ m 2 ), which are fed with single still-frames I(t). The other two models ($$M_1$$ M 1 , $$M_2$$ M 2 ) are modifications of the former ones consisting on the addition of a stage which makes use of 3D convolutions to process temporal information. $$M_1$$ M 1 , $$M_2$$ M 2 are fed with triplets of frames ($$I(t-1)$$ I ( t - 1 ) , I(t), $$I(t+1)$$ I ( t + 1 ) ) to produce the segmentation for I(t). Results The proposed method was evaluated using a custom dataset of 11 videos (2673 frames) which were collected and manually annotated from 6 patients. We obtain a Dice similarity coefficient of 0.80, outperforming previous state-of-the-art methods. Conclusion The obtained results show that spatial-temporal information can be effectively exploited by the ensemble model to improve hollow lumen segmentation in ureteroscopic images. The method is effective also in the presence of poor visibility, occasional bleeding, or specular reflections.

scholarly journals Dental implant placement with inferior alveolar nerve repositioning in severely resorbed mandibles: a retrospective multicenter study of implant success and survival rates, and lower lip sensory disturbances

2021 ◽  
Vol 7 (1) ◽  
Author(s):  
George Deryabin ◽  
Simonas Grybauskas
Keyword(s):  
Dental Implants ◽  
Dental Implant ◽  
Healing Process ◽  
Survival Rates ◽  
Inferior Alveolar Nerve ◽  
Reliable Technique ◽  
Lower Lip ◽  
Neurosensory Disturbances ◽  
Sensory Disturbances

Abstract Background The purpose of this study was to analyze medium-to-long-term implant success and survival rates, and lower lip sensory disturbance after placement of dental implants with simultaneous inferior alveolar nerve (IAN) repositioning. Methods Fifteen patients (3 men, 12 women) treated in two centers were included in this retrospective study. The ages of the participants ranged from 19 to 68. A total of 48 dental implants were placed in 23 posterior mandibular segments simultaneously with IAN transposition or lateralization. The residual bone above the IAN ranged from 0.5 to 7.0 mm. Crestal bone changes were measured using cone beam computed tomography (CBCT) images. Disturbance of the IAN was evaluated subjectively using a modified questionnaire. Results The healing process was uneventful in fourteen patients. In one patient, spontaneous fracture of the operated mandible occurred on tenth day after the surgery. The implant in the fracture line was removed at the time of open reduction and fixation. One more implant was lost after 5 years of loading. Therefore, the overall dental implant survival rate was 95.8%, whereas all implants in function were judged as successful after a follow-up period of 1 to 10 years. Transient neurosensory disturbances (ND) were observed in all patients who underwent IAN lateralization and IAN transposition. At follow-up times of 3 years, 5 years, and 10 years, weak hypoesthesia remained in two subjects treated with IAN transposition. None of the patients developed neuropathic pain after the procedure. Conclusions Within the limitations of this study, we conclude that reconstruction of severely resorbed mandibles with dental implants in conjunction with IAN repositioning is an effective and reliable technique. Although neurosensory disturbances are the most common complication after surgery, they tend to resolve over time. Advanced surgical skills are required to perform this technique.

scholarly journals Quantitative salivary gland SPECT/CT using deep convolutional neural networks

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Junyoung Park ◽  
Jae Sung Lee ◽  
Dongkyu Oh ◽  
Hyun Gee Ryoo ◽  
Jeong Hee Han ◽  
...  
Keyword(s):  
Computed Tomography ◽  
Salivary Gland ◽  
Single Photon ◽  
Automatic Segmentation ◽  
Photon Emission ◽  
Operating Time ◽  
Emission Computed Tomography ◽  
Dice Similarity Coefficient ◽  
Deep Convolutional Neural Networks ◽  
Gland Function

AbstractQuantitative single-photon emission computed tomography/computed tomography (SPECT/CT) using Tc-99m pertechnetate aids in evaluating salivary gland function. However, gland segmentation and quantitation of gland uptake is challenging. We develop a salivary gland SPECT/CT with automated segmentation using a deep convolutional neural network (CNN). The protocol comprises SPECT/CT at 20 min, sialagogue stimulation, and SPECT at 40 min post-injection of Tc-99m pertechnetate (555 MBq). The 40-min SPECT was reconstructed using the 20-min CT after misregistration correction. Manual salivary gland segmentation for %injected dose (%ID) by human experts proved highly reproducible, but took 15 min per scan. An automatic salivary segmentation method was developed using a modified 3D U-Net for end-to-end learning from the human experts (n = 333). The automatic segmentation performed comparably with human experts in voxel-wise comparison (mean Dice similarity coefficient of 0.81 for parotid and 0.79 for submandibular, respectively) and gland %ID correlation (R2 = 0.93 parotid, R2 = 0.95 submandibular) with an operating time less than 1 min. The algorithm generated results that were comparable to the reference data. In conclusion, with the aid of a CNN, we developed a quantitative salivary gland SPECT/CT protocol feasible for clinical applications. The method saves analysis time and manual effort while reducing patients’ radiation exposure.

scholarly journals Anesthetic Efficacy of a Combination of 0.5 M Mannitol Plus 127.2 mg of Lidocaine With 50 μg Epinephrine in Inferior Alveolar Nerve Blocks: A Prospective Randomized, Single-Blind Study

2013 ◽  
Vol 60 (1) ◽  
pp. 3-10 ◽  
Author(s):  
Steven Smith ◽  
Al Reader ◽  
Melissa Drum ◽  
John Nusstein ◽  
Mike Beck
Keyword(s):  
Postoperative Pain ◽  
Inferior Alveolar Nerve ◽  
Blind Study ◽  
Maximum Output ◽  
Nerve Blocks ◽  
Posterior Teeth ◽  
Solution Deposition ◽  
The Times ◽  
Pulpal Anesthesia ◽  
Single Blind

Abstract The purpose of this prospective, randomized, single-blind study was to determine the anesthetic efficacy of 127.2 mg lidocaine with 50 μg epinephrine compared to 127.2 mg lidocaine with 50 μg epinephrine plus 0.5 M mannitol in inferior alveolar nerve (IAN) blocks. Forty subjects randomly received 2 IAN blocks consisting of a 3.18 mL formulation of 127.2 mg lidocaine with 50 μg epinephrine and a 5 mL formulation of 127.2 mg lidocaine with 50 μg epinephrine (3.18 mL) plus 0.5 M mannitol (1.82 mL) in 2 separate appointments spaced at least 1 week apart. Mandibular anterior and posterior teeth were blindly electric pulp tested at 4-minute cycles for 60 minutes postinjection. Pain of solution deposition and postoperative pain were also measured. No response from the subject to the maximum output (80 reading) of the pulp tester was used as the criterion for pulpal anesthesia. Total percent pulpal anesthesia was defined as the total of all the times of pulpal anesthesia (80 readings) over the 60 minutes. One hundred percent of the subjects had profound lip numbness with both inferior alveolar nerve blocks. The results demonstrated that a 5 mL formulation of 127.2 mg lidocaine with 50 μg epinephrine plus 0.5 M mannitol was significantly better than the 3.18 mL formulation of 127.2 mg lidocaine with 50 μg epinephrine for all teeth. Solution deposition pain and postoperative pain were not statistically different between the lidocaine/mannitol formulation and the lidocaine formulation without mannitol. We concluded that adding 0.5 M mannitol to a lidocaine with epinephrine formulation was significantly more effective in achieving a greater percentage of total pulpal anesthesia than a lidocaine formulation without mannitol.

scholarly journals Orthodontic camouflage of skeletal Class III malocclusion with miniplate: a case report

2016 ◽  
Vol 21 (4) ◽  
pp. 89-98 ◽  
Author(s):  
Marcel Marchiori Farret ◽  
Milton M. Benitez Farret ◽  
Alessandro Marchiori Farret
Keyword(s):  
Orthognathic Surgery ◽  
Class Iii ◽  
Class Iii Malocclusion ◽  
Lower Lip ◽  
Facial Profile ◽  
Skeletal Class ◽  
Mandibular Teeth ◽  
One Year ◽  
Angle Class Iii Malocclusion

ABSTRACT Introduction: Skeletal Class III malocclusion is often referred for orthodontic treatment combined with orthognathic surgery. However, with the aid of miniplates, some moderate discrepancies become feasible to be treated without surgery. Objective: To report the case of a 24-year-old man with severe skeletal Angle Class III malocclusion with anterior crossbite and a consequent concave facial profile. Methods: The patient refused to undergo orthognathic surgery; therefore, orthodontic camouflage treatment with the aid of miniplates placed on the mandibular arch was proposed. Results: After 18 months of treatment, a Class I molar and canine relationship was achieved, while anterior crossbite was corrected by retraction of mandibular teeth. The consequent decrease in lower lip fullness and increased exposure of maxillary incisors at smiling resulted in a remarkable improvement of patient's facial profile, in addition to an esthetically pleasing smile, respectively. One year later, follow-up revealed good stability of results.

Sign in / Sign up

Export Citation Format

Share Document