The utility of behavioral biometrics in user authentication and demographic characteristic detection: a scoping review

BACKGROUND

Objective measures of screen time are necessary to better understand the complex relationship between screen time and health outcomes. However, current objective measures of screen time (e.g., passive sensing applications) are limited in identifying the user of the mobile device, a critical limitation in children's screen time research where devices are often shared across a family. Behavioral biometrics, a technology that uses embedded sensors on modern mobile devices to continuously authenticate users, could be used to address this limitation.

OBJECTIVE

The purpose of this scoping review was to summarize the current state of behavioral biometric authentication and synthesize these findings within the scope of applying behavioral biometric technology to screen time measurement.

METHODS

We systematically searched five databases (Web of Science Core Collection, Inspec in Engineering Village, Applied Science & Technology Source, IEEE Xplore, PubMed), with the last search in September of 2022. Eligible studies were on the authentication of the user or the detection of demographic characteristics (age, gender) using built-in sensors on mobile devices (e.g., smartphone, tablet). Studies were required to use the following methods for authentication: motion behavior, touch, keystroke dynamics, and/or behavior profiling. We extracted study characteristics (sample size, age, gender), data collection methods, data stream, model evaluation metrics, and performance of models, and additionally performed a study quality assessment. Summary characteristics were tabulated and compiled in Excel. We synthesized the extracted information using a narrative approach.

RESULTS

Of the 14,179 articles screened, 122 were included in this scoping review. Of the 122 included studies, the most highly used biometric methods were touch gestures (n = 76) and movement (n = 63), with 30 studies using keystroke dynamics and 6 studies using behavior profiling. Of the studies that reported age (47), most were performed exclusively in adult populations (n = 34). The overall study quality was low, with an average score of 5.5/14.

CONCLUSION

The field of behavioral biometrics is limited by the low overall quality of studies. Behavioral biometric technology has the potential to be used in a public health context to address the limitations of current measures of screen time; however, more rigorous research must be performed in child populations first.

SYSTEMATIC REVIEW REGISTRATION

The protocol has been pre-registered in the Open Science Framework database ( https://doi.org/10.17605/OSF.IO/92YCT ).

Teaching families to manage intake of candy in the home: Results from a feasibility study using multiphase optimization strategy (MOST)

BACKGROUND

Excess intake of desserts/sweets high in added sugars, such as candy, is linked with greater obesity risk. Parents often limit their childrens' intake of these sweet foods using controlling feeding practices, such as restriction; yet, restrictive feeding practices are counterproductive for childrens' self-regulation of energy intake.

OBJECTIVE

This study developed a family-based behavioural intervention that taught parents alternatives to restrictive feeding practices and encouraged children to consume candy in moderation.

METHODS

Using the multiphase optimization strategy (MOST), parent-child dyads (N = 37) were randomized into one of eight conditions that included a combination of intervention components delivered over 4 weeks: home supply, parent shared decision making, child mindfulness and child attention control strategies.

RESULTS

Retention rate at follow-up was high (95%). Among parents who received parent shared decision making, 86.4% reported the structured-based candy routine they set with their child was easy to follow. Most children reported child mindfulness (95%) and attention control (89.5%) strategies were easy to play. Children recalled 4.1 ± 1.8 of the six mindfulness strategies and 2.7 ± 1.6 of the five attention control strategies at follow-up. Eating in the absence of hunger tended to be lower for children who received parent shared decision making and child mindfulness components.

CONCLUSION

This intervention was feasible and well-implemented in the home environment. Findings will inform future, larger interventions designed to test similar strategies on childrens; eating behaviours and self-regulation.

Plasma cortisol and white blood cell responses in different breeds of bulls: a comparison of two methods of castration

To determine plasma cortisol and white blood cell response to castration, Angus (n = 12, 21.4 mo of age), Hereford (n = 6, 21.2 mo of age), and Brahman (n = 24, 20.3 mo of age) bulls nearing maturity were either left intact as uncastrated controls (CON), surgically castrated (SUR) after lidocaine, or castrated by latex rubber banding (BAN). Before and through 35 d after castration (castration = d 0), animals were weighed and blood samples were collected for analysis of cortisol and total white blood cell (WBC) count at 2-, 3-, or 7-d intervals. There was a treatment x breed interaction for ADG from d 0 to 7 (P < .05). From d 0 to 14, 0 to 21, 0 to 28, and 0 to 35, ADG tended to be lower for SUR and BAN animals than for CON animals (castrated vs CON, P < or = .13). No significant differences in ADG were observed between SUR and BAN animals during these times. On d 0, from just before treatment to just after treatment, plasma cortisol concentration increased 3.2 ng/mL for SUR and .1 ng/mL for BAN (SEM = +/- .5 ng/mL; SUR vs BAN, P < .03). From d 0 pretreatment to d 2 after treatment, plasma cortisol concentration increased 1.5 ng/mL for castrated (SUR = 2.0 and BAN = 1.1 ng/mL) and decreased 1.6 ng/mL for CON (SEM = +/- .7 ng/mL; P < .04). Plasma cortisol concentration was negatively correlated (P < .001) with BW (r = -.17) and BW change (r = -.19).(ABSTRACT TRUNCATED AT 250 WORDS)

Bull influences on conception percentage and calving date in Angus Hereford, Brahman and Senepol single-sire herds

Bull breeding soundness parameters, semen characteristics and sexual behavior were evaluated for effects on reproductive performance in single-sire beef herds. A total of 155 cow herds (Angus, 50 herds; Hereford, 40 herds; Brahman, 46 herds; and Senepol, 19 herds) bred to bulls of the same breed were observed for 8 yr. All bulls produced adequate quality semen and had scrotal circumference (SC)>or=30 cm. Reproductive performance was evaluated by the conception rate (CON), conception rate during the first 21 d of the breeding season (21dCON), mean calving date (MCD), and mean calving date of the first half of the herd to calve (HHCD). Correlations were determined between breeding soundness parameters and reproductive performance for all bulls combined, by breed, and by age. The Cp statistic was used to select models for the effects of parameters on CON, 21dCON, MCD and HHCD. Breeding season length and breed had significant effects. The percentages of normal cells, proximal droplets, detached heads and the semen score (motility plus percentage of normal cells) had a significant effect on CON when all bulls were considered. After the effect of season was deleted, the most significant parameter affecting CON in the Brahman was the percentage of detached sperm heads. In the Angus, motility was significantly correlated with all reproductive performance indices. In the Hereford, breeding soundness examination score (BSE) was positively correlated with 21dCON.

Direct and maternal genetic effects due to the introduction of Bos taurus alleles into Brahman cattle in Florida: I. Reproduction and calf survival

Pregnancy rate, calf survival rate to weaning and calf age at weaning of several types of crossbred cows (2/3 or more Brahman) were compared to those of straightbred Brahman and Angus cows over a 12-yr period at Subtropical Agricultural Research Station near Brooksville, FL. The purpose of this study was to determine the relative importance of additive vs nonadditive genetic effects on reproductive and calf survival traits in a population of cattle whose foundation was selected on the basis of superior reproductive performance under harsh environmental conditions. Best linear unbiased estimates (BLUE) of direct additive effect (measured as the deviation of Brahman additive breed effect from Angus) for pregnancy rate and calf age, measured as traits of the dam, were 6 +/- 3% and -7.2 +/- 2.1 d, respectively. Thus, Bos taurus germ plasm did not increase pregnancy rate but resulted in an earlier calving date. The BLUE of nonadditive (intralocus) direct genetic effects measured as deviations from intralocus group genetic effects in the parental breeds on pregnancy rate and calf age at weaning were 25 +/- 4% and -6.4 +/- 2.5 d. Nonadditive effects on pregnancy rate were the primary cause of the superior reproductive rates observed in Brahman crossbred cows. Calf survival was considered to be a trait of the calf, and BLUE of direct additive, direct nonadditive, maternal additive and maternal nonadditive genetic effects was obtained. Only maternal nonadditive genetic effects were found to have a significant effect on survival rate (9 +/- 4%).

Direct and maternal genetic effects due to the introduction of Bos taurus alleles into Brahman cattle in Florida: II. Preweaning growth traits

Records of birth weight (BW), weaning weight (WW) and condition score (CS) from 1,467 Brahman and Brahman X Angus crossbred calves from Brahman and crossbred Brahman sires and Brahman, crossbred Brahman and Angus dams were collected at the Subtropical Agricultural Research Station, Brooksville, Florida, from 1971 to 1982. Best linear unbiased estimates (BLUE) of Brahman sire and dam group additive genetic effects (as deviations from Angus) and Brahman X Angus dam and calf group nonadditive (intralocus) genetic effects (as deviations from intralocus group genetic effects in the parental breeds) were obtained. Linear combinations of these were used to compute direct and maternal Brahman additive and Brahman X Angus nonadditive (intralocus) group genetic effects. The respective BLUE of these four effects were 5.99 +/- 2.08, -5.70 +/- 1.91, .52 +/- 1.81 and 2.85 +/- .72 kg for BW; 9.60 +/- 10.29, 8.76 +/- 9.47, 9.47 +/- 8.96 and 20.95 +/- 3.56 kg for WW; and -1.10 +/- .55, 1.64 +/- .50, 1.47 +/- .47 and .05 +/- .19 units for CS. Linear combinations of the BLUE of sire, dam and calf group genetic effects can be used to predict the genetic worth of crossbred groups composed of any combination of Brahman and Angus breeding. Nonadditive maternal group genetic effects were the most important factor for BW and WW, whereas nonadditive direct group genetic effects were the most important for CS.

Treatment of basal-cell carcinomas with a carbon-dioxide laser

Treatment of basal-cell carcinomas with a carbon-dioxide laser was studied in 24 lesions for adequacy of treatment, the healing process, and the eventual cosmetic result after twelve months. There was a recurrence rate of 50% at the end of one year; healing occurred with mild degrees of hypopigmentation, erythema, and depression; and the cosmetic result was rated good or better than good in 10 of 11 lesions available for judgment. The high recurrence rate was due, we think, to our cautious use of the laser, which, we think nevertheless, warrants further trial with more aggressive application.