Sex assessment using the radius bone in a French sample when applying various statistical models.

Sex estimation of skeletal remains is one of the most important tasks in forensic anthropology. The radius bone is useful to develop standard guidelines for sex estimation across various populations and is an alternative when coxal or femoral bones are not available.The aim of the present study was to assess the sexual dimorphism from radius measurements in a French sample and compare the predictive accuracy of several modelling techniques, using both classical statistical methods and machine learning algorithms.A total of 78 left radii (36 males and 42 females) were used in this study. Sixteen measurements were made. The modelling techniques included a linear discriminant analysis (LDA), flexible discriminant analysis (FDA), regularised discriminant analysis (RDA), penalised logistic regression (PLR), random forests (RF) and support vector machines (SVM).The different statistical models showed an accuracy of classification that is greater than 94%. After selection of variables, the accuracies increased to 97%. The measurements made at the proximal part of the radius (sagittal and transversal diameters of the head, and sagittal diameter of the neck), at distal part (maximum width of the distal epiphysis) and of the entire bone (maximum length) stand out among the various models.The present study suggests that the radius bone constitutes a valid alternative for sex estimation of skeletal remains with comparable classification accuracies to the pelvis or femur and that the non-classical statistical models may provide a novel approach to sex estimation from the radius bone. However, the extrapolation of the current results cannot be made without caution because our sample was composed of very aged individuals.

Sex estimation from long bones: a machine learning approach.

Sex estimation from skeletal remains is one of the crucial issues in forensic anthropology. Long bones can be a valid alternative to skeletal remains for sex estimation when more dimorphic bones are absent or degraded, preventing any estimation from the first intention methods. The purpose of this study was to generate and compare classification models for sex estimation based on combined measurement of long bones using machine learning classifiers. Eighteen measurements from four long bones (radius, humerus, femur, and tibia) were taken from a total of 2141 individuals. Five machine learning methods were employed to predict the sex: a linear discriminant analysis (LDA), penalized logistic regression (PLR), random forest (RF), support vector machine (SVM), and artificial neural network (ANN). The different classification algorithms using all bones generated highly accuracy models with cross-validation, ranging from 90 to 92% on the validation sample. The classification with isolated bones ranked between 83.3 and 90.3% on the validation sample. In both cases, random forest stands out with the highest accuracy and seems to be the best model for our investigation. This study upholds the value of combined long bones for sex estimation and provides models that can be applied with high accuracy to different populations.

Sex estimation from the greater sciatic notch: a comparison of classical statistical models and machine learning algorithms.

The greater sciatic notch (GSN) is a useful element for sex estimation because it is quite resistant to damage, and thus it can often be assessed even in poorly preserved skeletons. This study aimed to develop statistical models for sex estimation based on visual and metric analyses of the GSN, and additional variables linked to the GSN. A total of 60 left coxal bones (30 males and 30 females) were analysed. Fifteen variables were measured, and one was a morphologic variable. These 16 variables were used for the comparison of six statistical models: linear discriminant analysis (LDA), regularized discriminant analysis (RDA), penalized logistic regression (PLR) and flexible discriminant analysis (FDA), and two machine learning algorithms, support vector machine (SVM) and artificial neural network (ANN). The statistical models were built in two steps: firstly, only with the GSN variables (group 1), and secondly, with the whole variables (group 2), in order to see if the models including all the variables performed better. The overall accuracy of the models was very close, ranging from 0.92 to 0.97 using specific GSN variables. When additional variables starting from the deepest point of GSN are available, it is worth to use them, because the accuracy increases. PLR (after optimization of parameters) stands out from other statistical models. The position of the deepest point of GSN (Fig. 2) probably plays a crucial role for the sexual dimorphism, as stated by the good performance of the visual assessment of this point and the fact that the A2 angle (posterior angle with the deepest point of the GSN as the apex) is included in all models.

Technical note: Interest of focused fields in post-mortem computed tomography using photorealistic images for age at death estimation from the pubic symphysis.

INTRODUCTION: The morphological assessment of the pubic symphysis using the Suchey-Brooks method is considered a reliable age at death indicator. Age at death estimation methods can be adapted to the images obtained from post-mortem computed tomography (PMCT). The aim of this study is to evaluate the utility of pubic symphysis photorealistic images obtained through Global illumination rendering (GIR) for age at death estimation from whole-body PMCT and from focused PMCT on the pubic bone. MATERIALS AND METHODS: We performed virtual age at death estimation using the Suchey Brooks method from both the whole-body field of view (Large Field of View: LFOV) and the pubis-focused field of view (Small and Field of View: SFOV) of 100 PMCT. The 3D photorealistic images were evaluated by three forensic anthropologists and the results were statistically evaluated for accuracy of the two applied PMCT methods and the intra- and inter-observer errors. RESULTS: When comparing the two acquisitions of PMCT, the accuracy rate reaches 98.5% when using a pubic-focused window (SFOV) compared to 86% with a whole-body window (LFOV). Additionally, the intra- and inter-observer variability has demonstrated that the focused window provides better repeatability and reproducibility. CONCLUSION: Adding a pubic-focused field of view to standard PMCT and processing it with GIR appears to be an applicable technique that increases the accuracy rate for age at death estimation from the pubic symphysis.

Risk prediction of kalaemia disturbance and acute kidney injury after total knee arthroplasty: use of a machine learning algorithm.

INTRODUCTION: Total knee arthroplasty (TKA) is a procedure associated with risks of electrolyte and kidney function disorders, which are rare but can lead to serious complications if not correctly identified. A routine check-up is very often carried out to assess the seric ionogram and kidney function after TKA, that rarely requires clinical intervention in the event of a disturbance. The aim of this study was to identify perioperative variables that would lead to the creation of a machine learning model predicting the risk of kalaemia disorders and/or acute kidney injury after total knee arthroplasty. HYPOTHESIS: A predictive model could be constructed to estimate the risk of kalaemia disorders and/or acute kidney injury after total knee arthroplasty. MATERIAL AND METHODS: This single-centre retrospective study included 774 total knee arthroplasties (TKA) operated on between January 2020 and March 2023. Twenty-five preoperative variables were incorporated into the machine learning model and filtered by a first algorithm. The most predictive variables selected were used to construct a second algorithm to define the overall risk model for postoperative kalaemia and/or acute kidney injury (K+ A). Two groups were formed of K+ A and non-K+ A patients after TKA. A univariate analysis was performed and the performance of the machine learning model was assessed by the area under the curve representing the sensitivity of the model as a function of 1 - specificity. RESULTS: Of the 774 patients included who had undergone TKA surgery, 46 patients (5.9%) had a postoperative kalaemia disorder requiring correction and 13 patients (1.7%) had acute kidney injury, of whom 5 patients (0.6%) received vascular filling. Eight variables were included in the machine learning predictive model, including body mass index, age, presence of diabetes, operative time, lowest mean arterial pressure, Charlson score, smoking and preoperative glomerular filtration rate. Overall performance was good with an area under the curve of 0.979 [CI95% 0.938 - 1.02], sensitivity was 90.3% [CI95% 86.2 - 94.4] and specificity 89.7% [CI95% 85.5 - 93.8]. The tool developed to assess the risk of impaired kalaemia and/or acute kidney injury after TKA is available on https://arthrorisk.com. CONCLUSION: The risk of kalaemia disturbance and postoperative acute kidney injury after total knee arthroplasty could be predicted by a model that identifies low-risk and high-risk patients based on eight pre- and intraoperative variables. This machine learning tool is available on a web platform accessible for everyone, easy to use and has a high predictive performance. The aim of the model was to better identify and anticipate the complications of dyskalaemia and postoperative acute kidney injury in high-risk patients. Further prospective multicentre series are needed to assess the value of a systematic postoperative biochemical work-up in the absence of risk predicted by the model. LEVEL OF EVIDENCE: IV; retrospective study of case series.