Objective:To investigate the clinical value of electronic laryngoscope combined with narrow band imaging（NBI） endoscope and its classification in the diagnosis of vocal cord leukoplakia. Methods:A retrospective analysis was performed on 115 cases of patients treated in the Department of Otolaryngology, Head and Neck Surgery, Nanjing Drum Tower Hospital from September 2020 to November 2022. All 115 cases were diagnosed with vocal cord leukogramma using the electronic laryngoscopy and narrow band imaging endoscopy, followed by pathological examination in the outpatient tissue biopsy. The morphological characteristics of vocal cord leukoplakia and the correlation between narrow band imaging classification and pathological results were investigated. Results:Among 115 cases of vocal cord leukoplakia, 46 cases（40.00%） were diagnosed as benign lesions. Low grade intraepithelial neoplasia occurred in 29 cases（25.22%）. High-grade intraepithelial neoplasia（including carcinoma in situ） occurred in 22 cases（19.13%）. Invasive carcinoma（including suspected invasive carcinoma） was found in 18 cases（15.65%）. There were no statistical differences in the unilateral and bilateral distribution of vocal cord leukoplakia and pathological results（P>0.05）, but there were statistical differences in the size, thickness, lesion uniformity, clear boundary, pre-invasion commissure, symmetry,age over 55 years old, morphological classification, NBI classification and pathological results distribution（P<0.05）. The two-by-two comparison among the three groups of morphological classification（flat type, raised type, rough type） showed that P<0.017 was only compared between flat type and rough type, and P>0.017 was compared between the other two groups. The pairwise comparison among the three groups of NBI classification（Ⅲ, Ⅳ, Ⅴ） was statistically significant（P<0.017）. There was a high correlation between NBI classification and pathological diagnosis, and the correlation coefficient was 0.705（P<0.05）. The risk of high intraepithelial neoplasia and cancerization in type Ⅳ was 9.125 times higher than that in type Ⅲ, and the risk of high intraepithelial neoplasia and cancerization in type Ⅴ was 271.078 times higher than that in type Ⅲ. The area under the curve of morphological classification and NBI classification were 0.672 and 0.896, respectively. Conclusion:There is a high match and correlation between NBI classification and pathological diagnosis. Electronic laryngoscope combined with narrow band imaging endoscope has a high diagnostic value for vocal cord leukoplakia, and a strong predictive ability for malignant leukoplakia.
Humans ; Vocal Cords/pathology* ; Male ; Female ; Leukoplakia/diagnosis* ; Narrow Band Imaging/methods* ; Retrospective Studies ; Laryngoscopy/methods* ; Middle Aged ; Laryngeal Neoplasms/diagnosis* ; Aged ; Laryngoscopes ; Adult

Objective:To evaluate the diagnostic accuracy of the convolutional neural network（CNN） in diagnosing nasopharyngeal carcinoma using endoscopic narrowband imaging. Methods:A total of 834 cases with nasopharyngeal lesions were collected from the People's Hospital of Guangxi Zhuang Autonomous Region between 2014 and 2016. We trained the DenseNet201 model to classify the endoscopic images, evaluated its performance using the test dataset, and compared the results with those of two independent endoscopic experts. Results:The area under the ROC curve of the CNN in diagnosing nasopharyngeal carcinoma was 0.98. The sensitivity and specificity of the CNN were 91.90% and 94.69%, respectively. The sensitivity of the two expert-based assessment was 92.08% and 91.06%, respectively, and the specificity was 95.58% and 92.79%, respectively. There was no significant difference between the diagnostic accuracy of CNN and the expert-based assessment （P=0.282, P=0.085）. Moreover, there was no significant difference in the accuracy in discriminating early-stage and late-stage nasopharyngeal carcinoma（P=0.382）. The CNN model could rapidly distinguish nasopharyngeal carcinoma from benign lesions, with an image recognition time of 0.1 s/piece. Conclusion:The CNN model can quickly distinguish nasopharyngeal carcinoma from benign nasopharyngeal lesions, which can aid endoscopists in diagnosing nasopharyngeal lesions and reduce the rate of nasopharyngeal biopsy.
Humans ; Nasopharyngeal Carcinoma ; Narrow Band Imaging ; China ; Neural Networks, Computer ; Nasopharyngeal Neoplasms/diagnostic imaging*

Objective:To study the application value of narrow-band imaging in the diagnosis of laryngopharyngeal reflux. Methods:A total of 275 patients admitted to the inpatient department or laryngoscopy room of the Otolaryngology Head and Neck Surgery Department of the First Affiliated Hospital of Harbin Medical University from September 2022 to April 2023 due to throat discomfort were selected as the research subjects. All of them completed RSI, RFS scoring scales and electronic laryngoscopy（including ordinary white light and NBI）. According to the expert consensus of LPRD in 2022, RSI and RFS scoring scale were used as diagnostic criteria to divide them into LPR group and non-LPR group. Chi-square test was used to analyze the differences of positive rates of characteristic manifestations under NBI among different groups. The consistency of NBI and scale diagnostic methods was analyzed by Kappa, and RSI and RFS scoring were used as diagnostic criteria, The diagnostic efficiency of NBI method was analyzed. Results:There were 190 people in the LPR group, 157 of whom showed characteristic performance under the NBI mode, with a positive rate of 82.6%（157/190）; there were 85 people in the non-LPR group, with a positive rate of 18.8%（16/85）. There was a statistically significant difference in the positive rate between the two groups（χ²=102.47, P<0.05）. The consistency rate between RSI, RFS and NBI was 82.2%（226/275）. Kappa consistency analysis was used, and Kappa=0.605（P<0.05）, indicating good consistency between the two diagnostic methods. Using RSI and RFS as diagnostic criteria for LPR, the sensitivity of NBI diagnostic method was 82.6%（157/190）, specificity 81.2%（69/85）, positive predictive value 90.8%（157/173） and negative predictive value 67.6%（69/102）. Conclusion:Narrow-band imaging, as a new endoscopic imaging technique, can show small changes in mucosal surface micro vessels and play an important role in the diagnosis of laryngopharyngeal reflux.
Humans ; Laryngopharyngeal Reflux/diagnostic imaging* ; Narrow Band Imaging ; Laryngoscopy/methods* ; Pharynx ; Predictive Value of Tests

Objective: To construct the diagnostic model of superficial esophageal squamous cell carcinoma (ESCC) and precancerous lesions in endoscopic images based on the YOLOv5l model by using deep learning method of artificial intelligence to improve the diagnosis of early ESCC and precancerous lesions under endoscopy. Methods: 13, 009 endoscopic esophageal images of white light imaging (WLI), narrow band imaging (NBI) and lugol chromoendoscopy (LCE) were collected from June 2019 to July 2021 from 1, 126 patients at the Cancer Hospital, Chinese Academy of Medical Sciences, including low-grade intraepithelial neoplasia, high-grade intraepithelial neoplasia, ESCC limited to the mucosal layer, benign esophageal lesions and normal esophagus. By computerized random function method, the images were divided into a training set (11, 547 images from 1, 025 patients) and a validation set (1, 462 images from 101 patients). The YOLOv5l model was trained and constructed with the training set, and the model was validated with the validation set, while the validation set was diagnosed by two senior and two junior endoscopists, respectively, to compare the diagnostic results of YOLOv5l model and those of the endoscopists. Results: In the validation set, the accuracy, sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of the YOLOv5l model in diagnosing early ESCC and precancerous lesions in the WLI, NBI and LCE modes were 96.9%, 87.9%, 98.3%, 88.8%, 98.1%, and 98.6%, 89.3%, 99.5%, 94.4%, 98.2%, and 93.0%, 77.5%, 98.0%, 92.6%, 93.1%, respectively. The accuracy in the NBI model was higher than that in the WLI model (P<0.05) and lower than that in the LCE model (P<0.05). The diagnostic accuracies of YOLOv5l model in the WLI, NBI and LCE modes for the early ESCC and precancerous lesions were similar to those of the 2 senior endoscopists (96.9%, 98.8%, 94.3%, and 97.5%, 99.6%, 91.9%, respectively; P>0.05), but significantly higher than those of the 2 junior endoscopists (84.7%, 92.9%, 81.6% and 88.3%, 91.9%, 81.2%, respectively; P<0.05). Conclusion: The constructed YOLOv5l model has high accuracy in diagnosing early ESCC and precancerous lesions in endoscopic WLI, NBI and LCE modes, which can assist junior endoscopists to improve diagnosis and reduce missed diagnoses.
Artificial Intelligence ; Endoscopy/methods* ; Esophageal Neoplasms/pathology* ; Esophageal Squamous Cell Carcinoma/diagnostic imaging* ; Humans ; Narrow Band Imaging ; Precancerous Conditions/diagnostic imaging* ; Sensitivity and Specificity

Humans ; Laryngeal Neoplasms/diagnostic imaging* ; Laryngoscopes ; Narrow Band Imaging ; Papilloma/diagnostic imaging*

Objective: To explore the possibility of using artificial intelligence (AI) technology based on convolutional neural network (CNN) to assist the clinical diagnosis of laryngeal squamous cell carcinoma (LSCC) through deep learning algorithm. Methods: A deep CNN was developed and applied in narrow band imaging (NBI) endoscopy of 4 799 patients with laryngeal lesions, including 3 168 males and 1 631 females, aged from 21 to 87 years, from 2015 to 2017 in Beijing Tongren Hospital, Capital Medical University. A simple randomization method was used to select the laryngeal NBI images of 2 427 patients (1 388 benign lesions and 1 039 LSCC lesions) for the training and correction the CNN model. The remaining laryngeal NBI images of 2 372 patients (including 1 276 benign lesions and 1 096 LSCC lesions) were used as validation data set to compare performance between CNN and otolaryngologists. SPSS 21.0 software was used for Chi-square test to calculate the accuracy, sensitivity and specificity of AI and otolaryngologists. The area under the curve (AUC) of receiver operating curve (ROC) was used to evaluate the diagnostic ability of the algorithm for NBI images. Results: The accuracy, sensitivity and specificity for NBI predictions were respectively 90.91% (AUC=0.96), 90.12% and 91.53%, which were equivalent to those for otolaryngologists' predictions (accuracy, sensitivity and specificity were (91.93±3.20)%, (91.33±3.25)% and (93.02±2.59)%, t values were 0.64, 0.75 and 1.17, and P values were 0.32, 0.28 and 0.21, respectively). The diagnostic efficiency of CNN was significantly higher than that of otolaryngologists (0.01 vs. 5.50, t =9.15, P<0.001). Conclusion: AI based on deep CNN is effective for using in the laryngeal NBI image diagnosis, showing a good application prospect in the diagnosis of LSCC.
Adult ; Aged ; Aged, 80 and over ; Artificial Intelligence ; Endoscopy ; Female ; Head and Neck Neoplasms ; Humans ; Male ; Middle Aged ; Narrow Band Imaging ; Neural Networks, Computer ; Sensitivity and Specificity ; Squamous Cell Carcinoma of Head and Neck ; Young Adult

Esophageal Neoplasms/diagnostic imaging* ; Humans ; Laryngeal Neoplasms/diagnostic imaging* ; Laryngoscopes ; Narrow Band Imaging ; Neoplasms, Multiple Primary

OBJECTIVE: To investigate the effect of NBI assisted white light transurethral resection of bladder tumor (TURBT) in the treatment of bladder urothelial carcinoma and to summarize the experience of narrow band imaging (NBI) operation. METHODS: Patients with bladder urothelial carcinoma were selected, and TURBT was performed after anesthesia. First of all, the bladder tumor was found and resected under white light. Then we replaced with NBI, looked for suspicious lesions and resected them, The specimens excised under white light and NBI were collected separately. The number, location and pathological results of the lesions under white light were recorded, and the residual lesions under NBI were also recorded. To evaluate the effect of NBI, the ratio of residual bladder tumor was calculated. The cases were divided into three groups according to the time sequence. The clinical data of each group were collected and the learning curve of TURBT under NBI assisted white light was observed. RESULTS: A prospective study of 45 patients with bladder tumor from April 2018 to January 2020, including 32 males and 13 females, aged from 23 to 89 years, with an average age of 65.2 years. All the operations were successfully completed, without obvious complications after operation. Nine cases were single and 36 cases were multiple. The maximum diameter of the tumors was 0.5 to 4.0 cm, with an average of 2.2 cm. The histopathology of the resected tissue under white light was urothelial carcinoma, and 19 cases (42.2%) were pathologically positive by NBI resection. The 45 cases were divided into three groups according to the time sequence, 15 cases in each group. The true positive rate of NBI was 33.3%, 46.7% and 46.7%, respectively, and the false positive rate was 60.0%, 46.7% and 26.7%, respectively in the three groups. CONCLUSION TURBT is an effective way to treat bladder urothelial cancer, NBI is an effective supplement of white light, which can increase the detection rate of bladder cancer and reduce post-operative recurrence. The NBI light source has a certain learning curve. With the increase of cases, the false-positive rate of NBI is gradually reduced. After the NBI operator has rich experience, the recognition degree of flat tumor is gradually improved under white light, and the residual rate of NBI is reduced after the removal under white light.
Adult ; Aged ; Aged, 80 and over ; Cystoscopy ; Female ; Humans ; Male ; Middle Aged ; Narrow Band Imaging ; Neoplasm Recurrence, Local ; Prospective Studies ; Self-Control ; Urinary Bladder Neoplasms/surgery* ; Young Adult

PURPOSE: Gastric adenocarcinoma of the fundic gland type (chief cell predominant type) (GA-FG-CCP) was first reported as a rare adenocarcinoma found in the normal fundic mucosa. Recent studies have proposed the possibility that GA-FG-CCPs were also generated in the atrophic mucosa after Helicobacter pylori (HP) eradication therapy. However, little is known on the endoscopic findings of GA-FG-CCP generated in the atrophic mucosa due to its extreme rarity. MATERIALS AND METHODS: A total of 8 patients who underwent endoscopic submucosal resection and were diagnosed with GA-FG-CCP generated in the HP-uninfected mucosa (4 cases, HP-uninfected group) or HP-eradicated atrophic mucosa (4 cases, HP-eradicated group) were retrospectively analyzed, and their endoscopic findings, including magnifying endoscopy with narrow band imaging (M-NBI), and pathological features were compared. RESULTS: While GA-FG-CCPs in the 2 groups displayed similar macroscopic appearance, M-NBI demonstrated that characteristic microvessels (tapered microvessels like withered branches) were specifically identified in the HP-eradicated group. Pathological investigation revealed that a decreasing number of fundic glands and thinned foveolar epithelium covering tumor ducts were thought to lower the thickness of the covering layer over tumor ducts in the HP-eradicated group. Moreover, dilation of vessels just under the surface of the lesions contributed to the visualization of microvessels by M-NBI. CONCLUSIONS: The change in background mucosa due to HP infection influenced the thickness of the covering layer over the tumor ducts and M-NBI finding of GA-FG-CCP.
Adenocarcinoma ; Endoscopy ; Epithelium ; Helicobacter pylori ; Helicobacter ; Humans ; Microvessels ; Mucous Membrane ; Narrow Band Imaging ; Retrospective Studies ; Stomach Neoplasms

In the last few years, interest in the optical diagnosis of colorectal polyps has increased among gastroenterologists. Several studies have shown that the optical diagnosis of small colorectal polyps is safe and feasible in routine clinical practice and is comparable to histopathology. The Narrow-band Imaging International Colorectal Endoscopic Classification provides a validated criterion for the classification of neoplastic and nonneoplastic polyps as well as polyps with deep submucosal invasion using narrow band imaging during real-time colonoscopy. The aim of the present review is to assess the current evidence for and limitations of optical diagnosis and to propose a systematic approach for transferring research findings to patient care.
Classification ; Colonic Polyps ; Colonoscopy ; Diagnosis* ; Narrow Band Imaging ; Optical Imaging ; Patient Care ; Polyps*