Article Preview
TopIntroduction
Miscarriage is the spontaneous loss of a pregnancy before the 20th week and presents the most common hurtful pregnancy outcome. As miscarriage is an irreversible phenomenon, prediction and prevention are still the only way to deal with this problem, which has negative psychological consequences for the woman and her partner as well (Kong et al., 2010). Some studies report that 50 percent of pregnancies will end in miscarriage (Arck et al., 2008).
Well-established risk factors for miscarriage include history of miscarriage, increased maternal age and infertility (Khalil et al., 2013; Waldron, n.d.). Also, several behavioral and social risks contribute in the increasing of the risk of miscarriage. Smoking (Pineles, Park, and Samet, 2014), alcohol drinking and caffeine intake (Avalos et al. 2014; Chen et al., 2016) are the main examples, but they still unconfirmed because some studies show these associations in the context of nausea which reduce the risk of miscarriage (Wen et al., 2001).
Stress and emotional wellbeing remain also important in pregnancy. In fact, the combination between life events and stress degree of the person increases risk of getting miscarriage (Khan et al., 2017). Emotions like anxiety, shock, sadness, anger, blame, depression, sleep disturbance have all been described as emotional responses to pregnancy loss (Mills et al., 2014). On other aspect, women are asked to participate in regular and moderate exercises while pregnant. Physical activity is defined by using either individual characteristics of a woman’s occupation such as standing, stooping, lifting, or walking. However, extreme or strong activity of the body has been associated with an elevated risk of miscarriage (Wong et al., 2010).
There is also increasing interest in the role that eating well plays in pregnancy (Women, 2005). Food safety is important for everyone and especially for pregnant women, because of her susceptibility to getting an illness during pregnancy. Eating in a restaurant, snack or cafe presents a higher risk of developing certain foodborne illnesses to pregnant women and their unborn children, and that can conduct to a miscarriage. Obesity is also associated with an increased risk of pregnancy complications. Indeed, obesity represents an independent cause of miscarriage (Pasquali, 2006), with rates that vary between 17 and 27 percent.
It becomes clear that several risk factors contribute to miscarriages. The research problem this paper tries to address is how reality mining, Big Data analytics and machine learning algorithms can help pregnant women to avoid getting miscarriage.
The idea is to take advantage of the ease with which mobiles and sensors can capture daily data and apply data mining techniques and big data predictive analytics to this data to make pregnant women’s life safer and healthier. Thanks to sensors and mobile phones, we can collect data from citizens, systems, and general things that weren’t previously possible (Asri, Mousannif, and Al Moatassime, 2018; Wang et al., n.d.).
The rest of this paper is illustrated over a few sections starting with a brief literature review followed by an overview of miscarriage prediction and detailed prediction process in section 3. Section 4 presents the data conceptual modeling; while section 5 presents Kmeans algorithm as data mining tool for clustering data. Section 6 illustrates the environment of the experiment, metrics, and experimental results. Finally, we conclude the paper by a conclusion and future work.