scholarly journals Deep Convolution Neural Network for Laryngeal Cancer Classification on Contact Endoscopy-Narrow Band Imaging

Sensors ◽  
2021 ◽  
Vol 21 (23) ◽  
pp. 8157
Author(s):  
Nazila Esmaeili ◽  
Esam Sharaf ◽  
Elmer Jeto Gomes Ataide ◽  
Alfredo Illanes ◽  
Axel Boese ◽  
...  
Keyword(s):  
Laryngeal Cancer ◽  
High Performance ◽  
Performance Metrics ◽  
Narrow Band ◽  
Narrow Band Imaging ◽  
Visual Assessment ◽  
Fine Tuning ◽  
Deep Convolutional Neural Networks ◽  
Laryngeal Mucosa ◽  
Unseen Data

(1) Background: Contact Endoscopy (CE) and Narrow Band Imaging (NBI) are optical imaging modalities that can provide enhanced and magnified visualization of the superficial vascular networks in the laryngeal mucosa. The similarity of vascular structures between benign and malignant lesions causes a challenge in the visual assessment of CE-NBI images. The main objective of this study is to use Deep Convolutional Neural Networks (DCNN) for the automatic classification of CE-NBI images into benign and malignant groups with minimal human intervention. (2) Methods: A pretrained Res-Net50 model combined with the cut-off-layer technique was selected as the DCNN architecture. A dataset of 8181 CE-NBI images was used during the fine-tuning process in three experiments where several models were generated and validated. The accuracy, sensitivity, and specificity were calculated as the performance metrics in each validation and testing scenario. (3) Results: Out of a total of 72 trained and tested models in all experiments, Model 5 showed high performance. This model is considerably smaller than the full ResNet50 architecture and achieved the testing accuracy of 0.835 on the unseen data during the last experiment. (4) Conclusion: The proposed fine-tuned ResNet50 model showed a high performance to classify CE-NBI images into the benign and malignant groups and has the potential to be part of an assisted system for automatic laryngeal cancer detection.

Endoscopic diagnosis of laryngeal cancer and precancerous lesions by narrow band imaging

2010 ◽  
Vol 125 (3) ◽  
pp. 288-296 ◽  
Author(s):  
X-G Ni ◽  
S He ◽  
Z-G Xu ◽  
L Gao ◽  
N Lu ◽  
...  
Keyword(s):  
Sensitivity And Specificity ◽  
Laryngeal Cancer ◽  
Narrow Band ◽  
Narrow Band Imaging ◽  
Precancerous Lesions ◽  
Type I ◽  
Severe Dysplasia ◽  
Laryngeal Lesions ◽  
Type V

AbstractObjective:To investigate the characteristics of the laryngeal mucosal microvascular network in suspected laryngeal cancer patients, using narrow band imaging, and to evaluate the value of narrow band imaging endoscopy in the early diagnosis of laryngeal precancerous and cancerous lesions.Patients and methods:Eighty-five consecutive patients with suspected precancerous or cancerous laryngeal lesions were enrolled in the study. Endoscopic narrow band imaging findings were classified into five types (I to V) according to the features of the mucosal intraepithelial papillary capillary loops assessed.Results:A total of 104 lesions (45 malignancies and 59 nonmalignancies) was detected under white light and narrow band imaging modes. The sensitivity and specificity of narrow band imaging in detecting malignant lesions were 88.9 and 93.2 per cent, respectively. The intraepithelial papillary capillary loop classification, as determined by narrow band imaging, was closely associated with the laryngeal lesions' histological findings. Type I to IV lesions were considered nonmalignant and type V lesions malignant. For type Va lesions, the sensitivity and specificity of narrow band imaging in detecting severe dysplasia or carcinoma in situ were 100 and 79.5 per cent, respectively. In patients with type Vb and Vc lesions, the sensitivity and specificity of narrow band imaging in detecting invasive carcinoma were 83.8 and 100 per cent, respectively.Conclusion:Narrow band imaging is a promising approach enabling in vivo differentiation of nonmalignant from malignant laryngeal lesions by evaluating the morphology of mucosal capillaries. These results suggest endoscopic narrow band imaging may be useful in the early detection of laryngeal cancer and precancerous lesions.

scholarly journals Deep Learning Applied to White Light and Narrow Band Imaging Videolaryngoscopy: Toward Real‐Time Laryngeal Cancer Detection

2021 ◽  
Author(s):  
Muhammad Adeel Azam ◽  
Claudio Sampieri ◽  
Alessandro Ioppi ◽  
Stefano Africano ◽  
Alberto Vallin ◽  
...  
Keyword(s):  
Deep Learning ◽  
Real Time ◽  
Laryngeal Cancer ◽  
Cancer Detection ◽  
White Light ◽  
Narrow Band ◽  
Narrow Band Imaging

S195 – Laryngeal Observation of Early Cancer with NBI System

2008 ◽  
Vol 139 (2_suppl) ◽  
pp. P252-P141
Author(s):  
Masahiko Higashikawa
Keyword(s):  
Laryngeal Cancer ◽  
Narrow Band ◽  
Early Cancer ◽  
Narrow Band Imaging ◽  
Ktp Laser ◽  
Glottic Carcinoma ◽  
Mucosal Edema ◽  
Illumination Method ◽  
False Vocal ◽  
Post Therapy

Objectives The Narrow Band Imaging (NBI) is an illumination method for medical endoscopes that can visualize the micro-vascular structure of the organ surface. The effectiveness of NBI has been reported in detecting the oropharyngeal and hypopharyngeal neoplasm lesion. The aim of this study is to identify the usefulness of NBI in laryngeal lesion, especially in observing cases of post-operated or post-irradiated early laryngeal cancers. Methods 16 patients of early laryngeal cancer after treatment for at least 6 months were entered in this study: 1 case of T1s, 8 cases of T1a, 3 cases of T1b, and 4 cases of T2. 3 cases of T1a underwent cordectomy under direct laryngoscope using KTP laser. 13 cases were treated with irradiation. Results The cases for which NBI was suggested to be useful were: T1b, undergoing laser surgery, supra-glottic carcinoma, appearing hyper-adduction of the false vocal fold during phonating, and causing severe mucosal edema after irradiation. Conclusions NBI system may play an important role in the observation of post-therapy of early laryngeal cancer.

THE VALUE OF NARROW BAND IMAGING ENDOSCOPY IN DIAGNOSIS OF HYPOPHARYNGEAL AND LARYNGEAL CANCER AND FOLLOWING - UP POST TREATMENT

2018 ◽  
Vol 8 (6) ◽  
pp. 114-122
Author(s):  
Thong Le Chi ◽  
Thanh Dang ◽  
Nam Tran Phuong
Keyword(s):  
Laryngeal Cancer ◽  
Narrow Band ◽  
Narrow Band Imaging ◽  
Hypopharyngeal Cancer ◽  
Post Treatment ◽  
Predominant Type ◽  
Mucosal Edema ◽  
Contrast Medium Injection ◽  
The Mean ◽  
Type V

Background: To evaluate the value of narrow band imaging (NBI) endosocopy in diagnosis of hypopharyngeal and laryngeal cancer and following – up post treatment. Material and methods: A total of 75 patients included 36 patients with hypopharyngeal cancer and 39 patients with laryngeal cancer who had diagnosed at Department of Otoloryngology – Hue Central Hospital from 5/2017 to 5/2018. A prospective cohort study was conducted. Results: The age group 51 - 60 years occurred most often, 33.3%, the mean age was 62.1 ± 13.4. The UICC stage III was 65.3%. Tumor was in ulcerlarative and infiltrate form (89.4%), edema and inflammation of magrin tumor (41.3%), invasive (58.7%). Intrapapillary capillary loops – IPCL - type V was predominant, type V-n was 46.7%. The tumor with IPCL type V-n had strong enhancement (51.3%) and moderate enhancement (44.4%) after contrast medium injection on CT scan. One month after treatment, there were 33.3% of tumor – free, 53.7% of mucosal edema and 13% tumor size-decreasing on NBI image. Conclusion: NBI endoscopy is an useful tool for diagnosing of hypopharyngeal and laryngeal cancer and following – up post treatment. Key words: narrow band imaging endoscopy, hypopharyngeal cancer, laryngeal cancer

Narrow band imaging in transoral laser microsurgery (TLM) in moderately advanced (T2, T3) glottic cancer

2018 ◽  
Vol 72 (5) ◽  
pp. 17-23 ◽  
Author(s):  
Krzysztof Piersiala ◽  
Hanna Klimza ◽  
Joanna Jackowska ◽  
Anna Majewska ◽  
Małgorzata Wierzbicka
Keyword(s):  
Laryngeal Cancer ◽  
Predictive Value ◽  
Narrow Band Imaging ◽  
Intraoperative Imaging ◽  
Glottic Cancer ◽  
Resection Margins ◽  
Severe Dysplasia ◽  
Transoral Laser Microsurgery ◽  
Laser Microsurgery ◽  
Laryngeal Mucosa

Introduction: Treatment planning in T2, T3 laryngeal carcinoma is based on clinical assessment and radiological imaging. However, to delineate precise mucosal margins for transoral laser microsurgery (TLM), a high class, sophisticated endoscopy is indispensable. Narrowband imaging (NBI) which is an optical filter technology, seems to be a useful adjunctive tool in marking superficial margins. Materials and Methods: A total of 98 patients diagnosed with HNSCC underwent cordectomies and were enrolled in the evaluation. T2 and T3 stage cancer were diagnosed in 90 and 8 patients, respectively. Intraoperatively, prior to the first laser shot, all anatomical sites were endoscopically evaluated by WL and NBI. Results: In 10/98 patients (10.2%), 10 samples were taken based only on NBI findings to guarantee better delineation of superficial margins. The result of histology revealed moderate dysplasia in 4 cases (40%), severe dysplasia in 2 (20%), carcinoma in situ in 3 (30%) and hyperkeratosis in 1 (10%). Based on presented results, combined NBI/WL endoscopy reached the sensitivity of 100%, specificity 98.88%, positive predictive value 90%, negative predictive value 100% and accuracy 98.98%. All patients had clear margins according to definitive histology results. Discussion: In this paper, we aimed to assess the usefulness of NBI in intraoperative imaging of laryngeal mucosa and delineation of superficial margins in patients with selected T2 and T3 laryngeal cancer treated with TLM. We proved in our study that with the support of NBI endoscopy, it is possible to increase the accuracy of superficial resection margins in patients with moderately advanced laryngeal cancer (T2, T3).

scholarly journals Cyclist Effort Features: A Novel Technique for Image Texture Characterization Applied to Larynx Cancer Classification in Contact Endoscopy—Narrow Band Imaging

Diagnostics ◽  
2021 ◽  
Vol 11 (3) ◽  
pp. 432
Author(s):  
Nazila Esmaeili ◽  
Axel Boese ◽  
Nikolaos Davaris ◽  
Christoph Arens ◽  
Nassir Navab ◽  
...  
Keyword(s):  
Feature Extraction ◽  
Image Classification ◽  
Laryngeal Cancer ◽  
Classification Accuracy ◽  
Narrow Band ◽  
Narrow Band Imaging ◽  
Image Texture ◽  
Finish Line ◽  
Cad System ◽  
Textural Characterization

Background: Feature extraction is an essential part of a Computer-Aided Diagnosis (CAD) system. It is usually preceded by a pre-processing step and followed by image classification. Usually, a large number of features is needed to end up with the desired classification results. In this work, we propose a novel approach for texture feature extraction. This method was tested on larynx Contact Endoscopy (CE)—Narrow Band Imaging (NBI) image classification to provide more objective information for otolaryngologists regarding the stage of the laryngeal cancer. Methods: The main idea of the proposed methods is to represent an image as a hilly surface, where different paths can be identified between a starting and an ending point. Each of these paths can be thought of as a Tour de France stage profile where a cyclist needs to perform a specific effort to arrive at the finish line. Several paths can be generated in an image where different cyclists produce an average cyclist effort representing important textural characteristics of the image. Energy and power as two Cyclist Effort Features (CyEfF) were extracted using this concept. The performance of the proposed features was evaluated for the classification of 2701 CE-NBI images into benign and malignant lesions using four supervised classifiers and subsequently compared with the performance of 24 Geometrical Features (GF) and 13 Entropy Features (EF). Results: The CyEfF features showed maximum classification accuracy of 0.882 and improved the GF classification accuracy by 3 to 12 percent. Moreover, CyEfF features were ranked as the top 10 features along with some features from GF set in two feature ranking methods. Conclusion: The results prove that CyEfF with only two features can describe the textural characterization of CE-NBI images and can be part of the CAD system in combination with GF for laryngeal cancer diagnosis.

The value of narrow band imaging for early detection of laryngeal cancer

2008 ◽  
Vol 266 (7) ◽  
pp. 1017-1023 ◽  
Author(s):  
Akihito Watanabe ◽  
Masanobu Taniguchi ◽  
Hitoshi Tsujie ◽  
Masao Hosokawa ◽  
Masahiro Fujita ◽  
...  
Keyword(s):  
Early Detection ◽  
Laryngeal Cancer ◽  
Narrow Band ◽  
Narrow Band Imaging
Sign in / Sign up

Export Citation Format

Share Document