High-resolution images were obtained for all specimens. A two-dimensional slice showing the ventricular septum of a fetal heart is shown in Fig. 1. The figure clearly shows the myocardial structure and displays an inverted Y-shaped dark structure at the crest of the ventricular septum indicating the conduction bundle. The conduction bundle contains more fat tissues and appears darker in XPCT than the neighboring myocardium [7]. A video containing this slice is provided as Supplementary video 1. Figure 2 shows the left anterior descending artery and its first septal branch in the same heart. A video containing this slice is provided as Supplementary video 2. A three-dimensional volume-rendered reconstruction of the same specimen with special attention paid to the ventricular septum is provided as Supplementary video 3.
Comments
We have shown that XPCT provides high-resolution, three-dimensional images without destroying or altering the property of human heart specimens.
Alternative Imaging Modalities
Other modalities for visualizing the structures of soft tissues include histological observation, absorption contrast computed tomography (ACCT), and magnetic resonance imaging (MRI). However, each of these modalities has its shortcomings.
Histological light microscopic observation is the standard technique for structural investigation, but a major limitation of this technique is the destruction of the specimen by sectioning, which renders the specimen unusable for other studies. ACCT has a high spatial resolution of up to 0.7 μm [8]; however, its contrast sensitivity in imaging soft tissues is about 1000 times poorer compared with XPCT [9]. Remedies for low contrast sensitivity include staining and injection of contrast agent into the blood vessels. However, the use of stains such as osmium, inorganic iodine, and phosphotungstic acid may alter the chemical and mechanical properties of the samples, and hence precluding subsequent studies. Furthermore, the homogeneous injection of contrast agent into the heart postautopsy is technically difficult because the heart chambers have already been incised for macroscopic investigation. Consequently, intact imaging of human heart specimens by ACCT is impractical. MRI may be used for intact soft tissue imaging; however, its spatial resolution of about 40 μm yields less detailed images compared with XPCT.
Alternative XPCT Techniques
Several XPCT imaging techniques have been developed, including the propagation-based technique, analyzer-based technique, crystal interferometry, grating interferometry, and the grating non-interferometric technique. We adopted grating interferometry because it provides adequate field of view, as well as high-density sensitivity and a dynamic range, and does not suffer from interferometer drift. A comprehensive review of each technique can be found elsewhere [10].
In our current XPCT imaging system, investigation of the whole adult human heart is not possible due to two limitations. The first limitation is unsatisfactory spatial resolution. The size of the specimen is limited by the XPCT system’s effective field of view, which is defined by the X-ray detector, grating area size, and X-ray beam size. Since the X-ray detector is composed of a digital camera with pixel array, the field of view is defined by the pixel size and number of pixels. This means that a larger pixel size makes it possible to obtain a larger field of view. However, a larger pixel size decreases the quality of spatial resolution. The height of the specimen is limited to 30 mm due to the X-ray beam size of our current XPCT imaging system. It may be possible to measure a larger specimen up to 100 mm in width using an X-ray detector with a larger field of view and a larger grating area size at the expense of spatial resolution. The second limitation is increased X-ray absorption. A larger specimen must be placed in a larger saline-filled canister for XPCT imaging. If we use a canister of 100 mm in thickness, the X-rays irradiated into the canister are mostly absorbed by saline and the specimen. It would be necessary to use higher X-ray energy for X-ray transmission when imaging a specimen as large as 100 mm in diameter.
Future Clinical Contribution
In spite of extensive research, the structure of hearts with congenital heart diseases has not been elucidated. For example, a heart with isomerism may have anterior atrioventricular bundle, posterior atrioventricular bundle, or twin atrioventricular sling. The location of the conduction bundle cannot be definitively predicted in an isomeric heart [11]. Furthermore, definitive imaging criteria to distinguish myocardial non-compaction and dilated cardiomyopathy have yet to be established [12]. Surgical injury to the coronary arterial branches supplying the conduction bundle is postulated to induce heart block. The clinical impact of coronary arterial branch injury from myocardial resection or ventricular septal defect enlargement remains unclear [13]. Using XPCT, we will examine as many heart specimens with congenital heart diseases as possible. The structures of the conduction tissues, cardiac musculature, and coronary vessels, as well as their relationships, can be easily studied. We believe that XPCT is a powerful tool for elucidating unanswered questions concerning congenital heart diseases.
Clinical application of XPCT is gaining traction [14]; however, imaging of the in vivo heart is challenging as it is surrounded by the air-filled alveoli of the lungs, which cause random refraction of irradiated X-rays. Moreover, the ribs absorb X-rays strongly. As a result, detection of phase-shift values inside the heart is more difficult compared with using an isolated heart. Future development of the XPCT system for in vivo human heart imaging can be expected.