Machine Learning Models for Prediction of Sex Based on Lumbar Vertebral Morphometry.

BACKGROUND: Identifying skeletal remains has been and will remain a challenge for forensic experts and forensic anthropologists, especially in disasters with multiple victims or skeletal remains in an advanced stage of decomposition. This study examined the performance of two machine learning (ML) algorithms in predicting the person's sex based only on the morphometry of L1-L5 lumbar vertebrae collected recently from Romanian individuals. The purpose of the present study was to assess whether by using the machine learning (ML) techniques one can obtain a reliable prediction of sex in forensic identification based only on the parameters obtained from the metric analysis of the lumbar spine. METHOD: This paper built and tuned predictive models with two of the most popular techniques for classification, RF (random forest) and XGB (xgboost). Both series of models used cross-validation and a grid search to find the best combination of hyper-parameters. The best models were selected based on the ROC_AUC (area under curve) metric. RESULTS: The L1-L5 lumbar vertebrae exhibit sexual dimorphism and can be used as predictors in sex prediction. Out of the eight significant predictors for sex, six were found to be particularly important for the RF model, while only three were determined to be important by the XGB model. CONCLUSIONS: Even if the data set was small (149 observations), both RF and XGB techniques reliably predicted a person's sex based only on the L1-L5 measurements. This can prove valuable, especially when only skeletal remains are available. With minor adjustments, the presented ML setup can be transformed into an interactive web service, freely accessible to forensic anthropologists, in which, after entering the L1-L5 measurements of a body/cadaver, they can predict the person's sex.

Thanatochemical Study of Glycated Hemoglobin in Diabetic Status Assessment.

Background and objectives. In forensic medicine, the postmortem determination of glycated hemoglobin (HbA1c) helps identify undiagnosed cases of diabetes or cases with uncontrolled glycemic status. In order to contribute to the solidification of thanatochemistry, both globally and especially nationally, we aimed to determine this biomarker postmortem, for the first time in our institution, in order to identify undiagnosed pre-mortem diabetics, as well as those with inadequate glycemic control. Materials and Methods. Our research consisted of analyzing a total number of 180 HbA1c values, 90 determinations from the peripheral blood and 90 from the central blood. The determination of HbA1c was performed by means of a fully automatic analyzer (HemoCue HbA1c 501), certified by the National Glycohemoglobin Standardization Program (NGSP)/Diabetes Control and Complications Trial (DCCT) and calibrated according to the standards developed by the International Federation of Clinical Chemistry (IFCC). According to ADA criteria, HbA1c values can provide us with the following information about the diagnosis of diabetes: normal 4.8-5.6%; prediabetes 5.7-6.4%; diabetes ≥ 6.5%. Results. A considerable number of cases with an altered glycemic status (cases that had HbA1c values equal to or greater than 5.7%) were identified-51% demonstrable by peripheral blood determinations and 41% by central blood determinations. Notably, 23 people with diabetes (25%) were identified by analyzing the peripheral blood; 18 other people with diabetes (20%) were identified by analyzing the central blood. Conclusions. Our study managed to confirm the antemortem diagnosis of DM using a simple point-of-care analyzer and applying standardized and certified criteria on HbA1c levels measured postmortem. We also identified a considerable number of cases with DM in patients with no antemortem history of glucose imbalance-at least 20% more cases. Although the two different sites used for blood collection showed a strong statistical correlation, it seems that the peripheral site could have a higher sensibility in detecting postmortem altered glycemic status.

PREDICTORS OF SUICIDE RISK IN INCARCERATED MALE OFFENDERS: THE ROLE OF PERSONALITY DISORDERS.

AIM: This paper summarizes the results of a cross-sectional study aimed at exploring the role of personality disorders in predicting the suicide risk among males imprisoned for various offenses. MATERIAL AND METHODS: The data set used in analyses was based on the responses of 124 males imprisoned for various violent or non-violent offenses (M(age) = 32.39 years, SD = 8.43 years). Male offenders were interviewed using a form for demographic data and information on suicidal history. Also, they completed the Suicide Behaviors Questionnaire-Revised, as part of a broader protocol used in an ongoing study focused on the psychosocial and pharmacological aspects relevant to the diagnosis and management of personality disorders among inmates. RESULTS: At the time of data collection, forty-three respondents (34.7%) were diagnosed with at least one personality disorder. Data were analyzed using hierarchical multiple linear regression. The frequency of suicide attempts during the periods of freedom and during incarceration were significant predictors of suicide risk. Moreover, the diagnosis of personality disorder brought a significant addition to the power of final regression model, explaining almost 8% of the variance in suicide risk among the inmates. CONCLUSIONS: The findings highlight the importance of implementing prevention programs in correctional institutions, with screening and monitoring of inmates diagnosed with severe mental disorders as an important component.

A possibility to identify the vehicle driver through complex forensic and criminalistic expertise--case report.

Traffic accidents can have more or less dramatic consequences that involve penal and civil responsibility with amplitude extending over long periods of time. In many cases, substitution of the driver with the passenger in order to avoid criminal responsibility is often remarked. The substitution takes place with the passenger's agreement or, in cases with dramatic consequences (coma or death), without his/her consent. These situations are encountered in civil cases regarding insurance fraud. In addition to forensic medical expertise, to aid the experts, mathematical modeling and computer simulation of the dynamics of vehicle passengers is a tool that completes the criminal expertise of traffic accidents. This paper presents the method of identification of the person driving the vehicle based on the computer simulation of vehicle occupants' dynamics.