Automatic segmentation of coronary lumen and external elastic membrane in intravascular ultrasound images using 8-layer U-Net.

BACKGROUND: Intravascular ultrasound (IVUS) is the golden standard in accessing the coronary lesions, stenosis, and atherosclerosis plaques. In this paper, a fully automatic approach by an 8-layer U-Net is developed to segment the coronary artery lumen and the area bounded by external elastic membrane (EEM), i.e., cross-sectional area (EEM-CSA). The database comprises single-vendor and single-frequency IVUS data. Particularly, the proposed data augmentation of MeshGrid combined with flip and rotation operations is implemented, improving the model performance without pre- or post-processing of the raw IVUS images. RESULTS: The mean intersection of union (MIoU) of 0.937 and 0.804 for the lumen and EEM-CSA, respectively, were achieved, which exceeded the manual labeling accuracy of the clinician. CONCLUSION: The accuracy shown by the proposed method is sufficient for subsequent reconstruction of 3D-IVUS images, which is essential for doctors' diagnosis in the tissue characterization of coronary artery walls and plaque compositions, qualitatively and quantitatively.

An 8-layer residual U-Net with deep supervision for segmentation of the left ventricle in cardiac CT angiography.

BACKGROUND AND OBJECTIVES: Accurate segmentation of left ventricle (LV) is a fundamental step in evaluation of cardiac function. Cardiac CT angiography (CCTA) has become an important clinical diagnostic method for cardio-vascular disease (CVD) due to its non-invasive, short exam time, and low cost. To obtain the segmentation of the LV in CCTA scans, we present a deep learning method based on an 8-layer residual U-Net with deep supervision. METHODS: Based on the original 4-layer U-Net, our method deepened the network to eight layers, which increased the fitting capacity of the network, thus greatly improved its LV recognition capability. Residual blocks were incorporated to optimize the network from the increased depth. Auxiliary paths as deep supervision were introduced to supervise the intermediate information to improve the segmentation quality. In this study, we collected CCTA scans of 100 patients. Eighty patients with 1600 discrete slices were used to train the LV segmentation and the remaining 20 patients with 400 discrete slices were used for testing our method. An interactive graph cut algorithm was utilized reliably to annotate the LV reference standard that was further confirmed by cardiologists. Online data augmentation was performed in the training process to improve the generalization and robustness of our method. RESULTS: Compared with the segmentation results from the original U-Net and FC-DenseNet56 with Dice similarity coefficient (DSC) of 0.878±0.230 and 0.897±0.189, respectively, our method demonstrated higher segmentation accuracy and robustness for varying LV shape, size, and contrast, achieving DSC of 0.927±0.139. Without online data augmentation, our method resulted in inferior performance with DSC of 0.911±0.170. In addition, compared with the provided results from other existing studies in the LV segmentation of cardiac CT images, our method achieved a competitive performance for the LV segmentation. CONCLUSIONS: The proposed 8-layer residual U-Net with deep supervision accurately and efficiently segments the LV in CCTA scans. This method has potential advantages to be a reliable segmentation method and useful for the evaluation of cardiac function in the future study.

Ultrasonography Monitoring of Trauma-Induced Heterotopic Ossification: Guidance for Rehabilitation Procedures.

Traumatic injury is one of varying causes of heterotopic ossification (HO). After HO occurrence, rehabilitation training need alterations to avoid the aggravation of HO. Therefore, monitoring of HO development plays an important role in the rehabilitation procedure. The aims of this study are to evaluate the post-traumatic HO occurring at various joints, to describe the features of HO development in ultrasound images, and to provide a guidance for the orthopedist to make individualized rehabilitation therapy. Eight subjects with the post-traumatic HO were recruited in this study. The joints on the injured side was examined by plain radiography. The joints on the injured side and the corresponding sites on the uninjured sides were scanned by ultrsonography. The HO tissues were segmented automatically using a semi-supervised segmentation algorithm. Then the HO tissues were evaluated in comparison with the corresponding region of the uninjured side. During the development stage of immature HO, ultrasonography was sensitive to observe the involved soft tissue and the calcification of HO. The characteristics of HO tissues in ultrasound image included the hyperechoic mass occasionally accompanied with acoustic shadow and the irregular muscular architecture. It was found that the mean grayscale value of HO was significantly higher (p < 0.001) than that of the uninjured side at the middle and late stages. During the development period of HO, the HO grayscale value gradually increased and the mean grayscale of value of mature HO was significantly higher (p < 0.05) than that of immature HO. According to the information of HO provided by ultrasound, the orthopedist properly adjusted the rehabilitation treatment. The results demonstrated that the visualization of HO using ultrasonography revealed the development of HO in the muscle tissues around the injured joints and thus provide a guidance for the orthopedist to make individualized rehabilitation therapy. Ultrasound could be a useful imaging modality for quantitative evaluation of HO during the rehabilitation of traumatic injury.

Trichostatin A (TSA) improves the development of rabbit-rabbit intraspecies cloned embryos, but not rabbit-human interspecies cloned embryos.

The interspecies somatic cell nuclear transfer (iSCNT) technique for therapeutic cloning gives great promise for treatment of many human diseases. However, the incomplete nuclear reprogramming and the low blastocyst rate of iSCNT are still big problems. Herein, we observed the effect of TSA on the development of rabbit-rabbit intraspecies and rabbit-human interspecies cloned embryos. After treatment with TSA for 6 hr during activation, we found that the blastocyst rate of rabbit-rabbit cloned embryos was more than two times higher than that of untreated embryos; however, the blastocyst rate of TSA-treated rabbit-human interspecies cloned embryos decreased. We also found evident time-dependent histone deacetylation-reacetylation changes in rabbit-rabbit cloned embryos, but not in rabbit-human cloned embryos from fusion to 6 hr after activation. Our results suggest that TSA-treatment does not improve blastocyst development of rabbit-human iSCNT embryos and that abnormal histone deacetylation-reacetylation changes in iSCNT embryos may account for their poor blastocyst development.

Time course of meiotic progression after transferring primary spermatocyte into ooplasm at different stages.

This study attempted to investigate the time course of meiotic progression after transferring primary spermatocyte (PS) into ooplasm at different maturing stages. In present experiments, PSs were introduced into maturing ooplasts or oocytes by electrofusion. Higher fusion rate was obtained by phytohemagglutinin (PHA) agglutination than by perivitelline space (PVS) insertion. When the ooplasms prepared at 0, 2, 5, and 8.5 hr of in vitro maturation (IVM) were used as recipients and PSs were used as donors, the reconstructed cells extruded the first polar body (PB1) approximately 8.5, 7, 5.5, and 3 hr after electrofusion, respectively. Especially, when ooplasm cultured for 8.5 hr in vitro after GV removal was fused with PS, the PB1 was emitted 7-11 hr after electrofusion. Additionally, the PB1 extrusions of GV and pro-MI oocytes fertilized with PSs were 2.5 hr earlier than control oocytes. The results suggest that (1) PSs undergo the first meiosis in different time courses when introduced into ooplasm at different maturing stages; (2) GV material plays an important role in determining the timing of PB1 extrusion; and (3) first meiotic division of GV and pro-MI oocytes can be accelerated by introducing PS.

Interspecies nuclear transfer of Tibetan antelope using caprine oocyte as recipient.

Interspecies nuclear transfer is an invalulable tool for studying nucleus-cytoplasm interactions; and at the same time, it provides a possible alternative to clone endangered animals whose oocytes are difficult to obtain. In the present study, we investigated the possibility of cloning Tibetan antelope embryos using abattoir-derived caprine oocytes as recipients. Effects of culture conditions, enucleation timing, and donor cell passages on the in vitro development of Tibetan antelope-goat cloned embryos were studied. Maternal to zygotic transition timing of interspecies Tibetan antelope embryos was also investigated using two types of cloned embryos, Tibetan antelope-rabbit and Tibetan antelope-goat embryos. Our results indicate that: (1) goat oocyte is able to reprogram somatic cells of different genus and supports development to blastocyst in vitro. (2) Coculture system supported the development of Tibetan antelope-goat embryos to blastocyst rate stage (4.0%), while CR1aa alone did not. (3) When MII phase enucleated caprine cytoplast and TII phase enucleated caprine cytoplast were used as recipients, the fusion rate and blastocyst rate of hybrid embryos were not statistically different (73.9% vs. 67.4%; 4.0% vs. 1.1%). (4) When donor cells at 3-8 passages were used, 2.9% hybrid embryos developed to blastocysts, while none developed to blastocysts when cells at 10-17 passages were used. (5) There may be a morula-to-blastocyst block for Tibetan antelope-goat, while there may be an 8- to 16-cell block for Tibetan antelope-rabbit embryos.

Increased Th1/Th2 (IFN-gamma/IL-4) Cytokine mRNA ratio of rat embryos in the pregnant mouse uterus.

Somatic cell nuclei can be dedifferentiated in ooplasm from another species, and interspecies cloned embryos can be implanted into the uteri of surrogates. However, no full pregnancies have been achieved through interspecific mammalian cloning. Rat blastocysts transferred into mouse uteri provide a unique model for studying the causes of interspecific pregnancy failure. In this study, intraspecific pregnancy (mouse-mouse) and interspecific pregnancy (rat-mouse) models were established. On Day 9 of pregnancy, the fetoplacental units were separated from the uterine implantation sites and the expression of messenger (m)RNA was quantitated by real-time PCR. We compared the mRNA expression levels of type-1 T helper (Th1) and type-2 T helper (Th2) cytokines, interferon-gamma (IFN-gamma), and interleukin-4 (IL-4) in fetoplacental units between intraspecific and interspecific pregnancy groups. The mRNA expression of IFN-gamma in the fetoplacental units of the interspecific pregnancy group was significantly higher than that of the intraspecific pregnancy group (P<0.05). The mRNA expression of IL-4 in the interspecific pregnancy group was significantly lower than that in the intraspecific pregnancy group (P<0.05). We also analyzed the ratio of IFN-gamma/IL-4 mRNA, and an increased IFN-gamma/IL-4 mRNA ratio was observed in the interspecific pregnancy compared with that in the intraspecific pregnancy group. The IFN-gamma and IL-4 mRNA expressions indicate that there is a Th1/Th2 imbalance in the feto-maternal interface of interspecific pregnancies. Bias of Th1 cytokine dominance may be a barrier to reproductive success between species.

NuMA distribution and microtubule configuration in rabbit oocytes and cloned embryos.

The assembly of microtubules and the distribution of NuMA were analyzed in rabbit oocytes and early cloned embryos. Alpha-tubulin was localized around the periphery of the germinal vesicle (GV). After germinal vesicle breakdown (GVBD), multi-arrayed microtubules were found tightly associated with the condensed chromosomes and assembled into spindles. After the enucleated oocyte was fused with a fibroblast, microtubules were observed around the introduced nucleus in most reconstructed embryos and formed a transient spindle 2-4 h post-fusion (hpf). A mass of microtubules surrounded the swollen pseudo-pronucleus 5 hpf and a normal spindle was formed 13 hpf in cloned embryos. NuMAwas detected in the nucleus in germinal vesicle-stage oocytes, and it was concentrated at the spindle poles in both meiotic and mitotic metaphase. In both donor cell nucleus and enucleated oocyte cytoplasm, NuMA was not detected, while NuMA reappeared in pseudo-pronucleus as reconstructed embryo development proceeded. However, no evident NuMA staining was observed in the poles of transient spindle and first mitotic spindle in nuclear transfer eggs. These results indicate that NuMA localization and its spindle pole tethering function are different during rabbit oocyte meiosis and cloned embryo mitosis.

Rabbit oocyte cytoplasm supports development of nuclear transfer embryos derived from the somatic cells of the camel and Tibetan antelope.

This study was designed to examine the ability of rabbit metaphase II oocyte cytoplasm to support the development of interspecies nuclear transfer embryos reconstructed using donor nuclei from different species. Skin fibroblast cells from a camel and Tibetan antelope were used as donor nuclei. As a first step, we investigated the efficiency of different activation protocols by comparing the parthenogenetic development of rabbit oocytes. The protocol that yielded the highest blastocyst rate was used to activate the reconstructed embryos in nuclear transfer experiments. In addition, the effect of donor cell serum starvation on the development of the reconstructed embryo was also examined. More than half of the karyoplast-cytoplast couplets could be fused, and about one third of the reconstructed embryos were capable of completing first cleavage, regardless of the species of donor nuclei. Some of the cleaving reconstructed embryos were even capable of progressing further and developing to the blastocyst stage (1.4-8.7% for the Tibetan antelope and 0-7.5% for the camel, respectively). Our results suggest that the mechanisms regulating early embryo development may be conserved among mammalian species and some factors existing in rabbit oocyte cytoplasm for somatic nucleus reprogramming and dedifferentiation may not be species-specific. Rabbit oocyte cytoplasm can reprogram donor nuclei regardless of the origin of the nucleus and support in vitro development to an advanced stage.

Dynamic changes in NuMA and microtubules in monkey-rabbit nuclear transfer embryos.

Previous reports have indicated that failure in cloning monkey is attributed to the removal of nuclear mitotic apparatus (NuMA) during enucleation and subsequent abnormal organization of mitotic apparatus. This study investigated the transformation and assembly of tubulin and NuMA protein during the first cell cycle of cloned monkey embryos reconstructed by using enucleated rabbit oocytes as recipients. After the oocyte fused with a fibroblast, extensive microtubule organization was observed around the introduced nucleus in most reconstructed embryos, suggesting the introduction of a somatic cell centrosome. A high proportion of fibroblast nuclei transferred into non-activated oocytes underwent premature chromosome condensation (PCC), transient spindle organization and chromosomes separation, followed by the formation of two pronucleus-like structures. In contrast, fibroblast nuclei in pre-activated ooplasm rarely underwent PCC, but formed a swollen pronucleus-like structure. Normal spindles were observed in about one third of the cloned embryos reconstructed by both methods. After transferring monkey fibroblasts into NuMA-removed enucleated rabbit oocytes, NuMA was localized in pseudo-pronuclei and gradually moved to mitotic spindle poles at the first mitotic spindle poles. NuMA antibody microinjection resulted in spindle disorganization and chromosome misalignment, but did not significantly affect early cleavage. Our findings indicate that: 1. NuMA in donor monkey fibroblast may contribute to form a normal spindle in enucleated rabbit oocyte; 2. when non-activated cytoplasts and pre-activated cytoplasts are used as recipients, the donor nuclei undergo different morphological changes, but yield similar early embryo development; 3. although abnormal spindle organization and chromosome alignment may cause low efficiency of animal cloning, these abnormalities do not significantly affect early cleavage.

Apoptosis in rabbit embryos produced by fertilization or nuclear transfer with fibroblasts and cumulus cells.

In this study, we investigated the development, the cell number of the blastocyst, and apoptosis in rabbit nuclear transfer (NT) embryos derived from adult fibroblasts and cumulus cells as compared with embryos derived from in vivo fertilization and in vitro culture. The developmental rate and the total cell number of the blastocyst were significantly lower in NT embryos than in fertilized embryos (FEs). The type of donor cells did not affect the embryonic developmental rate and the total cell number of blastocysts in NT groups. The present study investigated the onset and the frequency of apoptosis in NT embryos and FEs by using a terminal deoxynucleotidyl transferase-mediated dUTP nick and labeling (TUNEL) assay. The earliest positive TUNEL signals were detected at the eight-cell stage in NT embryos and at the morula stage in FEs. The apoptotic index of the total blastocysts, the inner cell mass and the trophoderm was greatly higher in the NT embryos than in FEs. Moreover, the apoptotic index of the blastocyst from fibroblasts was significantly higher than that of the blastocyst from cumulus cells.

Inhibitory effects of cAMP and protein kinase C on meiotic maturation and MAP kinase phosphorylation in porcine oocytes.

The regulation of MAP kinase phosphorylation by cAMP and protein kinase C (PKC) modulators during pig oocyte maturation was studied by Western immunoblotting. We showed that both forskolin and IBMX inhibited MAP kinase phosphorylation and meiosis resumption in a dose-dependent manner, and this inhibitory effect was overcome by the protein phosphatase inhibitor, okadaic acid. Pharmacological PKC activator phorbol myristate acetate or physiological PKC activator diC8 also delayed MAP kinase phosphorylation and meiosis resumption, and their effect was abrogated by PKC inhibitors, staurosporine, and calphostin C. The results suggest that meiotic resumption is inhibited by elevation of cAMP or delayed by activation of PKC probably via down-regulation of MAP kinase activation, which is mediated by protein phosphatase, during pig oocyte maturation.