Development of a novel mobile application to assist in the diagnosis of Fetal Alcohol Syndrome (FAS)

点击翻译按钮获取中文摘要

Project Summary/Abstract. Fetal alcohol spectrum disorder (FASD) is a major public health issue resulting from prenatal alcohol exposure (PAE), impacting between 1-5% of the US population and perhaps more than 630,000 children globally each year. When assessing for FASD the following facial features must to be evaluated as part of the diagnostic process: palpebral fissure length (PFL), thinness of the upper vermillion, and the smoothness of the philtrum. PFL is typically assessed with a metric ruler or photometrically, while the lip and philtrum are evaluated against a series of five Likert scale photos. Each of these assessments has methodological problems often resulting in unreliable application of specific diagnostic criteria. There is also a reluctance on the part of health care providers to diagnose FASD because they don’t feel confident in their skills, resulting in under- or misdiagnosis. Furthermore, the number of affected individuals is far greater than the capacity of diagnosticians, especially when considering the global impact of PAE. Some of the issues involved in the diagnosis of FASD could be remedied by simple-to-use mobile apps that would take much of the guesswork or heuristic elements out of the diagnostic equation, provide evaluation based upon expert opinion, and can be used by health care providers globally. Several aims of this project revolve around replacing the standard Likert scale lip-philtrum guide with an animation, morphing from a full to a thin upper lip on a continuous slider scale that the user controls while examining a subject. A similar morphing tool is proposed for the assessment of the philtrum. This would eliminate much of the confusion that exists between the rankings required for a fetal alcohol syndrome (FAS) or partial FAS (pFAS) diagnosis. The 5-point Likert scale doesn’t provide enough guidance and agreement even among experts is often mediocre. Rather than having a scale with only five demarcations, where really only the demarcation between a Rank 3 and 4 matters for a diagnosis, the apps allow for a more accurate assessment by providing a continuous distribution removing all ambiguity. Dysmorphologists who are expert at diagnosing FAS will determine the demarcation points for an upper lip and a smooth philtrum compatible with a diagnosis of FAS or pFAS for use in the app. Bias in the current diagnostic methods will also be assessed. PFL is an extremely difficult measurement even for experts, and it is proposed that using a smartphone camera with 3D capability would provide a more accurate way to assess this cardinal feature. These apps could be used by health care providers anywhere in the world to receive expert opinion feedback almost instantly to assist them in making a difficult yet critically important clinical diagnosis. We are in a unique position to accomplish these endeavors, as we have both the clinical and technical expertise to develop the apps, and access to data to validate the findings. Given the scope of the problem and the lack of access to expert diagnosticians, these mobile health applications should overcome at least some of the issues involved in the diagnosis of FASD.