Congenital anomalies (also referred as birth defects, congenital disorders, or congenital malformations) are mostly related to cardiac defects, neural tube defects, and Down syndrome (World Health Organization 2012; Lowry et al. 2013). The congenital heart disease (CHD) problem, as the most common severe congenital anomaly found in neonates, has significant impact not only on infant morbidity and mortality, but also on health care cost in children and adults. Although the etiology is exactly unknown, the incidence of CHD is closely related to other associated major and minor non-cardiac anomalies. This is mostly associated with potential genes and molecular mechanisms, exogenous, and other multifactorial factors that affect embryonic development phase, such as maternal risk and environmental factors (Gembruch 1997; Sander et al. 2006; Baardman et al. 2012). Therefore, CHD cases necessitate to be managed by prenatal screening procedure to improve the survival rate of the fetus. This is realized by performing an in-depth fetal cardiovascular evaluation by using fetal imaging device (Gardiner 2001; Garne et al. 2001).

In the literature, to the authors’ knowledge, fetal heart can only be monitored by using cardiac magnetic resonance (CMR) and fetal echocardiography (Sklansky 2004; Prakash et al. 2010). Fetal echocardiography, as the main obstetric imaging screening for fetus, is considered to be the most effective fetal monitoring system with excellent diagnostic accuracy to diagnose cardiac anomalies prenatally (Meyer-Wittkopf et al. 2001; Chew et al. 2007; Deng and Rodeck 2004), due to its non-radiation exposure, real-time capability, economic rate, and ease of access. The main purpose of fetal echocardiography is to perform assessment of fetal cardiac structure and function, detect complex structural malformation in early stage, and recognize the presence of a defect by using a specialized sonography system. With the combination of expertise knowledge and imaging instrument, most forms of major heart disease have become recognizable by prenatal ultrasound (World Health Organization 2012; Allan et al. 1994; Jaeggi et al. 2001; Lagopoulos et al. 2010). Such screening examination is proven to optimize fetal cardiac monitoring by identifying and characterizing the progression of congenital anomalies before delivery and to aid management decisions in delivery, treatment, or interventions (Garne et al. 2001; Levi 2002; Carvalho 2002; Lee 2013; Carvalho et al. 2013; Ayres 1997; Chitty and Pandya 1997).

In this review, we attempt to provide supplementary information about technical aspects in fetal echocardiography by bringing up the discussion into newer issues in image processing and analysis methods for developing 2-D fetal echocardiography applications. We also restrict our discussion specifically in 2-D fetal echocardiography for diagnostic purpose only. Therefore, topics related to interventional echocardiography is not included in this review. We organize the article by presenting a brief introduction about general aspects in fetal echocardiography. A comparison between CMR and fetal echocardiography is also given here. Section 3.2 reviews about fetal heart imaging modalities, CMR and fetal echocardiography, including the new developments. In Sect. 3.3, a brief overview on basic clinical routine and standards in 2-D fetal echocardiography is portrayed. Furthermore, a thorough discussion on imaging and image processing to solve problems in 2-D fetal echocardiography images is revealed. The discussion includes transducers and settings, speckle reduction, image segmentation, and image analysis. Furthermore, a concise evaluation of these techniques is given to describe the advantages and drawbacks of the techniques. Finally, possible future directions for clinical and research applications are proposed.

The major issue in choosing fetal heart imaging is about noninvasive modality. Although echocardiography remains the core of noninvasive cardiac imaging, the existence of CMR is potential to take part in improving the quality of the diagnosis (Prakash et al. 2010). In this section, we perform a simple comparison study of the two modalities, CMR and echocardiography, to characterize the diagnosis constraints.

The development of the fetal heart begins at conception and completely formed by 8 weeks into pregnancy. Accordingly, CHD occurs during this development phase. Since the anatomy and physiology of the fetal heart is different from those of pediatric or adult, the abnormality is also unalike. In comparison with pediatric heart, the extension of fetoplacental circulation brings about difference in the assessment of the function in the fetal heart. Furthermore, due to extracardiac abnormalities and chromosomal defects, specific lesions may differ in characteristics. The fetal heart has specific abnormality spectrum, cardiac anatomy complexity, specific positioning, small size, and circulation differences which are represented by placental circulation of the blood to the fetal heart and return back to the placenta. Based on these characteristics, the requirement of an ideal fetal echocardiography for CHD imaging should be able to define all anatomical aspects of cardiac structure and evaluate physiological consequences of CHD. Therefore, performing an optimal fetal echocardiographic screening requires special professional prenatal sonographer with cardiac anatomy knowledge and suitable echocardiography device that generates high-resolution real-time images. Thanks to the development of the sophisticated imaging tool, detailed information on the fetal heart structure, function, and time-related events has become available (Gardiner 2001; Prakash et al. 2010; Lee et al. 2008).

The Clinical Standards Committee (CSC) of International Society of Ultrasound in Obstetrics and Gynecology (ISUOG) has a concern to develop practice guidelines and consensus statements that provide health care practitioners with a consensus-based approach for diagnostic imaging (Lee et al. 2008). The guidelines for performing basic and advanced procedures of fetal echocardiography have been largely documented and published in Lee et al. (2008), Barboza et al. (2002), Allan (2004). These procedures highlighted the scanning period based on gestational age, some technical factors for scanning, and cardiac examination in basic and extended methods (Lee et al. 2008). On the subject of our review on image processing perspective, we concentrate our discussion on scanning technical aspects.

Regarding the technical aspects for the examination, the echocardiography equipment for diagnostic purpose strongly depends on the application. Therefore, the choice of echocardiography application differs technically according to clinical constrains, such as maternal and fetal physical status, type of CHD diagnosis that needs to be obtained, gestational age, region of interest to be scanned, and so on that have been clearly revealed in the fetal echocardiography guidelines (Lee 2013) as well as technical constrains (type of transducer, technical settings, supporting equipment, acquisition and recording technique, processing system, data dimension visualization, and other technical requirements). However, we restrict our review only on the configuration for the aforementioned applications.

Rapid advancement in transducer technology and computer system has transmuted the 2-D imaging perspective into higher dimension of imaging system. Although the use of the 2-D echocardiography is considered to be old-fashioned, such scheme still becomes the gold standard for prenatal imaging of the fetal heart and situs (Rychik 2004). The important discussion in 2-D fetal echocardiography can be categorized into two groups: clinical and technical issues. Clinical issues encompass two points: challenges in performing diagnostic screening and the choice of transducers, while technical issues cover three aspects: ultrasound devices, acquisition techniques, and post-processing methods.

In clinical practice, 2-D fetal echocardiography is still powerful in clinical routine and plays a main role in obstetric diagnostics, especially when the availability of sophisticated multidimensional imaging modalities is limited. In diagnostic purpose, the so-called grayscale imaging scheme is the basis of a reliable fetal cardiac examination. In this scheme, 2-D fetal echocardiography is particularly used to observe the ‘basic view’ and ‘extended basic view’ screening for the four-chamber view, which allows assessment of abnormalities involving the atria and the ventricles, as well as right and left ventricular outflow tract and great artery, for more effective screening for CHD. However, many anomalies can still be missed (Gardiner 2001; Jaeggi et al. 2001; Lee 2013; Barboza et al. 2002; Allan 2004; Rychik 2004). Then, again, some clinical drawbacks on the 2-D-based method may occur hampering the scanning reliability. Occasionally, despite the regular 2-D fetal echocardiography procedure yields no abnormality results, particular undetected cardiac anomalies are still found sometime after birth (Sklansky 2004; Buskens et al. 1996). Additionally, separated assessment of the fetal heart and circulation may cause incomprehensive observation. This occurs because the pathophysiology of pregnancy to study uteroplacental circulation is performed by obstetrician using Doppler, while the morphological aspect of cardiac development is investigated by cardiologist by means of M-mode and Doppler (Gardiner 2001; Campbell et al. 1983; Allan et al. 1982, 1987). On the subject of the aforesaid clinical drawbacks, some factors have been thought to contribute to such problems, such as tremendously operator dependent with unreliable sonographic window. This limitation may be part of the cause of the commonly time-consuming acquisition.

On the choice of transducer, the accuracy of a fetal echocardiogram is affected by the scanning time during gestation when the study is performed and the transducer type to obtain the correct fetal heart characteristic. Basically, transvaginal echocardiography and transabdominal echocardiography are the two standard modalities currently being used in prenatal screening for congenital cardiac and extracardiac defects. Both transducers have been fostered primarily by the introduction of higher-frequency and higher-resolution ultrasound probes, so that they suit with early screening. However, transvaginal echocardiography is more applicable for observing cardiac defects at earlier gestational age of pregnancy (Gardiner 2001; Ayres 1997; Chitty and Pandya 1997; Budorick and Millman 2000).

In technical point of view, selecting the right transducer for certain application is indispensable to generate a correct and reliable image. In this regard, we draw our attention to define the appropriate transducer type, settings, and applications. The basic requirement of the 2-D fetal echocardiography transducer is excellent B mode, with a good cine-loop facility for scrolling back frame by frame and capturing the frame of interest as well as real-time scanning capability. Afterward, a special preset of a transducer for evaluating fetal heart is high frame rate, decreased persistence, and increased compression. The high frame rate is achieved by parallel processing, where the transducer transmits one line and receives two, resulting in a doubling of previous rates. The standardized frame rate setting for fetal echocardiography is 4–5 MHz. When the frame rate is decreased, the spatial resolution will be compensated as well (Gardiner 2001; Turan et al. 2009). Furthermore, the transducer frequency is the other feature that plays an important role in ultrasound image quality, as the higher the frequency, the higher the resolution and the greater anatomical detail can be obtained. While for the lower frequency, it is related to increased penetration of the sound beam (Sklansky 2004). In resolution setting, high-spatial, high-temporal, and high-contrast resolutions guarantee the accuracy of the diagnostic of most cardiac anomalies. The spatial resolution determines the anatomical details of the cardiac structures, while the temporal resolution reveals the motion factor (Turan et al. 2009; Chaubal and Chaubal 2009; Ng and Swanevelder 2011). For contrast resolution, it can be achieved by setting the dynamic range into low value. Additionally, the use of contrast agent is found to be effective in generating better contrast (Senior et al. 2009). However, it is considered to be unwise for the use of contrast agent in fetal echocardiography. The existence of artifacts is also a common phenomenon that determines the quality and diagnostic value of the image. Definitely, it is also necessary to provide color Doppler, pulsed Doppler, and continuous-wave Doppler for blood flow visualization and quantification. The availability of advanced fetal echocardiography applications, such as STIC, tissue Doppler, and multiplanar imaging, is certainly more advantageous (Chaubal and Chaubal 2009).