Automated evaluation of masseter muscle volume: deep learning prognostic approach in oral cancer.

BACKGROUND: Sarcopenia has been identified as a potential negative prognostic factor in cancer patients. In this study, our objective was to investigate the relationship between the assessment method for sarcopenia using the masseter muscle volume measured on computed tomography (CT) images and the life expectancy of patients with oral cancer. We also developed a learning model using deep learning to automatically extract the masseter muscle volume and investigated its association with the life expectancy of oral cancer patients. METHODS: To develop the learning model for masseter muscle volume, we used manually extracted data from CT images of 277 patients. We established the association between manually extracted masseter muscle volume and the life expectancy of oral cancer patients. Additionally, we compared the correlation between the groups of manual and automatic extraction in the masseter muscle volume learning model. RESULTS: Our findings revealed a significant association between manually extracted masseter muscle volume on CT images and the life expectancy of patients with oral cancer. Notably, the manual and automatic extraction groups in the masseter muscle volume learning model showed a high correlation. Furthermore, the masseter muscle volume automatically extracted using the developed learning model exhibited a strong association with life expectancy. CONCLUSIONS: The sarcopenia assessment method is useful for predicting the life expectancy of patients with oral cancer. In the future, it is crucial to validate and analyze various factors within the oral surgery field, extending beyond cancer patients.

Machine Learning Analysis of Gaze Data for Enhanced Precision in Diagnosing Oral Mucosal Diseases.

The diagnosis of oral mucosal diseases is a significant challenge due to their diverse differential characteristics. Risk assessment of lesions by visual examination is a complex process due to the lack of definitive guidelines. This study aimed to improve this process by creating a diagnostic algorithm using gaze data acquired during oral mucosal disease examinations. A total of 78 dentists were included in this study. Tobii Pro Nano® (Tobii Technology) was used to acquire gaze data during clinical photographic visual examinations. Advanced analysis tools such as support vector machines and heatmaps were used to visualize the gazing tendencies of a group of skilled oral surgeons, focusing on the number of gazes per region and the gazing time ratios. The preliminary findings showed the possibility of visualizing gazing tendencies and identifying areas of importance for diagnosis. The classification of intraoral photographs based on gross features revealed the existence of an optimal examination method for each category and diagnostically significant areas. This novel approach to analyzing gaze data has the potential to refine diagnostic techniques and increase both accuracy and efficiency.

Accuracy and dimensional reproducibility by model scanning, intraoral scanning, and CBCT imaging for digital implant dentistry.

BACKGROUND: Recently, it has become possible to analyze implant placement position using the digital matching data of optical impression data of the oral cavity or plaster models with cone beam computed tomography (CBCT) data, and create a highly accurate surgical guide. It has been reported that CBCT measurements were smaller than the actual values, termed shrinkage. Matching of digital data is reliable when the plaster model or intraoral impression values show shrinkage at the same rate as the CBCT data. However, if the shrinkage rate is significantly different, the obtained digital data become unreliable. To clarify digital matching reliability, we examined dimensional reproducibility and shrinkage in measurements obtained with a model scanner, intra-oral scanner (iOS), and CBCT. MATERIALS AND METHODS: Three implants that were arranged in a triangle were fixed in an acrylic plate. The distance between each implants were measured using model scanner, iOS, and CBCT. The actual size measured by electronic caliper was regarded as control. RESULTS: All values measured with CBCT were significantly smaller than that of model scanner, iOS, and control (p<0.001). The model scanner shrinkage was 0.37-0.39%, iOS shrinkage was 0.9-1.4%, and CBCT shrinkage was 1.8-6.9%. There were statistically significant differences among the shrinkage with iOS, CBCT, and model scanner (p<0.001). CONCLUSION: Our findings showed that all measurements obtained with those modalities showed shrinkage as compared to the actual values. In addition, CBCT shrinkage was largest among three different measuring methods. They indicated that data matching between CBCT and scanner measurements requires attention in regard to the reliability of values obtained with those devices.

Relationship Between Kinesthetic/Visual Motor Imagery Difficulty and Event-Related Desynchronization/Synchronization.

Motor imagery (MI) is divided into two types: kinesthetic (KMI) and visual (VMI). To estimate the MI that an examinee performs, event-related desynchronization (ERD) or event-related synchronization (ERS) is used to characterize KMI or VMI via electroencephalogram (EEG). However, no definitive method using ERD/ERS via EEG has been established yet to estimate the type of MI performed. This is because the MI performed by the examinee is not always the same as that instructed by the examiner. One of the reasons for this mismatch is the difficulty of MI, especially KMI. However, almost no reported studies have considered MI difficulty to estimate MI type. Therefore, in this study, we examined the relationship between MI difficulty and the ERD/ERS pattern corresponding to the type of MI in the case of single flexion of the right index finger (SFRIF). The results showed that for a subject who felt MI was less difficult, the α-band ERD value (αERD) at the electrode of the occipital area (O1 or O2 site) of the KMI instruction was significantly smaller than that of the VMI instruction. On the contrary, for a subject who felt MI was very difficult, αERD at the O1 or O2 site on the KMI instruction was similar to that of the VMI instruction. In addition, for the subject who felt MI was easy, the αERDs at the O1 or O2 site on the KMI and VMI instructions were similar to those on the movement execution (ME) and movement observation (MO) instructions, respectively. Therefore, in the case of SFRIF, it was suggested that MI difficulty could be estimated by ERD/ERS patterns in the occipital area. This was supported by referring to the ME and MO ERD/ERS patterns in the occipital area.