Augmented reality app boasts capacity in estimating children’s weight

An augmented reality mobile application (app), called the Paediatric Augmented Reality Scale (PARS), performs slightly better than the Broselow tape (BT) and Paediatric Advanced Weight Prediction in the Emergency Room extra-long and extra-large (PAWPER-XL) tape methods in estimating body weight of children, a Korea study has shown.

“PARS-estimated body weights correlated fairly accurately with the actual body weights,” the researchers said. “PARS holds potential utility in paediatric emergencies.”

Children aged 1 month to 12 years who visited the emergency department of the tertiary university hospital in Bucheon, South Korea, between July 2021 and February 2022 were included in this prospective, single-blinded cross-sectional study. The research team assessed the measurement agreement and performance of three methods, namely BT, PAWPER-XL, and PARS.

A total of 1,090 participants (mean age 4.1 years, mean height 102.7 cm, mean weight 18.8 kg) were enrolled, of whom 639 (58.6 percent) were boys. [Ann Acad Med Singap 2023;52:660-668]

PARS demonstrated a lower mean absolute percentage error (9.60 percent) and root mean square percentage error (3.02 percent) when compared with BT and PAWPER-XL. The augmented reality app also outperformed the two methods by achieving a higher proportion of weights estimated within 10 percent of the actual weight (63.21 percent vs 57.25 percent and 62.47 percent, respectively).

Moreover, the intraclass correlation coefficients for the actual and estimated weights were 0.973 with PARS, 0.952 with BT, and 0.969 with PAWPER-XL (p<0.001).

“We developed the PARS, a mobile application that combines the strengths of the body habitus-adjusted, length-based weight estimation method with the accessibility and convenience of smartphone augmented reality,” the researchers said. “The performance of PARS was modestly better than that of the BT and PAWPER-XL tape.”

Strengths

PARS utilizes a length-based weight estimation that is akin to BT and PAWPER-XL. Its strengths lie in its accessibility, portability, and accuracy.

“Smartphone medical applications are widely used by healthcare providers worldwide,” the researchers said. “PARS will assist healthcare providers in emergency situations inside and outside of hospitals.” [P T 2014;39:356-364]

In addition, PARS does not need additional reference objects like other smartphone apps. Finally, it considers both underweight and overweight statuses. [Prehosp Emerg Care 2020;24:441-450; J Clin Monit Comput 2018;32:571-578]

“It is worth noting that previous smartphone applications have also demonstrated high PW10 [proportion of weights estimated within 10 percent of the actual weight] values,” the researchers said. [Pediatr Emerg Care 2022;38:e1257-e1261]

“For accurate weight estimation, it is crucial for PARS to fulfil the PW10 (60–70 percent) and PW20 (90–95 percent) criteria,” they noted. [Trauma Emerg Care 2017;2:1-8; J Pediatr 2015;167:173-7.e1]

PARS had a PW10 value of 63.21 percent and a PW20 value of 88.72 percent for all patients. These estimates were comparable to the reference values and even outperformed those of PAWPER-XL, according to the researchers.

“Augmented reality combines digital information with the real environment and enables accurate measurement of object dimensions,” they said. “[It] is widely utilized in various medical fields, including surgical planning and telemedicine, due to its high level of convenience and accessibility.” [J Digit Imaging 2019;32:38-53; JMIR Mhealth Uhealth 2019;7:e10967; J Med Internet Res 2019;21:e12368]