JMIR Publications

ABSTRACT

Background:

Development of medical devices is an area faced with multiple challenges, resulting in a high number of developments not reaching the patients. Neonatal jaundice (NNJ) is an important cause of newborn morbidity and mortality and results in more than 100.000 deaths annually. It is important to detect the condition at an early stage, but currently there is a lack of accurate, available, and affordable tools.

Objective:

We wanted to develop a novel tool to assess NNJ that gave accurate results, was approved as a medical device, was easy to use, and that would be able to produce at an affordable price, even for low resource settings.

Methods:

We used an iterative approach to develop a smartphone-based system to detect NNJ. We worked in parallel on both technical development, and clinical and usability testing, and initiated the regulatory processes for certification from the start of the project. We updated the system for each iteration of the system and the final version underwent a clinical validation study on healthy, term born newborns aging 1 to 15 days, before all documentation was submitted for conformity assessment to obtain CE-certification.

Results:

We developed a system that included a smartphone app, a color calibration card, and a server. Three iterations of the system were developed, and the final version was approved as a medical device after Medical Device Regulations (MDR). 201 infants were included in the validation study, and bilirubin estimates were highly correlated to blood levels (r=.84). The system had a high sensitivity to detect severe jaundice (97%) and maintained a high specificity (64%).

Conclusions:

A smartphone-based system shows large potential for use in screening of NNJ. An iterative approach with work along different axes in parallel, and with focus on regulatory approval from the beginning can be used to develop market-ready mHealth solutions. Clinical Trial: clinicaltrials.gov, identifiers NCT03007563 and NCT04182555