Landmark annotation through feature combinations: a comparative study on cephalometric images with in-depth analysis of model's explainability.

OBJECTIVES: The objectives of this study are to explore and evaluate the automation of anatomical landmark localization in cephalometric images using machine learning techniques, with a focus on feature extraction and combinations, contextual analysis, and model interpretability through Shapley Additive exPlanations (SHAP) values. METHODS: We conducted extensive experimentation on a private dataset of 300 lateral cephalograms to thoroughly study the annotation results obtained using pixel feature descriptors including raw pixel, gradient magnitude, gradient direction, and histogram-oriented gradient (HOG) values. The study includes evaluation and comparison of these feature descriptions calculated at different contexts namely local, pyramid, and global. The feature descriptor obtained using individual combinations is used to discern between landmark and nonlandmark pixels using classification method. Additionally, this study addresses the opacity of LGBM ensemble tree models across landmarks, introducing SHAP values to enhance interpretability. RESULTS: The performance of feature combinations was assessed using metrics like mean radial error, standard deviation, success detection rate (SDR) (2 mm), and test time. Remarkably, among all the combinations explored, both the HOG and gradient direction operations demonstrated significant performance across all context combinations. At the contextual level, the global texture outperformed the others, although it came with the trade-off of increased test time. The HOG in the local context emerged as the top performer with an SDR of 75.84% compared to others. CONCLUSIONS: The presented analysis enhances the understanding of the significance of different features and their combinations in the realm of landmark annotation but also paves the way for further exploration of landmark-specific feature combination methods, facilitated by explainability.

Effect of selective trimming of nasoavleolar molding on maxillary arch growth in patients with unilateral cleft lip and palate.

AIM: Presurgical Nasoalveolar molding (PNAM) is a technique used for cleft lip and palate patients prior to cheiloplasty. However, concerns exist regarding its negative impact on maxillary arch growth.This study aimed to assess the effect of selective trimming in NAM on maxillary arch growth in patients with unilateral cleft lip palate. METHODOLOGY: The retrospective observational study analyzed the study casts of 30 patients before and after undergoing nasoalveolar molding treatment. Study casts which were repositories of the institute were analyzed and segregated as Group A: NAM given with selective trimming of the appliance, and Group B: NAM without selective trimming of the appliance. Pre and post-treatment casts were assessed digitally at L-L' (Intercleft segment width), C-C' (Intercanine width), T-T' (Alveolar arch width), L-TT' (Alveolar arch length from major cleft segment), and L'-TT' (Alveolar arch length from minor cleft segment) RESULTS: A notable significant difference between Group A's and B's mean Alveolar arch width and Intercleft segment width was seen. Whereas parameters such as intercanine width, and alveolar arch length as functions from major and minor segments showed no significant variance. CONCLUSION: Although NAM has been known to affect the growth of the maxillary arch, this study proposes that techniques such as selective trimming can help counteract this drawback.

Is Training Working Memory in Children with Learning Disabilities a Viable Solution? A Systematic Review.

Background: Working memory (WM) is one of the most influential cognitive functions in encoding, registering, and retrieving information. It influences the learning process in children. Its role becomes essential, especially in a child with a learning disability (LD). Researchers worldwide are giving much prominence to WM, especially in devising cognitive retraining strategies for better cognitive functioning and academic attainment in these children. This current study aims to explore globally used instruments to measure this construct and review effective WM training models in the cognitive rehabilitation of children with LD. This study used a systematic review, availing the elaborate "Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA)" guidelines. Summary: The databases of Google Scholar, PubMed, and Web of Science were searched thoroughly, and those studies, which met the inclusion criteria, were considered for this review. Out of 770 studies found with keywords, only six met the inclusion criteria and were selected for a detailed analysis. The outcome of the current review provides trustworthy evidence of poor performance, especially in tasks involving verbal and executive WM in children with all types of learning disabilities (LD) and difficulties. The studies reviewed support the hypothesis that WM can improve with training and significantly improve children's academic attainment. Key Message: Further this review recommends that research and efforts must go into devising these cognitive training techniques. Children have high cerebral plasticity; hence, using cognitive training (emphasizing WM training and other cognitive functions) with them would enhance their cognitive functioning and capacity, improving their academic performance.

Lateral Cephalometric Landmark Annotation Using Histogram Oriented Gradients Extracted from Region of Interest Patches.

Introduction: Two-dimensional cephalometric image analysis plays a crucial role in orthodontic diagnosis and treatment planning. While deep learning-based algorithms have emerged to automate the laborious task of anatomical landmark annotation, their effectiveness is hampered by the challenges of acquiring and labelling clinical data. In this study, we propose a model that leverages conventional machine learning techniques to enhance the accuracy of landmark detection using limited dataset. Materials and methods: Our methodology involves coarse localization through region of interest (ROI) extraction and fine localization utilizing histogram-oriented gradient (HOG) feature. The image patch containing landmark pixels is classified using the light gradient boosting machine (LGBM) algorithm. To evaluate our model's performance, we conducted rigorous tests on the ISBI Cephalometric dataset and Dental Cepha dataset, aiming to achieve accuracy within a 2 mm radial precision range. We also employed cross-validation to assess our approach, providing a robust evaluation. Results: Our model's performance on the ISBI Cephalometric dataset showed an accuracy rate of 77.11% within the desired 2 mm radial precision range. The cross-validation results further confirmed the effectiveness of our approach, yielding a mean accuracy of 78.17%. Additionally, we applied our model to the Dental Cepha dataset, where we achieved a remarkable landmark detection accuracy of 84%. Conclusion: The results demonstrate that traditional machine learning techniques can be effective for accurate landmark detection in cephalometric images, even with limited data. Our findings highlight the potential of these techniques for clinical applications, where large datasets of labelled images may not be available.

Establishing Cephalometric Norms in Primary Dentition Using Comprehensive Craniofacial Growth Analysis - A Digital Cephalometric Study

ABSTRACT Objective: To establish cephalometric norms in primary dentition among males and females using novel customized Comprehensive Cephalometric Growth (CCG) Analysis. Material and Methods: The study was conducted on 67 subjects with a mean age of 5.5 yrs. Digital lateral cephalometric radiographs were obtained using Planmeca Pro One. The digital images were then transferred to Nemoceph software. Craniofacial Growth (CCG) Analysis was configured in the software with five sub-groups. This sub-grouping was done such that related components were grouped together and comprehensively; it would provide an assessment of every component of the craniofacial region that could be affected either by treatment maneuver or growth process. The same was used for the cephalometric analysis and to determine the cephalometric norms in the primary dentition. Results: Certain linear measurements were higher among males when compared to females. However, most measurements remained similar among males and females during this age group. The CCG analysis provided a comprehensive knowledge of the craniofacial parameters during the growth process. Conclusion: The cephalometric norms during primary dentition thus established using Comprehensive Craniofacial Growth analysis would provide the data for early diagnosis and treatment planning in interceptive orthodontic treatment procedures.