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.

The role of in-person focus groups in the management of urinary incontinence in women from a mixed-methods, randomized pilot study.

PURPOSE: To assess the quantitative and qualitative components of in-person focus groups as a potential intervention for female patients with urinary incontinence. METHODS: Women over the age of 18 seeking treatment for UI were randomized to standard care with focus group participation or to standard care alone. All participants completed validated questionnaires: MESA, UDI-6, OAB-SAT-q, PGI-S, PGI-I, SQoL-F, PHQ-9, IPAQ at the beginning and conclusion of the study. Questionnaires were analyzed with repeated measures of ANOVA models in an intention-to-treat manner. Three moderated focus group sessions were held and audio recorded. Recordings were transcribed and categorized by frequency into themes using grounded theory methodology. RESULTS: A total of ten control and eight intervention participants agreed to participate. Seven women attended all three focus group sessions and were included in the final analysis. Transcripts from focus group sessions observed women identified most with (1) urinary incontinence (UI) as a chronic disease, (2) shame managing UI, and (3) social constraints of toileting. Participants self-reported appreciation of focus group participation and desire for on-going sessions. Analysis of the questionnaires did not demonstrate statistically significant differences. CONCLUSION: Data ascertained from questionnaires were unable to demonstrate a meaningful effect in improved treatment outcomes for control and intervention participants. Grounded theory analysis of transcripts identified four primary themes: (1) appreciation of the focus group, (2) UI as a gendered issue, (3) lack of public awareness, and (4) history of negative provider interactions. All focus group participants self-reported interest in attending future focus group sessions.

Identification of preferred reminder systems and patient factors to promote adherence in the management of urinary incontinence.

Objective: To investigate comfort level and preferences of automated reminder systems (mail, email, text message, phone call, patient-portal message, and/or smartphone application) to promote adherence to recommended therapies for patients seeking care for urinary incontinence (UI) at our urology clinic in Phoenix, Arizona. Methods: Anonymous surveys were distributed in English to adult patients with UI from 4/2019-5/2019. Patient demographics, UI type, and access to and use of the Internet, smartphone and patient-portal were assessed. Using a Likert scale, patients indicated level of comfort with each reminder system and numerically ranked each system. Statistical analyses were performed to identify patient characteristics associated with reminder modality and determine significance in ranking of systems. Results: Fifty-seven patients (67.3 ± 16.3 years) completed the survey with an 87% response rate. Text-message and phone call reminder modalities were ranked the highest compared to other modalities (p < 0.05). A Chi-squared test showed no correlation between preferred reminder system modality and type of incontinence, age, gender, race/ethnicity, or language spoken (p > 0.05). Internet usage and access significantly correlated with preference for smartphone application and patient-portal message reminder systems (p < 0.05). Conclusion: Patients reported they were extremely comfortable with all communication modalities, except for smartphone applications in which patients were the least comfortable. The modalities most preferred by patients were phone call and text message and least preferred were patient portal and smart phone application. In conclusion, phone calls and text messages were the most preferred communication modality, with smart phone applications as the least comfortable. Innovation: This study demonstrates the potential utility of specific reminder modalities for patients seeking help with treatment adherence.