Maxillectomy patients' speech and performance of contemporary speaker-independent automatic speech recognition platforms in Japanese.

BACKGROUND: Automatic speech recognition (ASR) can potentially help older adults and people with disabilities reduce their dependence on others and increase their participation in society. However, maxillectomy patients with reduced speech intelligibility may encounter some problems using such technologies. OBJECTIVES: To investigate the accuracy of three commonly used ASR platforms when used by Japanese maxillectomy patients with and without their obturator placed. METHODS: Speech samples were obtained from 29 maxillectomy patients with and without their obturator and 17 healthy volunteers. The samples were input into three speaker-independent speech recognition platforms and the transcribed text was compared with the original text to calculate the syllable error rate (SER). All participants also completed a conventional speech intelligibility test to grade their speech using Taguchi's method. A comprehensive articulation assessment of patients without their obturator was also performed. RESULTS: Significant differences in SER were observed between healthy and maxillectomy groups. Maxillectomy patients with an obturator showed a significant negative correlation between speech intelligibility scores and SER. However, for those without an obturator, no significant correlations were observed. Furthermore, for maxillectomy patients without an obturator, significant differences were found between syllables grouped by vowels. Syllables containing /i/, /u/ and /e/ exhibited higher error rates compared to those containing /a/ and /o/. Additionally, significant differences were observed when syllables were grouped by consonant place of articulation and manner of articulation. CONCLUSION: The three platforms performed well for healthy volunteers and maxillectomy patients with their obturator, but the SER for maxillectomy patients without their obturator was high, rendering the platforms unusable. System improvement is needed to increase accuracy for maxillectomy patients.

Digital design of open hollow obturator framework using existing dentures: A dental technique.

Computer-aided design and computer-aided manufacturing technology has been applied to the fabrication of maxillary obturator frameworks, both directly and indirectly. However, with earlier techniques, it was not possible to accurately determine the position of the framework conforming to the palate, an issue that has been resolved in current fabrication methods. Using the patient's existing denture, prosthodontists can determine where the framework should be positioned in the defect area. This allows the obturator bulb to be hollowed, thereby reducing weight and making adjustment easier. The most appropriate position for the finish line can be determined by accurately establishing the arrangement of the artificial teeth as well as the most appropriate polishing surface morphology. In maxillofacial prosthetics, restoring proper articulation and the swallowing function through rehabilitation is important, and determining the proper palatal morphology enables good tongue movement and facilitates the restoration of adequate function. The lighter weight contributes to protecting the remaining teeth and improves patient comfort.

Dental and bony changes following subtotal glossectomy for cancer: A clinical report.

The impact of tongue size and forces on surrounding tissues have been described, but clinical reports on the influence on dentoalveolar structures after glossectomy for cancer are lacking. This report describes the dental and bony changes after subtotal glossectomy for squamous cell carcinoma by comparing 3-dimensional models generated from computed tomography scans at various time points during the clinical course. Surface comparison showed lingual tipping of the teeth and areas of bone surface remodeling on the alveolar bone, basilar bone, coronoid process, and condyles. Early denture planning or splinting teeth before glossectomy could be encouraged in some patients, and clinical studies are recommended.

Predicting maximum occlusal force and tongue pressure using decision tree analysis in patients diagnosed with head and neck tumors.

STATEMENT OF PROBLEM: Preserving and restoring oral functions, especially mastication and swallowing, is important to the quality of life of patients being treated for head and neck tumors. Studies that help predict maximum occlusal force and tongue pressure during prosthetic treatment, necessary for providing comprehensive, appropriate treatment and encouraging patient adherence and confidence are lacking. PURPOSE: The purpose of this clinical study was to develop a decision tree model for predicting maximum occlusal force and tongue pressure in patients diagnosed with head and neck tumors that could help both experienced and less experienced prosthodontists and oral surgeons optimize the treatment plan and support patient compliance and their quality of life. MATERIAL AND METHODS: A total of 80 patients who had been treated for head and neck tumors were enrolled in the study. Their maximum occlusal force was measured using a pressure-sensitive film and tongue pressure using a tongue pressure measurement device. Data, including basic characteristics, were transferred to a comma separated values file, which was then imported into a statistical software package to produce a decision tree. The classification and regression tree method was used to construct a predictive model. RESULTS: The number of occlusal contacts associated with not wearing a prosthesis, flap reconstruction, radiotherapy, chemotherapy, the number of teeth present, age, tumor stage, and tumor type were found to be associated with maximum occlusal force, with a prediction accuracy of 96.3%, area under the receiver operating characteristic curve of 0.99, sensitivity of 97%, and specificity of 94%. The number of occlusal contacts associated with wearing and not wearing a prosthesis, tumor stage, age, radiotherapy, and surgery type were found to be associated with tongue pressure, with a prediction accuracy of 96.3%, area under the receiver operating characteristic curve of 0.97, sensitivity of 97%, and specificity of 93%. CONCLUSIONS: The decision tree model can be an effective tool for the prediction of maximum occlusal force and tongue pressure in patients diagnosed with head and neck tumors, helping both experienced and less experienced prosthodontists and oral surgeons to provide early, appropriate, and necessary treatment before starting prosthetic treatment and helping patients with treatment compliance and communication with medical staff.

Advancing maxillofacial prosthodontics by using pre-trained convolutional neural networks: Image-based classification of the maxilla.

PURPOSE: The study aimed to compare the performance of four pre-trained convolutional neural networks in recognizing seven distinct prosthodontic scenarios involving the maxilla, as a preliminary step in developing an artificial intelligence (AI)-powered prosthesis design system. MATERIALS AND METHODS: Seven distinct classes, including cleft palate, dentulous maxillectomy, edentulous maxillectomy, reconstructed maxillectomy, completely dentulous, partially edentulous, and completely edentulous, were considered for recognition. Utilizing transfer learning and fine-tuned hyperparameters, four AI models (VGG16, Inception-ResNet-V2, DenseNet-201, and Xception) were employed. The dataset, consisting of 3541 preprocessed intraoral occlusal images, was divided into training, validation, and test sets. Model performance metrics encompassed accuracy, precision, recall, F1 score, area under the receiver operating characteristic curve (AUC), and confusion matrix. RESULTS: VGG16, Inception-ResNet-V2, DenseNet-201, and Xception demonstrated comparable performance, with maximum test accuracies of 0.92, 0.90, 0.94, and 0.95, respectively. Xception and DenseNet-201 slightly outperformed the other models, particularly compared with InceptionResNet-V2. Precision, recall, and F1 scores exceeded 90% for most classes in Xception and DenseNet-201 and the average AUC values for all models ranged between 0.98 and 1.00. CONCLUSIONS: While DenseNet-201 and Xception demonstrated superior performance, all models consistently achieved diagnostic accuracy exceeding 90%, highlighting their potential in dental image analysis. This AI application could help work assignments based on difficulty levels and enable the development of an automated diagnosis system at patient admission. It also facilitates prosthesis designing by integrating necessary prosthesis morphology, oral function, and treatment difficulty. Furthermore, it tackles dataset size challenges in model optimization, providing valuable insights for future research.

Factors Related to Masticatory Rhythm in Patients with Oral Tumors.

Background: Older adults who have undergone surgery for oral tumors are at increased risk of impaired masticatory rhythm. This study investigated the correlations between masticatory rhythm, objective masticatory performance, and subjective masticatory performance as well as factors related to masticatory rhythm. Methods: The participants were 44 adults (24 men, 20 women; age range 42~90 years old) who had undergone maxillectomy, mandibulectomy, or glossectomy and were rehabilitated with a maxillofacial prosthesis. The number of functional contact teeth pairs was confirmed by intraoral examination. Chewing rate, cycle duration, coefficient of variation (CV) for cycle duration (reflecting the stability of masticatory rhythm), and mixing ability were measured simultaneously using a mastication movement rhythm tracking device during gum chewing. Maximum occlusal force was measured using the dental prescale system. Patients' perception of chewing ability was rated using a questionnaire. Results: The Spearman's rank correlation test revealed that mixing ability, patient-rated masticatory scores, cycle duration, CV for cycle duration, and maximum occlusal force showed significant correlations with chewing rate. Multiple linear regression analysis identified mixing ability and the CV for cycle duration as significant predictors of masticatory rhythm. Conclusions: Factors associated with a faster chewing rate were higher mixing ability and masticatory scores, greater maximum occlusal force, shorter cycle duration, and smaller CV for cycle duration. Stable masticatory rhythm and mixing ability are significant predictors of chewing rate. Poor masticatory performance and unstable masticatory rhythm can result in slower chewing and thus a higher risk of inadequate dietary intake.

Assessment of the color stability of two silicone elastomers for maxillofacial prostheses upon exposure to cold beverages.

PURPOSE: This study investigated the color stability of different commercially available silicone materials for facial prostheses upon exposure to everyday beverages. It was hypothesized that the beverages would not alter the color of the silicone under conditions simulating daily exposure. METHODS: Sixty specimens were fabricated using two materials - VST-50 and Silfy - and exposed to commercially available cold mineral water, tea, or coffee. Specimen color was evaluated on days 1 and 15. The color was measured with a spectrophotometer based on CIELab system and color differences were calculated as ΔE. Statistical analysis was performed using the Kruskal-Wallis test and Mann-Whitney U test. RESULTS: The mean ΔE values after 15 days of exposure to mineral water, cold green tea, and coffee were 1.016, 3.480, and 3.636 for VST-50 and 0.440, 0.798, and 1.425 for Silfy, respectively. Both materials showed significant differences in color, and VST-50 showed a greater color change than Silfy, especially for coffee. CONCLUSION: Pigmented silicone elastomers have low color stability, leading to an overall color change in silicone prostheses when exposed to pigmented beverages. Color changes in such prostheses can be mitigated by selecting materials with better color stability to extend their longevity.

Evaluation of the color stability of three maxillofacial silicone materials after exposure to beverages: An in vitro study.

Purpose: Oral cancer often requires treatments like surgical intervention, involving surgical resection of lips and other facial parts. For those patients, prosthetic rehabilitation following surgery promotes social reintegration. This study explores the color stability of various commercially available silicone materials for facial prostheses when exposed to everyday beverages like coffee and tea. The hypothesis is that these beverages can induce color changes in silicones, simulating conditions of daily use. Material and methods: A total of 90 specimens were fabricated. There were 30 specimens deefor each of 3 nonpigmented silicone elastomers: Silfy, A-2186-F and VST-50 and were immersed in tea or coffee at drinking temperature and evaluated for color changes at time points of 0 h, 1 h, and 6 h. The color was measured with a colorimeter that used the CIE L*a*b* system and the color change ΔΕ was calculated to quantify the color change. Statistical analysis was performed by using the Kruskal-Wallis test, the Mann-Whitney U test, and the Wilcoxon signed-rank Test. The threshold for statistical significance was P < .05. Results: Color change was observed for each of the silicones exposed to the beverages (P < .05). The value of ΔE was significantly higher for A-2186-F compared with VST-50 and Silfy (P < .05 in all beverages at 6 h). The color of the elastomers progressively changed while immersed in the beverages over 6 h. The change in color was significantly greater for coffee compared with tea (P < .05). Conclusions: The color stability of nonpigmented silicone elastomers is inherently low, which contributes to the overall color change of silicone prostheses when they are exposed to beverages that contain pigments. The elastomer A-2186-F had greatest values of ΔE among the materials tested. Color changes of silicone prostheses can be lessened by selecting materials with consideration of color stability to improve their longevity and extend their period of use.

Machine Learning Prediction of Tongue Pressure in Elderly Patients with Head and Neck Tumor: A Cross-Sectional Study.

Background: This investigation sought to cross validate the predictors of tongue pressure recovery in elderly patients' post-treatment for head and neck tumors, leveraging advanced machine learning techniques. Methods: By employing logistic regression, support vector regression, random forest, and extreme gradient boosting, the study analyzed an array of variables including patient demographics, surgery types, dental health status, and age, drawn from comprehensive medical records and direct tongue pressure assessments. Results: Among the models, logistic regression emerged as the most effective, demonstrating an accuracy of 0.630 [95% confidence interval (CI): 0.370-0.778], F1 score of 0.688 [95% confidence interval (CI): 0.435-0.853], precision of 0.611 [95% confidence interval (CI): 0.313-0.801], recall of 0.786 [95% confidence interval (CI): 0.413-0.938] and an area under the receiver operating characteristic curve of 0.626 [95% confidence interval (CI): 0.409-0.806]. This model distinctly highlighted the significance of glossectomy (p = 0.039), the presence of functional teeth (p = 0.043), and the patient's age (p = 0.044) as pivotal factors influencing tongue pressure, setting the threshold for statistical significance at p < 0.05. Conclusions: The analysis underscored the critical role of glossectomy, the presence of functional natural teeth, and age as determinants of tongue pressure in logistics regression, with the presence of natural teeth and the tumor site located in the tongue consistently emerging as the key predictors across all computational models employed in this study.

Effect of ethylene oxide unit number in bis-EMA on the physical properties of additive-manufactured occlusal splint material.

PURPOSE: To investigate the effects of the number of ethylene oxide units in bis-EMA on the physical properties of additively manufactured occlusal splints. METHODS: Seven experimental materials containing bis-EMAs with three and 10 ethylene oxide units (BE3 and BE10, respectively) were prepared at different BE10 content rates (BE10-0%, -20%, -30%, -40%, -50%, -60%, and -80%). Half the specimens of each material were aged in boiling water. Flexural strength (FS), flexural modulus (FM), fracture toughness (FT), microwear depth (MD), degree of conversion (DC), water sorption (WSP), water solubility (WSL), color difference between non-aged and aged series (ΔE), and translucency (TP) were evaluated. All the evaluated properties other than FS and MD were analyzed by 1-way ANOVA and Tukey's post hoc analysis, while FS and MD were analyzed by Kruskal-Wallis's test and Bonferroni correction (α=0.05). RESULTS: BE10-80% revealed the lowest FS (P < 0.01 for BE10-0%, -20%, and -30%) and FM (P < 0.01, for all), while revealing the highest DC, WSP, WSL (P < 0.01 for all) and TP (P < 0.01 for all other than BE10-60%). BE10-50% showed the highest FT (P < 0.01 for all). BE10-50%, -60%, and -80% revealed significantly lower ΔE than others (P < 0.01) and lower MD than BE10-0% (P < 0.05). Regardless of the BE10 content, FS, FM, and FT decreased with aging. CONCLUSIONS: The number of ethylene oxide units affects the physical properties of additively manufactured occlusal splints. The higher number of ethylene oxide units in bis-EMA enhanced the microwear resistance, DC, WSP, WSL, color stability, and translucency, whereas it deteriorated the FS and FM.