Head and neck squamous cell carcinoma (HNSC) exhibits genetic heterogeneity in etiologies, tumor sites, and biological processes, which significantly impact therapeutic strategies and prognosis. While the influence of human papillomavirus on clinical outcomes is established, the molecular subtypes determining additional treatment options for HNSC remain unclear and inconsistent. This study aims to identify distinct HNSC molecular subtypes to enhance diagnosis and prognosis accuracy. In this study, we collected three HNSC microarrays (n = 306) from the Gene Expression Omnibus (GEO), and HNSC RNA-Seq data (n = 566) from The Cancer Genome Atlas (TCGA) to identify differentially expressed genes (DEGs) and validate our results. Two scoring methods, representative score (RS) and perturbative score (PS), were developed for DEGs to summarize their possible activation functions and influence in tumorigenesis. Based on the RS and PS scoring, we selected candidate genes to cluster TCGA samples for the identification of molecular subtypes in HNSC. We have identified 289 up-regulated DEGs and selected 88 genes (called HNSC88) using the RS and PS scoring methods. Based on HNSC88 and TCGA samples, we determined three HNSC subtypes, including one HPV-associated subtype, and two HPV-negative subtypes. One of the HPV-negative subtypes showed a relationship to smoking behavior, while the other exhibited high expression in tumor immune response. The Kaplan–Meier method was used to compare overall survival among the three subtypes. The HPV-associated subtype showed a better prognosis compared to the other two HPV-negative subtypes (log rank, p = 0.0092 and 0.0001; hazard ratio, 1.36 and 1.39). Additionally, within the HPV-negative group, the smoking-related subgroup exhibited worse prognosis compared to the subgroup with high expression in immune response (log rank, p = 0.039; hazard ratio, 1.53). The HNSC88 not only enables the identification of HPV-associated subtypes, but also proposes two potential HPV-negative subtypes with distinct prognoses and molecular signatures. This study provides valuable strategies for summarizing the roles and influences of genes in tumorigenesis for identifying molecular signatures and subtypes of HNSC.

Objective: Obstructive sleep apnea (OSA) is a common sleep disorder characterized by upper airway obstruction during sleep, leading to hypoxia and serious health consequences. Conventional diagnostic methods such as polysomnography and drug-induced sleep endoscopy (DISE) are often costly, invasive, or time-consuming. This study aimed to develop a safe, rapid, and fully automated AI-assisted platform for OSA detection using nasopharyngoscopic videos acquired during wakefulness, and to propose diagnostic criteria comparable to the apnea–hypopnea index (AHI). Methods: Flexible nasopharyngoscopic videos of supine, awake patients were analyzed using an AI system comprising four Xception-based image classifiers to identify scan boundaries and classify anatomical regions (nasopharynx, velopharyngeal/oropharyngeal \[VO] wall, tongue base, and epiglottis \[TE]). Five U-Net-based semantic segmentation models were then applied to extract quantitative airway features. Key variables—including maximum and minimum airway cross-sectional areas, VO wall area ratio, and TE ratio—were entered into a support vector regression model to predict OSA. A total of 103 clinical samples (59 non-OSA, 44 OSA) were analyzed, with 35 cases reserved for testing. Results: Classification accuracy for nostril, VO/TE, vocal fold, and nasopharynx regions was 100%, 95.8%, 95%, and 98.5%, respectively. The mean intersection over union (IoU) for segmentation models reached 82.72%. The prediction model achieved 97.14% accuracy on the test set. OSA-associated thresholds were identified—VO wall area ratio < 0.41 and TE ratio > 36.97—all comparable to AHI-based diagnosis. The complete diagnostic workflow, including video upload, classification, segmentation, and prediction, was completed in an average of 85 seconds. Conclusion: This study is the first to implement a fully automated, AI-based dynamic endoscopic video analysis for OSA detection in awake patients. The system accurately predicts OSA and localizes potential obstruction sites in a non-invasive, real-time manner, offering a practical outpatient screening tool to help select candidates who require further evaluation with polysomnography or DISE.

This study deals with the development of a computer tomography (CT) system for automatic segmentation and quantitative analysis of the pulmonary bronchus. It includes three parts. Part I employed an adaptive median and four neighbors low pass filters to eliminate the noise of CT. Then, k-means clustering was used to segment the lung region in the CT data. In Part II, the pulmonary airway was segmented. The three-grade segmentation was employed to divide all pixels in the lung region into three uncertain grades, including air, blood vessels, and tissues, and uncertain portions. The airway wall was reformed using a border pixel weight mask. Afterwards, the seed was calculated automatically with the front-end image masking the aggregation position of the lung region as the input of the region growing to obtain the initial airway. Afterwards, the micro bronchi with different radii were detected using morphological grayscale reconstruction to modify the initial airway. Part III adopted skeletonization to simplify the pulmonary airway, keeping the length and extension direction information. The information was recorded in a linked list with the world coordinates based on the patients’ carina, defined by the directions of the carina to the top end of the trachea and right and left main bronchi. The whole set of bronchi was recognized by matching the target bronchus direction and world coordinates using hierarchical classification. The proposed system could detect the location of the pulmonary airway and detect 11 generations’ bronchi with a bronchus recognition capability of 98.33%. Meanwhile, 20 airway parameters’ measurement and 3D printing verification have been processed. The diameter, length, volume, angle, and cross-sectional area of the main trachea and the right and left bronchi, the cross-sectional area of the junction, the left bronchus length, and the right bronchus length have been calculated for clinical practice guidelines. The system proposed in this study simultaneously maintained the advantages of automation and high accuracy and contributed to clinical diagnosis.

Background/Objectives: Otitis media (OM), including acute otitis media (AOM) and chronic otitis media (COM), is a common middle ear disease that can lead to significant morbidity if not accurately diagnosed. Otoscopic interpretation remains subjective and operator-dependent, underscoring the need for objective and reproducible diagnostic support. Recent advances in artificial intelligence (AI) offer promising solutions for automated otoscopic image analysis. Methods: We developed an AI-based diagnostic framework consisting of three sequential steps: (1) semi-supervised learning for automatic recognition and semantic segmentation of tympanic membrane structures, (2) region-based feature extraction, and (3) disease classification. A total of 607 clinical otoscopic images were retrospectively collected, including normal ears (n = 220), AOM (n = 157), and COM with tympanic membrane perforation (n = 230). Among these, 485 images were used for training and 122 for independent testing. Semantic segmentation of five anatomically relevant regions was performed using multiple convolutional neural network architectures, including U-Net, PSPNet, HRNet, and DeepLabV3+. Following segmentation, color and texture features were extracted from each region and used to train a neural network-based classifier to differentiate disease states. Results: Among the evaluated segmentation models, U-Net demonstrated superior performance, achieving an overall pixel accuracy of 96.76% and a mean Dice similarity coefficient of 71.68%. The segmented regions enabled reliable extraction of discriminative chromatic and texture features. In the final classification stage, the proposed framework achieved diagnostic accuracies of 100% for normal ears, 100% for AOM, and 91.3% for COM on the independent test set, with an overall accuracy of 96.72%. Conclusions: This study demonstrates that a semi-supervised, segmentation-driven AI pipeline integrating feature extraction and classification can achieve high diagnostic accuracy for otitis media. The proposed framework offers a clinically interpretable and fully automated approach that may enhance diagnostic consistency, support clinical decision-making, and facilitate scalable otoscopic assessment in diverse healthcare screening settings for disease prevention and health education.