New Model and Public Online Prediction Platform for Risk Stratification of Vocal Cord Leukoplakia.

OBJECTIVE: To extract texture features from vocal cord leukoplakia (VCL) images and establish a VCL risk stratification prediction model using machine learning (ML) techniques. METHODS: A total of 462 patients with pathologically confirmed VCL were retrospectively collected and divided into low-risk and high-risk groups. We use a 5-fold cross validation method to ensure the generalization ability of the model built using the included dataset and avoid overfitting. Totally 504 texture features were extracted from each laryngoscope image. After feature selection, 10 ML classifiers were utilized to construct the model. The SHapley Additive exPlanations (SHAP) was employed for feature analysis. To evaluate the model, accuracy, sensitivity, specificity, and the area under the receiver operating characteristic (ROC) curve (AUC) were utilized. In addition, the model was transformed into an online application for public use and further tested in an independent dataset with 52 cases of VCL. RESULTS: A total of 12 features were finally selected, random forest (RF) achieved the best model performance, the mean accuracy, sensitivity, specificity, and AUC of the 5-fold cross validation were 92.2 ± 4.1%, 95.6 ± 4.0%, 85.8 ± 5.8%, and 90.7 ± 4.9%, respectively. The result is much higher than the clinicians (AUC between 63.1% and 75.2%). The SHAP algorithm ranks the importance of 12 texture features to the model. The test results of the additional independent datasets were 92.3%, 95.7%, 90.0%, and 93.3%, respectively. CONCLUSION: The proposed VCL risk stratification prediction model, which has been developed into a public online prediction platform, may be applied in practical clinical work. LEVEL OF EVIDENCE: 3 Laryngoscope, 2024.

Definition and analysis of gray matter atrophy subtypes in mild cognitive impairment based on data-driven methods.

Introduction: Mild cognitive impairment (MCI) is an important stage in Alzheimer's disease (AD) research, focusing on early pathogenic factors and mechanisms. Examining MCI patient subtypes and identifying their cognitive and neuropathological patterns as the disease progresses can enhance our understanding of the heterogeneous disease progression in the early stages of AD. However, few studies have thoroughly analyzed the subtypes of MCI, such as the cortical atrophy, and disease development characteristics of each subtype. Methods: In this study, 396 individuals with MCI, 228 cognitive normal (CN) participants, and 192 AD patients were selected from ADNI database, and a semi-supervised mixture expert algorithm (MOE) with multiple classification boundaries was constructed to define AD subtypes. Moreover, the subtypes of MCI were obtained by using the multivariate linear boundary mapping of support vector machine (SVM). Then, the gray matter atrophy regions and severity of each MCI subtype were analyzed and the features of each subtype in demography, pathology, cognition, and disease progression were explored combining the longitudinal data collected for 2 years and analyzed important factors that cause conversion of MCI were analyzed. Results: Three MCI subtypes were defined by MOE algorithm, and the three subtypes exhibited their own features in cortical atrophy. Nearly one-third of patients diagnosed with MCI have almost no significant difference in cerebral cortex from the normal aging population, and their conversion rate to AD are the lowest. The subtype characterized by severe atrophy in temporal lobe and frontal lobe have a faster decline rate in many cognitive manifestations than the subtype featured with diffuse atrophy in the whole cortex. APOE ε4 is an important factor that cause the conversion of MCI to AD. Conclusion: It was proved through the data-driven method that MCI collected by ADNI baseline presented different subtype features. The characteristics and disease development trajectories among subtypes can help to improve the prediction of clinical progress in the future and also provide necessary clues to solve the classification accuracy of MCI.

Single-cell RNA sequencing reveals the epithelial cell, fibroblast, and key gene alterations in chronic rhinosinusitis with nasal polyps.

Chronic rhinosinusitis with nasal polyps (CRSwNP) is a chronic inflammatory disease of the nasal mucosa, and epithelial-mesenchymal transition (EMT) is thought to be an essential process in the pathogenesis of CRSwNP. However, the mechanisms of epithelial and fibroblastic changes at the single-cell level are unclear. In this study, we investigated the epithelial cell, fibroblast, and key gene alterations in the development of CRSwNP. We revealed major cell types involved in CRSwNP and nasal mucosal inflammation formation, then mapped epithelial and fibroblast subpopulations. We showed that the apical and glandular epithelial cells and the ADGRB3+ and POSTN+ fibroblasts were the key cell subtypes in the progression of CRSwNP. Pseudotime and cell cycle analysis identified dynamic changes between epithelial cells and fibroblasts during its development. WFDC2 and CCL26 were identified as the key marker genes involved in the development of CRSwNP and were validated by IHC staining, which may provide a potential novel target for future CRSwNP therapy. ScRNA-seq data provided insights into the cellular landscape and the relationship between epithelial cells and fibroblasts in the progression of CRSwNP. WFDC2 and CCL26 were identified as the key genes involved in the development of CRSwNP and may be the potential markers for gene therapy.

Automatic prediction of obstructive sleep apnea event using deep learning algorithm based on ECG and thoracic movement signals.

BACKGROUND: Obstructive sleep apnea (OSA) is a sleeping disorder that can cause multiple complications. AIMS/OBJECTIVE: Our aim is to build an automatic deep learning model for OSA event detection using combined signals from the electrocardiogram (ECG) and thoracic movement signals. MATERIALS AND METHODS: We retrospectively obtained 420 cases of PSG data and extracted the signals of ECG, as well as the thoracic movement signal. A deep learning algorithm named ResNeSt34 was used to construct the model using ECG with or without thoracic movement signal. The model performance was assessed by parameters such as accuracy, precision, recall, F1-score, receiver operating characteristic (ROC), and area under the ROC curve (AUC). RESULTS: The model using combined signals of ECG and thoracic movement signal performed much better than the model using ECG alone. The former had accuracy, precision, recall, F1-score, and AUC values of 89.0%, 88.8%, 89.0%, 88.2%, and 92.9%, respectively, while the latter had values of 84.1%, 83.1%, 84.1%, 83.3%, and 82.8%, respectively. CONCLUSIONS AND SIGNIFICANCE: The automatic OSA event detection model using combined signals of ECG and thoracic movement signal with the ResNeSt34 algorithm is reliable and can be used for OSA screening.

A practical online prediction platform to predict the survival status of laryngeal squamous cell carcinoma after 5 years.

OBJECTIVE: Currently, there are few practical tools for predicting the prognosis of laryngeal squamous cell carcinoma (LSCC). This study aims to establish a model and a convenient online prediction platform to predict whether LSCC patients will survive 5 years after diagnosis, providing a reference for further evaluation of patient prognosis. METHODS: This is a retrospective study based on data collected from two centers. Center 1 included 117 LSCC patients with survival prognosis data, and center 2 included 33 patients, totaling 150 patients. All data were divided into independent training sets (60 %) and testing sets (40 %). Eight machine learning (ML) algorithms were used to establish models with 11 clinical parameters as input features. The accuracy, sensitivity, specificity, and receiver operating characteristic curve (ROC) of the testing set were used to evaluate the models, and the best model was selected. The model was then developed into a website-based 5-year survival status prediction platform for LSCC. In addition, we also used the SHapley Additive exPlanations (SHAP) tool to conduct interpretability analysis on the parameters of the model. RESULTS: The LSCC 5-year survival status prediction model using the support vector machine (SVM) algorithm achieved the best results, with accuracy, sensitivity, specificity, and area under the ROC curve (AUC) of 85.0 %, 87.5 %, 75.0 %, and 81.2 % respectively. The online platform for predicting the 5-year survival status of LSCC based on this model was successfully established. The SHAP analysis shows that the clinical stage is the most important feature of the model. CONCLUSION: This study successfully established a ML model and a practical online prediction platform to predict the survival status of laryngeal cancer patients after 5 years, which may help clinicians to better evaluate the prognosis of LSCC.

Deep learning technique to detect craniofacial anatomical abnormalities concentrated on middle and anterior of face in patients with sleep apnea.

OBJECTIVES: The aim of this study is to propose a deep learning-based model using craniofacial photographs for automatic obstructive sleep apnea (OSA) detection and to perform design explainability tests to investigate important craniofacial regions as well as the reliability of the method. METHODS: Five hundred and thirty participants with suspected OSA are subjected to polysomnography. Front and profile craniofacial photographs are captured and randomly segregated into training, validation, and test sets for model development and evaluation. Photographic occlusion tests and visual observations are performed to determine regions at risk of OSA. The number of positive regions in each participant is identified and their associations with OSA is assessed. RESULTS: The model using craniofacial photographs alone yields an accuracy of 0.884 and an area under the receiver operating characteristic curve of 0.881 (95% confidence interval, 0.839-0.922). Using the cutoff point with the maximum sum of sensitivity and specificity, the model exhibits a sensitivity of 0.905 and a specificity of 0.941. The bilateral eyes, nose, mouth and chin, pre-auricular area, and ears contribute the most to disease detection. When photographs that increase the weights of these regions are used, the performance of the model improved. Additionally, different severities of OSA become more prevalent as the number of positive craniofacial regions increases. CONCLUSIONS: The results suggest that the deep learning-based model can extract meaningful features that are primarily concentrated in the middle and anterior regions of the face.

New parameters measured via preoperative tonsil photos to evaluate the post-tonsillectomy pain: an analysis assisted by machine learning.

Background: Postoperative pain is the most common complication after tonsillectomy. We aimed to explore new parameters related to post-tonsillectomy pain, as well as to construct and validate a model for the preoperative evaluation of patients' risk for postoperative pain. Methods: Data collected from patients who underwent tonsillectomy by the same surgeon at Beijing Chaoyang Hospital from January 2019 to May 2022 were analyzed. Preoperative tonsil images from all patients were taken, and the ratios of the distance between the upper pole of the tonsil and the base of the uvula (L1 for the left side and R1 for the right side) to the width of the uvula (U1) or the length of the uvula (U2) were measured. The following six ratios were calculated: L1/U1, R1/U1, LR1/U1 (the add of L1 and R1, and then divide U1), L1/U2, R1/U2, LR1/U2 (the add of L1 and R1, and then divide U2). The post-tonsillectomy pain was recorded. In addition, machine learning (ML) algorithm and feature importance analysis were used to evaluate the value of the parameters. Results: A total of 100 patients were involved and divided into the training set (60%) and the validation set (40%). All six parameters are negatively correlated with post-tonsillectomy pain. The accuracy, sensitivity, and specificity of the model were 75.0%, 72.7%, and 77.8%, respectively. LR1/U1 and LR1/U2 are the most valuable parameters to evaluate post-tonsillectomy pain. Conclusions: We have discovered new parameters that can be measured using preoperative tonsil images to evaluate post-tonsillectomy pain. ML models based on these parameters could predict whether these patients will have intolerable pain after tonsillectomy and manage it promptly.

The Necessity of Subsequent Dental Treatment for Odontogenic Sinusitis After Endoscopic Sinus Surgery.

OBJECTIVE: To investigate the necessity of subsequent dental treatment for patients with odontogenic sinusitis (OS) after endoscopic sinus surgery (ESS). METHODS: OS patients who received ESS from January 2018 to May 2022 were eligible. Demographic data, medical history, Lund-Mackay (LM) computed tomography (CT) score, Lund-Kennedy (LK) endoscopic score, and OS-specific CT score were recorded and analyzed. All patients were followed up for 6 months after surgery. RESULTS: A total of 48 patients (30 males and 18 females) were enrolled. A total of 35 cases (72.9%) were caused by dental diseases (periapical periodontitis), and 13 cases (27.1%) were caused by iatrogenic factors (including tooth extraction or implantation). The overall treatment success rate was 97.9%, and the success rates in the dental disease OS group (97.1%) and the iatrogenic OS group (100%) were not significantly different (X2 = 0.329, P = .729). None of the patients in the dental disease group received treatment of the causative teeth within 6 months after the operation, except for one patient who received tooth extraction treatment 6 months after the operation. CONCLUSION: Without considering the impact of dental disease on patients' oral health, OS can be completely alleviated by ESS alone, even if no dental treatment is carried out after surgery.

Effect of Cold-Water Irrigation on the Coblation Site for Post-Tonsillectomy Pain: A Prospective Randomized Clinical Study.

OBJECTIVE: To explore the effect of cold-water irrigation on post-tonsillectomy pain after coblation. METHODS: Data from 61 adult patients who underwent coblation tonsillectomy in our hospital from January 2019 to December 2020 were collected, and the patients were randomly divided into the cold-water irrigation group (Group 1) and the room-temperature irrigation group (Group 2). Group 1 was irrigated with ice water mixed saline with the help of a pressure band, and Group 2 was irrigated with room-temperature saline. During the operation, we monitored the temperature of the operating cavity in real time. We recorded the postoperative pain for 11 consecutive days from the day and the 10th day after the operation. RESULTS: The postoperative pain score was significantly lower than that in Group 2, except on the 2nd, 3rd and 7th and 8th days after the operation. CONCLUSION: The perfusion of cold water during coblation tonsillectomy is helpful to reduce postoperative pain.

Developing and validating a multivariable machine learning model for the preoperative prediction of lateral lymph node metastasis of papillary thyroid cancer.

Background: At present, preoperative diagnosis of lateral cervical lymph node metastasis (LLNM) in patients with papillary thyroid carcinoma (PTC) mostly depends on the training and expertise of ultrasound doctors. A machine-learning model for predicting LLNM accurately before PTC surgery may help to determine the scope of surgery and reduce unnecessary surgical trauma. Methods: The data of patients with primary PTC who underwent thyroidectomy with lateral cervical lymph node surgery at Beijing Tongren Hospital between July 2009 and June 2021 were retrospectively analyzed. All patients had complete ultrasonic examination, clinical data, and definite pathology diagnosis of lymph nodes. LLNM was confirmed by postoperative pathology. The patients were randomly divided into a training set (155 cases) and a test set (98 cases) at a ratio of 6:4. Eleven parameters, including patient demographics, ultrasound results, and tumor-related conditions, were collected, and a prediction model was established using the support vector machine (SVM) algorithm. Several other machine-learning algorithms were also used to establish models for comparison. The accuracy, precision, recall, F1-score, sensitivity, specificity, Cohen's kappa value, and area under the receiver operating characteristic curve (AUC) were used to evaluate model performance. Results: A total of 87 males and 156 females were included in the study, aged 14-80 years. One hundred and four patients of them had LLNM and 139 did not have LLNM. The pandas Python library was used for the statistical analysis, and the Spearman coefficient was used to analyze the correlation between each parameter and the prediction index. The SVM model performed the best among all the models. Its accuracy, precision, recall, F1-score, sensitivity, specificity, Cohen's kappa value, and AUC were 90.8%, 91.0%, 90.8%, 90.8%, 87.5%, 94.0%, 81.6%, and 91.0%, respectively. Conclusions: This model can enable surgeons to improve the accuracy of ultrasonography in predicting LLNM without additional examination, thus avoiding missing positive lateral cervical lymph nodes and reducing the sequelae caused by unnecessary lateral neck dissection.

Surgical Intervention to Treat Pharyngolaryngeal Stenosis Caused by Behcet's Disease.

We aimed to summarize the surgical treatment for pharyngolaryngeal stenosis and discuss prognosis in patients with Behcet's disease. Six cases of pharyngolaryngeal stenosis caused by Behcet's disease were analyzed retrospectively. All underwent surgical treatment for pharyngolaryngeal stenosis after systematic medical treatment. The follow-up time for the 6 patients was between 1 and 12 years. Four of the 6 patients underwent adhesiolysis as their first procedure. Two of these 4 experienced recurrence of stenosis within 6 months and underwent flap repair as their second procedure. The remaining two patients underwent flap reconstruction as their first procedure and maintained good swallowing function. Three of the 6 patients underwent preoperative tracheotomy because of dyspnea. Tracheal decannulation was successful in all patients. None of the patients experienced recurrence after their final surgical procedure and all recovered to a near-normal condition. Pharyngolaryngeal stenosis caused by Behcet's disease is a rare but severe complication; surgical intervention should be considered in patients with dysphagia after systematic medical treatment.

A model for obstructive sleep apnea detection using a multi-layer feed-forward neural network based on electrocardiogram, pulse oxygen saturation, and body mass index.

PURPOSE: To develop and evaluate a model for obstructive sleep apnea (OSA) detection using an artificial neural network (ANN) based on the combined features of body mass index (BMI), electrocardiogram (ECG), and pulse oxygen saturation (SpO2). METHODS: Polysomnography (PSG) data for 148 patients with OSA and 33 unaffected individuals were included. A multi-layer feed-forward neural network (FNN) was used based on the features obtained from ECG, SpO2, and BMI. The receiver operating characteristic (ROC) curve and the metrics of accuracy, sensitivity, and specificity were used to evaluate the performance of the overall classification. Some other machine learning methods including linear discriminant, linear Support Vector Machine (SVM), Complex Tree, RUSBoosted Trees, and Logistic Regression were also used to compare their performance with the FNN. RESULTS: The accuracy, sensitivity, and specificity of the proposed multi-layer FNN were 97.8%, 98.6%, and 93.9%, respectively, and the area under the ROC curve was 97.0%. Compared with the other machine learning methods mentioned above, the FNN achieved the highest performance. CONCLUSIONS: The satisfactory performance of the proposed FNN model for OSA detection indicated that it is reliable to screen potential patients with OSA using the combined channels of ECG and SpO2 and also taking into account BMI. This strategy might be a viable alternative method for OSA diagnosis.

Upregulation of BTF3 affects the proliferation, apoptosis, and cell cycle regulation in hypopharyngeal squamous cell carcinoma.

PURPOSE: The BTF3 is involved in oncogenesis, while the biological roles in HSCC remain unclear. The aim of this study was to explore the impact of BTF3 knockdown on biological phenotypes of human HSCC in vivo and in vitro. METHODS: The expression of BTF3 was assessed in HSCC and normal tissues. In vitro experiments were performed to explore impact of BTF3 knockdown on biological phenotypes of human HSCC cell line, including proliferation, cell cycle, and apoptosis. Moreover, nude mice were used to evaluate growth of xenograft tumors. Finally, gene chip was used to explore the potential signaling pathways of BTF3, with confirmation of potential BTF3-related genes. RESULTS: Our results showed elevated expression of BTF3 was observed in HSCC tumors compared to paired adjacent normal tissues in 68 patients, and positively associated with lymph node metastasis and survival of this HSCC patient cohort. In addition, in vitro experiments showed that BTF3 knockdown significantly impaired regulation of proliferation, cell cycle, and apoptosis, potentially via ATM signaling pathway. Finally, in vivo experiments demonstrated that BTF3 functioned as an oncogene by promoting the development and progression of HSCC tumors, indicating its oncogenic role in HSCC. CONCLUSIONS: This study for the first time demonstrated that expression of BTF3 is upregulated in HSCC tumors and this upregulation is positively correlated with lymph node metastasis of this malignancy. The oncogenic role of BTF3 is further validated for tumor promotion and progression of HSCC in vivo, indicating that BTF3 is a potential therapeutic target and prognostic biomarker for HSCC.

Expression and clinical significance of basic transcription factor 3 in nasopharyngeal carcinoma.

Basic transcription factor 3 (BTF3), a transcription factor and modulator of apoptosis, is differentially expressed in carcinoma. To acquire further understanding of the involvement of BTF3 in carcinoma, the present study analyzed the expression of BTF3, as well as its role in cell function in nasopharyngeal carcinoma (NPC). BTF3 transcription rates in human NPC samples (n=46) and adjacent normal tissue samples (n=46) were analyzed using reverse transcription-quantitative polymerase chain reaction and immunohistochemistry. BTF3-silencing in NPC cells was performed via specific small interfering RNA molecules. The function of BTF3 was analyzed by proliferation assays and colony forming assays using a Cellomic assay system. The positive expression rates of BTF3 were significantly increased in cancerous tissues compared with those in adjacent tissues (P<0.05). In addition, BTF3-silencing decreased cell proliferation and colony formation (P<0.01) in TCA-8113 and 5-8F cells. BTF3 is overexpressed in NPC, and its silencing is associated with decreased cell proliferation and colony formation, enhanced apoptosis and cell cycle regulation of TCA-8113 and 5-8F cells.

Significance of MMP11 and P14(ARF) expressions in clinical outcomes of patients with laryngeal cancer.

BACKGROUND: To evaluate the association of MMP11 and P14(ARF) expression in laryngeal squamous cell carcinoma (LSCC) with clinical pathological characteristics and survival. METHODS: The mRNA and protein levels for both genes were determined in 65 LSCC patients. A log-rank test and Cox models were used to compare survival among different groups. RESULTS: The mRNA expressions of MMP11 and P14(ARF) were significantly different between LSCC and their corresponding adjacent tissues (All P < 0.001). The expressions of MMP11 and P14(ARF) were correlated with several clinical characteristics (All P < 0.05). Patients with low MMP11 and high P14(ARF) expression had significantly better survival compared with those with high MMP11 and low P14(ARF) expression, respectively (All P < 0.05). The patients with surgery only had significantly better survival than those with chemoradiotherapy (log rank: P = 0.016), particularly in patients with low MMP11 and high P14(ARF) expression (log rank: P = 0.006). Furthermore, multivariable analysis showed that patients with low MMP11 and high P14(ARF) expression alone had a significantly reduced risk of death compared with those with high MMP11 and low P14(ARF) expression. The reduced risk for overall death was pronounced for patients with low and high expression of both genes (HR, 0.2; 95% CI, 0.1-0.5) compared with any other co-expression status of both genes, particularly for patients with surgery only (HR, 0.1; 95% CI, 0.0-0.9). CONCLUSION: These results suggest that altered expression of MMP11 and P14(ARF) in tumors may individually, or in combination, predict poor prognosis of LSCC, particularly for patients with surgery only.

Polymorphisms of the XPC gene may contribute to the risk of head and neck cancer: a meta-analysis.

Polymorphisms of the XPC gene have been reported to be associated with an increased risk of head and neck cancer (HNC), though the exact biological effect is still unclear. Genetic association studies (GAS) investigating the associations between three common polymorphisms (PAT, Lys939Gln, and Ala499Val) of the XPC gene and HNC risk have produced contradictory and inconclusive results. The aim of this meta-analysis is to evaluate the contributions of these polymorphisms to the risk of HNC. A literature search was conducted in the PubMed, Embase, Web of Science, Cochrane Library, and China National Knowledge Infrastructure databases to indentify eligible studies. Pooled odds ratios (ORs) and 95% confidence intervals (95% CIs) were used to evaluate the strength of the associations under a fixed- or random-effect model according to heterogeneity test. Twelve case-control studies were included in this meta-analysis with a total of 3,078 HNC patients and 4,311 healthy controls. For XPC PAT, a significant overall association was found under all major genetic models. Stratified analyses further indicated significant associations in the Caucasian, population-based, non-PCR-RFLP, esophageal cancer and oral cancer subgroups. For XPC Lys939Gln, few significant results were found in either the overall analysis or stratified analyses. For XPC Ala499Val, the combined results revealed a significantly increased risk of HNC for carriers of the 499Val allele. This meta-analysis shows that the XPC PAT and Ala499Val polymorphisms may be associated with an increased risk of HNC, while XPC Lys939Gln may not be associated with HNC risk. Despite some limitations, this meta-analysis establishes solid statistical evidence for an association between XPC genetic polymorphisms and HNC risk that warrants further validation.

Association of TBX2 and P21 expression with clinicopathological features and survival of laryngeal squamous cell carcinoma.

BACKGROUND: The expression of P21 and TBX2 in laryngeal squamous cell carcinoma (LSCC) and their corresponding adjacent normal laryngeal tissues, as well as their association with clinical pathological features and survival remain unclear. METHOD: we used the RT-PCR and immunohistochemistry to detect their mRNA and protein levels in 75 LSCC patients. We also use log-rank test and Cox models to compare survival among different groups. RESULTS: The mRNA expression level of TBX2 was up-regulated, while P21 was down-regulated in LSCC compared with their matched adjacent laryngeal tissues (All P < 0.001). The expression of P21 was correlated with tumor stage, lymph node metastasis and smoking; and TBX2 expression was associated with lymph node metastasis, differentiation degree and smoking (All P < 0.05). Patients with high TBX2 and low P21 expression had significantly worse survival than those with low TBX2 and high P21 expression, respectively (All P < 0.05). A significant correlation between expression of TBX2 and P21 (Pearson, P < 0.05) was observed. Furthermore, multivariable analysis showed that patients with low TBX2 and high P21 expression alone had a significantly reduced risk for overall death compared with those with low TBX2 and high P21 expression. The risk for overall death was even lower for patients with both low and high expression of both genes than any other co-expression status of both genes (HR, 0.1; 95% CI, 0.0-0.9). CONCLUSION: These results suggest that abnormal expression of P21 and TBX2 in tumors may jointly, or individually, predict poor prognosis of LSCC.