Diagnostics

Background/Objectives: Intraoral scanners (IOSs) are essential tools in digital dentistry; however, their accuracy remains influenced by clinical conditions such as restricted access, patient movement, or intraoral moisture. Intraoral scanning is performed within a confined space that restricts scanner motion, potentially influencing maneuverability during data acquisition and, consequently, IOS performance. This study investigated the impact of maneuverability constraints on the trueness accuracy and efficiency of IOS under clinically representative intraoral conditions. Methods: Fifteen participants with no previous experience in intraoral scanning or device operation were recruited. Each participant scanned a maxillary full-dentition typodont model and a mandibular implant-containing typodont model using the Aoralscan 3 IOS. Scans were performed under two conditions: constrained intraoral scanning within a manikin and open-vision extraoral scanning on a bench-top. Trueness accuracy was evaluated using three parameters: the root mean square (RMS) deviation of the maxillary dentition, discrepancies in inter–scan body distances, and angular deviations of the scan bodies, each calculated by comparison with reference data obtained from an industrial-grade scanner. Scan time was recorded to assess time-based efficiency. Results: No significant differences were observed in RMS trueness, inter-implant distances, or implant angular deviations between intraoral and extraoral scans. Extraoral scanning significantly reduced scan times for both maxillary and mandibular models (p < 0.0001). Conclusions: Within the limitations of this study, maneuverability constraints alone may not significantly affect IOS trueness accuracy compared with open bench-top scanning. However, scanning efficiency was reduced under intraoral scanning constraints, with longer scan times observed among inexperienced operators. The potential influence of intraoral factors other than maneuverability on IOS accuracy under clinical conditions warrants further investigation.

Background/Objectives: Lung cancer remains the leading cause of cancer-related deaths worldwide, with Non-Small Cell Lung Cancer (NSCLC) accounting for the majority of cases, primarily Squamous Cell Carcinoma (SCC) and Adenocarcinoma (ADC). The aim of this systematic review is to summarise and critically appraise the performance of machine learning (ML)-based radiomics models in the differential diagnosis and overall survival analysis for lung SCC and ADC. Methods: PRISMA standards were followed in conducting the review. The quality of the studies was assessed using the Radiomics quality score (RQS) tool. Results: A total of 11 studies were included, demonstrating that deep learning and radiomics-based machine learning models significantly improve the non-invasive classification of lung squamous cell carcinoma and adenocarcinoma. Deep learning systems achieved an accuracy of 67–97%, and machine learning models showed an accuracy of 75–87%. The integration of radiomic features further enhanced diagnostic precision, showing strong potential for reliable histologic subtype differentiation. Conclusions: Machine learning-based radiomics models and deep learning significantly enhance the non-invasive, accurate differentiation of lung squamous and adenocarcinoma cell carcinoma when combined with clinical and pathological data.

Background: Parkinson’s disease (PD) is a neurodegenerative disorder that affects millions of people worldwide. Speech analysis has emerged as a non-invasive tool for automatic PD detection; however, the scarcity and homogeneity of available datasets often limit the generalization capability of machine learning models, motivating the use of data augmentation strategies to improve robustness. Methods: This study presents a data augmentation-based methodology for speech-based classification between PD patients and healthy control subjects. A deep learning model trained from scratch on Mel spectrograms is evaluated using augmentation techniques applied at both the waveform and time–frequency levels. Multiple training and model selection strategies are analyzed and model performance is assessed through internal validation as well as using an independent dataset Results: Experimental results show that carefully selected data augmentation techniques improve classification performance with respect to the non-augmented counterpart, achieving gains of up to 3% in accuracy. However, when evaluated on an independent dataset, these improvements do not consistently translate into better generalization. Conclusions: These findings demonstrate that, while data augmentation can effectively enhance model performance within a single dataset, this apparent robustness is not sufficient to guarantee generalization on independent speech corpora for PD detection.