A smartphone application with augmented reality for estimating weight in critically ill pediatric patients

Knowing child weight is critical while managing pediatric emergencies because dosing resuscitation drugs is generally based on weight. However, in many out-of-hospital and emergency department settings the child weight is unknown to the treating team and often it is not possible to weight the patients. Some of the conditions which make challenging to obtain a rapid and reliable measurement of the weight include on-going cardiopulmonary resuscitation (CPR), spinal immobilization, emergency airway management, and emergency delirium or agitation. Calculation of emergency drug doses, choosing the most appropriate equipment size and the energy level for defibrillation ideally requires knowing or to be able to accurately estimate the child’s weight.

For this reason, various weight estimation techniques were developed and published in the literature [1]. Current methods of weight estimation include visual estimation by parents or healthcare providers (physicians, nurses and paramedics), estimation by child age [7-9] and estimation by child length [10,11]. Despite evidences demonstrated that age-based methods have poor estimation accuracy [2, 3] more than twenty age-based formulae were created to estimate children body weight. Some of these formulae require relatively complex arithmetic calculations increasing the risk for errors in drug dosing, in particular, in a stressful resuscitation setting [4]. Among these multiple techniques, parental visual estimation is the most accurate [12].

In case of absence of parents, length-based strategies adjusted for body habitus reached an acceptable level of precision. Moreover, the Paediatric Advanced Life Support (PALS) guidelines published in 2015 suggests that it is reasonable to use a body-length tape with pre-calculated doses if the child’s weight is unknown (Class IIa, Level of Evidence C). Length-based color-coded emergency tapes were developed allowing estimation of body weight from child’s height. These tapes, used throughout the world for paediatric emergencies, are subdivided in nine colour zones based on the length/weight relationship. Each zone estimates the 50th percentile weight for length from standard weight curves and thus represents the median weight (i.e. ideal body weight). The tape is placed next to a supine child with one end at the level of the child head. The opposite end of the tape, reaching the feet of the child, corresponds to the colour code zone and approximates the weight in kilograms along with the pre-calculated doses.

Development of the PediTape app

Concerned by the risk derived by drug dosing errors in critically ill children we developed PediTape, the first smartphone application (app) that estimates child weight using the smartphone camera and augmented reality (AR). PediTape is an app developed only for Apple iOS devices with XCode development environment in Swift programming language.

How it works

PediTape app is very simple and easy to use. The app comes with a user-friendly user interface that permits to quickly obtain an estimation of the child weight using only the smartphone camera. After launching the app, the smartphone camera is immediately activated and displayed in full-screen with a yellow marker in the centre of the screen. As soon as the smartphone camera is available the augmented reality (AR) software starts to track a correspondence between the real-world space and the virtual space: this is the essential requirement for any AR experience and it is called world tracking. To create this correspondence the AR software uses a technique called visual-inertial odometry combining information from the device’s motion sensing hardware with computer vision analysis of the images obtained from the device’s camera [5]. World tracking also analyses and understands the contents of a real-world space using hit-testing methods detecting flat surfaces and main objects in the scene image. All these sophisticated steps are needed in order to display the measuring tape together with a live camera image so that the user experiences augmented reality with the illusion that the virtual tape is part of the real world. The world-tracking process may last from less than one second to numerous seconds based on the lightning and environment conditions. After the completion of this process the app is ready to measure the length of the child by determining the two extremities points, head and foot. To start measuring child height you have to point the marker either to the head or to the foot of the child (starting point) and tap the marker by keeping the smartphone camera steady. After this action a virtual tape anchored to the starting point is displayed in the screen. From now, moving the smartphone around the child will increase the length of the coloured tape like a self-retracting metal tape measure (Fig. 1). The next step is to tap while the marker is over the foot (or head if previously started from foot) defining the ending point. In the bottom of the screen, a button is displayed with the current measured length between the starting point and the ending point along with the corresponding colour code. By tapping this button, a new view is displayed containing medication dosages, equipment sizes and other critical calculations subdivided in four main categories: vital signs, airway management, resuscitation and other drugs (Figure 2A-B). Moreover, the app displays favourite items for quick access. When applicable the app provides name, dose for the corresponding estimated weight, route of administration, general dose and notes for the selected item. It also gives the possibility to the user to add personal notes, save the item as favourite and report errors (Figure 2C).

Figure 2. User interfaces from the PediTape application displaying the main view subdividing medications and equipment into four main categories: vital signs, airway management, resuscitation and other drugs (A), the detail view of the airway management category with an example of related items (B) and the medication or equipment details (name, dose, route of administration, general dose, and notes) for the estimated weight along with the possibility to add personal notes, save the item as favorite and report errors (C).

Potential limitations

When using augmented reality technology, we must always consider that world tracking is an inexact science and subjected to numerous user-dependent limiting factors despite this process achieves often an impressive accuracy. In order to obtain high-quality measures users must be aware of lighting conditions where the app operates and quality of acquired smartphone camera images. In fact, tracking quality is reduced when the camera is pointed at a blank surface or the scene is too dark. AR software is able to better understand the real-world environment if the smartphone is moving without excessive motion that otherwise would results in a blurred low-quality image and hence, would reduce tracking quality. Another key point to obtain an accurate measurement and thus an effective estimation of the weight is to allow the necessary time for surfaces detection.

Data collection

In prevision to carry out a clinical study PediTape comes with a built-in system (currently disabled) that allows collection of anonymous usage data. All measurements made (estimated height and weight) can be collected and stored in an online database with the possibility to provide supplementary details like the true height and weight of the child. The main purpose of this data collection is to establish the reliability and accuracy of this paediatric weight estimation app. An eventual data collection following data protection regulations is secondary to user agreement and ethical approval.

Future perspectives and next steps

The growing incidence of the childhood obesity pandemic [6] has recently led to an increased interest in developing weight estimation technique adjusted for body habitus.

As previously described this app comes with a built-in system that permits to collect measured estimated height and weight of the child. Following to a capillary diffusion between emergency and critical care providers and a proper authorization from user and ethical committee this app may potentially collect a huge amount of data in respect to the current law in terms of data protection. This would improve the length-based estimation of weight by implementing and training a machine learning regression model that features gender and body habitus. Another minor feature we are currently implementing is the possibility for the healthcare provider to add custom drugs or equipment to the app.

Conclusions

Length-based weight estimation methods should be the first line technique to accurately estimate weight of critically ill paediatric patients in respect to age-based methods. Like the currently available physical tapes, this app estimates the child’s weight through the smartphone camera by taking advantage of augmented reality and provides medications doses and equipment sizes to healthcare providers. PediTape app underwent internal accuracy tests but it has not been formally validated. There is the need to perform a clinical study considered that lightning conditions and images quality could influence the accuracy of this technique before using in the clinical setting.

References

  1. Young KD, Korotzer NC. Weight Estimation Methods in Children: A Systematic Review. Ann Emerg Med American College of Emergency Physicians 2016;68:441–451.e10.
  2. Marlow RDLD, Walton LJ. Accurate paediatric weight estimation by age: mission impossible? Arch Dis Child 2011;96.
  3. Wells M, Goldstein LN, Bentley A. It is time to abandon age-based emergency weight estimation in children! A failed validation of 20 different age-based formulas. Resuscitation 2017;116:73–83.
  4. Gavriel Salvendy(Ed.): Handbook of human factors and ergonomics (3rd ed.). Universal Access in the Information Society. 2007;5(4):421-
  5. ARKit | Apple Developer Documentation. [online] Available at: https://developer.apple.com/documentation/arkit [Accessed 25 Feb. 2019].
  6. Ogden CL, Carroll MD, Kit BK, Flegal KM. Prevalence of Childhood and Adult Obesity in the United States, 2011-2012. JAMA. 2014;311(8):806–814. doi:10.1001/jama.2014.732
  7. Luscombe MD, Owens BD, Burke D. Weight estimation in paediatrics: a comparison of the APLS formula and the formula ‘Weight=3(age) + 7’. Emerg Med J 2011;28(7):590–3
  8. Tinning K, Acworth J. Make your Best Guess: an updated method for paediatric weight estimation in emergencies. Emerg Med Australas 2007;19(6):528–34.
  9. Marlow RDLD, Walton LJ. Accurate paediatric weight estimation by age: mission impossible? Arch Dis Child 2011;96 (A1-A2)
  10. Luten R, Wears RL, Broselow J, Croskerry P, Joseph MM, Frush K (August 2002). “Managing the unique size-related issues of pediatric resuscitation: reducing cognitive load with resuscitation aids”. Academic Emergency Medicine. 9 (8): 840–7.
  11. Wells M, Coovadia A, Kramer E, Goldstein L. The PAWPER tape: A new concept tape-based device that increases the accuracy of weight estimation in children through the inclusion of a modifier based on body habitus. Resuscitation
  12. Krieser D, Nguyen K, Kerr D, Jolley D, Clooney M, Kelly AM. Parental weight estimation of their child’s weight is more accurate than other weight estimation methods for determining children’s weight in an emergency department?. Emerg Med J. 2007;24(11):756-9.