Using polyacrylamide hydrogel to adsorb chloride ions in cement-based materials.

Concrete material is an important engineering material for modern marine engineering construction, but the presence of chloride ions in sea sand and seawater can cause corrosion of reinforcing steel, which greatly endangers the safety of reinforced concrete structures. Gel is an environmental friendly functional material which has the functions of exchange and adsorption of ions. Therefore, in this paper, polyacrylamide (PAM) gels were prepared for chloride ions adsorption in a reinforced concrete system. The chloride ions adsorption behavior of PAM gel in simulated seawater and cement were investigated and the maximum adsorption capacity of chloride ions in simulated seawater was 32.67 mg g-1. In addition, compared with the cement sample without gel, the chloride ion content in the cement sample containing 1.5 wt% gel was reduced 46.8% at a depth of 0-2.5 mm from the sample's surface. The results showed that PAM gel can effectively adsorb the chloride ion and improve the chloride ion penetration resistance in cement because the three-dimensional network structure of PAM gels allowed chloride ions to enter the inside of the gel. This gel has potential applications in the field of marine construction.

Cross-Attention Based Multi-Resolution Feature Fusion Model for Self-Supervised Cervical OCT Image Classification.

Cervical cancer seriously endangers the health of the female reproductive system and even risks women's life in severe cases. Optical coherence tomography (OCT) is a non-invasive, real-time, high-resolution imaging technology for cervical tissues. However, since the interpretation of cervical OCT images is a knowledge-intensive, time-consuming task, it is tough to acquire a large number of high-quality labeled images quickly, which is a big challenge for supervised learning. In this study, we introduce the vision Transformer (ViT) architecture, which has recently achieved impressive results in natural image analysis, into the classification task of cervical OCT images. Our work aims to develop a computer-aided diagnosis (CADx) approach based on a self-supervised ViT-based model to classify cervical OCT images effectively. We leverage masked autoencoders (MAE) to perform self-supervised pre-training on cervical OCT images, so the proposed classification model has a better transfer learning ability. In the fine-tuning process, the ViT-based classification model extracts multi-scale features from OCT images of different resolutions and fuses them with the cross-attention module. The ten-fold cross-validation results on an OCT image dataset from a multi-center clinical study of 733 patients in China indicate that our model achieved an AUC value of 0.9963 ± 0.0069 with a 95.89 ± 3.30% sensitivity and 98.23 ± 1.36 % specificity, outperforming some state-of-the-art classification models based on Transformers and convolutional neural networks (CNNs) in the binary classification task of detecting high-risk cervical diseases, including high-grade squamous intraepithelial lesion (HSIL) and cervical cancer. Furthermore, our model with the cross-shaped voting strategy achieved a sensitivity of 92.06% and specificity of 95.56% on an external validation dataset containing 288 three-dimensional (3D) OCT volumes from 118 Chinese patients in a different new hospital. This result met or exceeded the average of four medical experts who have used OCT for over one year. In addition to promising classification performance, our model has a remarkable ability to detect and visualize local lesions using the attention map of the standard ViT model, providing good interpretability for gynecologists to locate and diagnose possible cervical diseases.

Design of a Frequency Selective Rasorber Based on a Band-Patterned Octagonal Ring.

In this study, a dual-polarization and low-profile frequency-selective rasorber (FSR) constructed from a novel band-patterned octagonal ring and dipole slot-type elements is investigated. We show the process of designing from a full octagonal ring to realize a lossy frequency selective surface of our proposed FSR, and it has a passband with low insertion loss between the two absorptive bands. An equivalent circuit for our designed FSR is modeled to explain the introduction of the parallel resonance. Surface current, electric energy, and magnetic energy of the FSR are further investigated to illustrate the working mechanism. Simulated results indicate that S11 < -10 dB bandwidth within 5.2-14.8 GHz, S21 > -3 dB passband within 9.62-11.72 GHz, lower absorptive bandwidth within 5.02-8.80 GHz, and upper absorptive bandwidth within 12.94-14.89 GHz are obtained under normal incidence. Meanwhile, our proposed FSR possesses the properties of dual-polarization and angular stability. To verify the simulated results, a sample with thickness of 0.097 λL is manufactured, and the results are experimentally verified.

Engineering orthotopic tumor spheroids with organ-specific vasculatures for local chemoembolization evaluation.

Developing a three-dimensional (3D) in vitro tumor model with vasculature systems suitable for testing endovascular interventional therapies remains a challenge. Here we develop an orthotopic liver tumor spheroid model that captures the organ-level complexity of vasculature systems and the extracellular matrix to evaluate transcatheter arterial chemoembolization (TACE) treatment. The orthotopic tumor spheroids are derived by seeding HepG2 cell colonies with controlled size and location surrounding the portal triads in a decellularized rat liver matrix and are treated by clinically relevant drug-eluting beads embolized in a portal vein vasculature while maintaining dynamic physiological conditions with nutrient and oxygen supplies through the hepatic vein vasculature. The orthotopic tumor model exhibits strong drug retention inside the spheroids and embolization location-dependent cellular apoptosis responses in an analogous manner to in vivo conditions. Such a tumor spheroid model built in a decellularized scaffold containing organ-specific vasculatures, which closely resembles the unique tumor microenvironment, holds the promise to efficiently assess various diagnostic and therapeutic strategies for endovascular therapies.

Peroxisome Proliferator-Activated Receptor α Attenuates Hypertensive Vascular Remodeling by Protecting Vascular Smooth Muscle Cells from Angiotensin II-Induced ROS Production.

Vascular remodeling is the fundamental basis for hypertensive disease, in which vascular smooth muscle cell (VSMC) dysfunction plays an essential role. Previous studies suggest that the activation of peroxisome proliferator-activated receptor α (PPARα) by fibrate drugs has cardiovascular benefits independent of the lipid-lowering effects. However, the underlying mechanism remains incompletely understood. This study explored the role of PPARα in angiotensin II (Ang II)-induced vascular remodeling and hypertension using VSMC-specific Ppara-deficient mice. The PPARα expression was markedly downregulated in the VSMCs upon Ang II treatment. A PPARα deficiency in the VSMC significantly aggravated the Ang II-induced hypertension and vascular stiffness, with little influence on the cardiac function. The morphological analyses demonstrated that VSMC-specific Ppara-deficient mice exhibited an aggravated vascular remodeling and oxidative stress. In vitro, a PPARα deficiency dramatically increased the production of mitochondrial reactive oxidative species (ROS) in Ang II-treated primary VSMCs. Finally, the PPARα activation by Wy14643 improved the Ang II-induced ROS production and vascular remodeling in a VSMC PPARα-dependent manner. Taken together, these data suggest that PPARα plays a critical protective role in Ang II-induced hypertension via attenuating ROS production in VSMCs, thus providing a potential therapeutic target for hypertensive diseases.

Cervical optical coherence tomography image classification based on contrastive self-supervised texture learning.

BACKGROUND: Cervical cancer (CC) seriously affects the health of the female reproductive system. Optical coherence tomography (OCT) emerged as a noninvasive, high-resolution imaging technology for cervical disease detection. However, OCT image annotation is knowledge-intensive and time-consuming, which impedes the training process of deep-learning-based classification models. PURPOSE: This study aims to develop a computer-aided diagnosis (CADx) approach to classifying in-vivo cervical OCT images based on self-supervised learning. METHODS: In addition to high-level semantic features extracted by a convolutional neural network (CNN), the proposed CADx approach designs a contrastive texture learning strategy to leverage unlabeled cervical OCT images' texture features. We conducted 10-fold cross-validation on the OCT image dataset from a multicenter clinical study on 733 patients from China. RESULTS: In a binary classification task for detecting high-risk diseases, including high-grade squamous intraepithelial lesion and CC, our method achieved an area-under-the-curve value of 0.9798 ± 0.0157 with a sensitivity of 91.17% ± 4.99% and a specificity of 93.96% ± 4.72% for OCT image patches; also, it outperformed two out of four medical experts on the test set. Furthermore, our method achieved 91.53% sensitivity and 97.37% specificity on an external validation dataset containing 287 three-dimensional OCT volumes from 118 Chinese patients in a new hospital using a cross-shaped threshold voting strategy. CONCLUSIONS: The proposed contrastive-learning-based CADx method outperformed the end-to-end CNN models and provided better interpretability based on texture features, which holds great potential to be used in the clinical protocol of "see-and-treat."

Effects of Involuntary and Voluntary Exercise in Combination with Acousto-Optic Stimulation on Adult Neurogenesis in an Alzheimer's Mouse Model.

Single-factor intervention, such as physical exercise and auditory and visual stimulation, plays a positive role on the prevention and treatment of Alzheimer's disease (AD); however, the therapeutic effects of single-factor intervention are limited. The beneficial effects of these multifactor combinations on AD and its molecular mechanism have yet to be elucidated. Here, we investigated the effect of multifactor intervention, voluntary wheel exercise, and involuntary treadmill running in combination with acousto-optic stimulation, on adult neurogenesis and behavioral phenotypes in a mouse model of AD. We found that 4 weeks of multifactor intervention can significantly increase the production of newborn cells (BrdU+ cells) and immature neurons (DCX+ cells) in the hippocampus and lateral ventricle of Aß oligomer-induced mice. Importantly, the multifactor intervention could promote BrdU+ cells to differentiate into neurons (BrdU+ DCX+ cells or BrdU+ NeuN+ cells) and astrocytes (BrdU+GFAP+ cells) in the hippocampus and ameliorate Aß oligomer-induced cognitive impairment and anxiety- and depression-like behaviors in mice evaluated by novel object recognition, Morris water maze tests, elevated zero maze, forced swimming test, and tail suspension test, respectively. Moreover, multifactor intervention could lead to an increase in the protein levels of PSD-95, SYP, DCX, NeuN, GFAP, Bcl-2, BDNF, TrkB, and pSer473-Akt and a decrease in the protein levels of BAX and caspase-9 in the hippocampal lysates of Aß oligomer-induced mice. Furthermore, sequencing analysis of serum metabolites revealed that aberrantly expressed metabolites modulated by multifactor intervention were highly enriched in the biological process associated with keeping neurons functioning and neurobehavioral function. Additionally, the intervention-mediated serum metabolites mainly participated in glutamate metabolism, glucose metabolism, and the tricarboxylic acid cycle in mice. Our findings suggest the potential of multifactor intervention as a non-invasive therapeutic strategy for AD to anti-Aß oligomer neurotoxicity.

Effects of involuntary treadmill running in combination with swimming on adult neurogenesis in an Alzheimer's mouse model.

Physical exercise plays a role on the prevention and treatment of Alzheimer's disease (AD), but the exercise mode and the mechanism for these positive effects is still ambiguous. Here, we investigated the effect of an aerobic interval exercise, running in combination with swimming, on behavioral dysfunction and associated adult neurogenesis in a mouse model of AD. We demonstrate that 4 weeks of the exercise could ameliorate Aß42 oligomer-induced cognitive impairment in mice utilizing Morris water maze tests. Additionally, the exercised Aß42 oligomer-induced mice exhibited a significant reduction of anxiety- and depression-like behaviors compared to the sedentary Aß42 oligomer-induced mice utilizing an Elevated zero maze and a Tail suspension test. Moreover, by utilizing 5'-bromodeoxyuridine (BrdU) as an exogenous cell tracer, we found that the exercised Aß42 oligomer-induced mice displayed a significant increase in newborn cells (BrdU+ cells), which differentiated into a majority of neurons (BrdU+ DCX+ cells or BrdU+NeuN+ cells) and a few of astrocytes (BrdU+GFAP+ cells). Likewise, the exercised Aß42 oligomer-induced mice also displayed the higher levels of NeuN, PSD95, synaptophysin, Bcl-2 and lower level of GFAP protein. Furthermore, alteration of serum metabolites in transgenic AD mice between the exercised and sedentary group were significantly associated with lipid metabolism, amino acid metabolism, and neurotransmitters. These findings suggest that combined aerobic interval exercise-mediated metabolites and proteins contributed to improving adult neurogenesis and behavioral performance after AD pathology, which might provide a promising therapeutic strategy for AD.

Shape-Anisotropic Microembolics Generated by Microfluidic Synthesis for Transarterial Embolization Treatment.

Particulate embolic agents with calibrated sizes, which employ interventional procedures to achieve endovascular embolization, have recently attracted tremendous interest in therapeutic embolotherapies for a wide plethora of diseases. However, the particulate shape effect, which may play a critical role in embolization performances, has been rarely investigated. Here, polyvinyl alcohol (PVA)-based shape-anisotropic microembolics are developed using a facile droplet-based microfluidic fabrication method via heat-accelerated PVA-glutaraldehyde crosslinking reaction at a mild temperature of 38 ° C. Precise geometrical controls of the microembolics are achieved with a nearly capsule shape through regulating surfactant concentration and flow rate ratio between dispersed phase and continuous phase in the microfluidics. Two specific models are employed, i.e., in vitro decellularized rabbit liver embolization model and in vivo rabbit ear embolization model, to systematically evaluate the embolization behaviors of the nonspherical microembolics. Compared to microspheres of the same volume, the elongated microembolics demonstrated advantageous endovascular navigation capability, penetration depth and embolization stability due to their comparatively smaller radial diameter and their central cylindrical part providing larger contact area with distal vessels. Such nonspherical microembolics present a promising platform to apply shape anisotropy to achieve distinctive therapeutic effects for endovascular treatments.

Facile formation of giant elastin-like polypeptide vesicles as synthetic cells.

We demonstrate the facile and robust generation of giant peptide vesicles by using an emulsion transfer method. These robust vesicles can sustain chemical and physical stresses. The peptide vesicles can host cell-free expression reactions by encapsulating essential ingredients. We show the incorporation of another cell-free expressed elastin-like polypeptide into the existing membrane of the peptide vesicles.

Visualization and Evaluation of Chemoembolization on a 3D Decellularized Organ Scaffold.

Transarterial chemoembolization (TACE) has emerged as the mainstay treatment for patients suffering from unresectable intermediate hepatocellular carcinoma and also holds the potential to treat other types of hypervascular cancers such as renal cell carcinoma. However, an in vitro model for evaluating both embolic performance and drug-release kinetics of the TACE embolic agents is still lacking since the current models greatly simplified the in vivo vascular systems as well as the extracellular matrices (ECM) in the organs. Here, we developed a decellularized organ model with preserved ECM and vasculatures as well as a translucent appearance to investigate chemoembolization performances of a clinically widely used embolic agent, i.e., a doxorubicin-loaded ethiodised oil (EO)-based emulsion. We, for the first time, utilized an ex vivo model to evaluate the liquid-based embolic agent in two organs, i.e., liver and kidneys. We found that the EO-based emulsion with enhanced stability by incorporating an emulsifier, i.e., hydrogenated castor oil-40 (HCO), showed an enhanced occlusion level and presented sustained drug release in the ex vivo liver model, suggesting an advantageous therapeutic effect for TACE treatment of hepatocellular carcinoma. In contrast, we observed that drug-release burst happened when applying the same therapy in the kidney model even with the HCO emulsifier, which may be explained by the presence of the specific renal vasculature and calyceal systems, indicating an unfavorable effect in the renal tumor treatment. Such an ex vivo model presents a promising template for chemoembolization evaluation before in vivo experiments for the development of novel embolic agents.

Analysis of the Coptis chinensis genome reveals the diversification of protoberberine-type alkaloids.

Chinese goldthread (Coptis chinensis Franch.), a member of the Ranunculales, represents an important early-diverging eudicot lineage with diverse medicinal applications. Here, we present a high-quality chromosome-scale genome assembly and annotation of C. chinensis. Phylogenetic and comparative genomic analyses reveal the phylogenetic placement of this species and identify a single round of ancient whole-genome duplication (WGD) shared by the Ranunculaceae. We characterize genes involved in the biosynthesis of protoberberine-type alkaloids in C. chinensis. In particular, local genomic tandem duplications contribute to member amplification of a Ranunculales clade-specific gene family of the cytochrome P450 (CYP) 719. The functional versatility of a key CYP719 gene that encodes the (S)-canadine synthase enzyme involved in the berberine biosynthesis pathway may play critical roles in the diversification of other berberine-related alkaloids in C. chinensis. Our study provides insights into the genomic landscape of early-diverging eudicots and provides a valuable model genome for genetic and applied studies of Ranunculales.

Inverse Design of Self-Assembling Diamond Photonic Lattices from Anisotropic Colloidal Clusters.

Colloidal nanoparticles with anisotropic interactions are promising building blocks for the fabrication of complex functional materials. A challenge in the self-assembly of colloidal particles is the rational design of geometry and chemistry to program the formation of a desired target structure. We report an inverse design procedure integrating Langevin dynamics simulations and evolutionary algorithms to engineer anisotropic patchy colloidal clusters to spontaneously assemble into a cubic diamond lattice possessing a complete photonic band gap. The combination of a tetrahedral cluster geometry and optimized placement of a single type of anisotropic interaction patch results in a colloidal building block predicted to assemble a cubic diamond lattice with more than 82% yield. This design represents an experimentally viable colloidal building block capable of high-fidelity assembly of a cubic diamond lattice.

In search of a novel chassis material for synthetic cells: emergence of synthetic peptide compartment.

Giant lipid vesicles have been used extensively as a synthetic cell model to recapitulate various life-like processes, including in vitro protein synthesis, DNA replication, and cytoskeleton organization. Cell-sized lipid vesicles are mechanically fragile in nature and prone to rupture due to osmotic stress, which limits their usability. Recently, peptide vesicles have been introduced as a synthetic cell model that would potentially overcome the aforementioned limitations. Peptide vesicles are robust, reasonably more stable than lipid vesicles and can withstand harsh conditions including pH, thermal, and osmotic variations. This mini-review summarizes the current state-of-the-art in the design, engineering, and realization of peptide-based chassis materials, including both experimental and computational work. We present an outlook for simulation-aided and data-driven design and experimental realization of engineered and multifunctional synthetic cells.

Structure, Function, Diversity, and Composition of Fungal Communities in Rhizospheric Soil of Coptis chinensis Franch under a Successive Cropping System.

Soil types and cropping systems influence the diversity and composition of the rhizospheric microbial communities. Coptis chinensis Franch is one of the most important medicinal plants in China. In the current study, we provide detailed information regarding the diversity and composition of rhizospheric fungal communities of the C. chinensis plants in continuous cropping fields and fallow fields in two seasons (winter and summer), using next-generation sequencing. Alpha diversity was higher in the five-year C. chinensis field and lower in fallow fields. Significant differences analysis confirmed more fungi in the cultivated field soil than in fallow fields. Additionally, PCoA of beta diversity indices revealed that samples associated with the cultivated fields and fallow fields in different seasons were separated. Five fungal phyla (Ascomycota, Basidiomycota, Chytridiomycota, Glomeromycota and Mucoromycota) were identified from the soil samples in addition to the unclassified fungal taxa and Cryptomycota, and among these phyla, Ascomycota was predominantly found. FUNGuild fungal functional prediction revealed that saprotroph was the dominant trophic type in all two time-series soil samples. Redundancy analysis (RDA) of the dominant phyla data and soil physiochemical properties revealed the variations in fungal community structure in the soil samples. Knowledge from the present study could provide a valuable reference for solving the continuous cropping problems and promote the sustainable development of the C. chinensis industry.

Computer-Aided Diagnosis in Histopathological Images of the Endometrium Using a Convolutional Neural Network and Attention Mechanisms.

Uterine cancer (also known as endometrial cancer) can seriously affect the female reproductive system, and histopathological image analysis is the gold standard for diagnosing endometrial cancer. Due to the limited ability to model the complicated relationships between histopathological images and their interpretations, existing computer-aided diagnosis (CAD) approaches using traditional machine learning algorithms often failed to achieve satisfying results. In this study, we develop a CAD approach based on a convolutional neural network (CNN) and attention mechanisms, called HIENet. In the ten-fold cross-validation on ∼3,300 hematoxylin and eosin (H&E) image patches from ∼500 endometrial specimens, HIENet achieved a 76.91 ± 1.17% (mean ± s. d.) accuracy for four classes of endometrial tissue, i.e., normal endometrium, endometrial polyp, endometrial hyperplasia, and endometrial adenocarcinoma. Also, HIENet obtained an area-under-the-curve (AUC) of 0.9579 ± 0.0103 with an 81.04 ± 3.87% sensitivity and 94.78 ± 0.87% specificity in a binary classification task that detected endometrioid adenocarcinoma. Besides, in the external validation on 200 H&E image patches from 50 randomly-selected female patients, HIENet achieved an 84.50% accuracy in the four-class classification task, as well as an AUC of 0.9829 with a 77.97% (95% confidence interval, CI, 65.27%∼87.71%) sensitivity and 100% (95% CI, 97.42%∼100.00%) specificity. The proposed CAD method outperformed three human experts and five CNN-based classifiers regarding overall classification performance. It was also able to provide pathologists better interpretability of diagnoses by highlighting the histopathological correlations of local pixel-level image features to morphological characteristics of endometrial tissue.

Inverse design of self-assembling colloidal crystals with omnidirectional photonic bandgaps.

Open colloidal lattices possessing omnidirectional photonic bandgaps in the visible or near-visible regime are attractive optical materials the realization of which has remained elusive. We report the use of an inverse design strategy termed landscape engineering that rationally sculpts the free energy self-assembly landscape using evolutionary algorithms to discover anisotropic patchy colloids capable of spontaneously assembling pyrochlore and cubic diamond lattices possessing complete photonic bandgaps. We validate the designs in computer simulations to demonstrate the defect-free formation of these lattices via a two-stage hierarchical assembly mechanism. Our approach demonstrates a principled strategy for the inverse design of self-assembling colloids for the bottom-up fabrication of desired crystal lattices.

Inhibitory effects of oxymatrine on hepatic stellate cells activation through TGF-ß/miR-195/Smad signaling pathway.

BACKGROUND: Oxymatrine (OM), a quinolizidine alkaloid extracted from a herb Sophorae Flavescentis Radix, has been used to treat liver fibrotic diseases. However, the mechanism of its anti-fibrosis effects is still unclear. TGF-ß/Smad signaling and miR-195 have been proved to paly an important role in hepatic stellate cells (HSCs) activation and liver fibrosis. In this study, we investigated whether OM could inhibit HSCs activation through TGF-ß1/miR-195/Smads signaling or not. METHODS: First, the effects of OM on HSC-T6 in different concentrations and time points were tested by MTT assay. We choose three appropriate concentrations of OM as treatment concentrations in following experiment. By Quantitative Real-time PCR and Western Blot, then we investigated the effect of OM on miR-195, Smad7 and α-SMA's expressions to prove the correlation between OM and the TGF-ß1/miR-195/Smads signaling. Last, miR-195 mimic and INF-γ were used to investigate the relation between miR-195 and OM in HSC activation. RESULTS: Our results showed that the proliferation of HSC was significantly inhibited when OM concentration was higher than 200 µg/mL after 24 h, 100 µg/mL after 48 h and 10 µg/mL after 72 h. The IC50 of OM after 24, 48 and 72 h were 539, 454, 387 µg/mL respectively. OM could down-regulate miR-195 and α-SMA (P < 0.01), while up-regulate Smad7 (P < 0.05). In HSC-T6 cells transfected with miR-195 mimic and pretreated with OM, miR-195 and α-SMA were up-regulated (P < 0.05), and Smad7 was down-regulated (P < 0.05) . CONCLUSIONS: Given these results, OM could inhibit TGF-ß1 induced activation of HSC-T6 proliferation in a dose-dependent and time-dependent manner to some extent. We proved that OM inhibited HSC activation through down-regulating the expression of miR-195 and up-regulating Smad7.

Computer-Aided Diagnosis of Label-Free 3-D Optical Coherence Microscopy Images of Human Cervical Tissue.

OBJECTIVE: Ultrahigh-resolution optical coherence microscopy (OCM) has recently demonstrated its potential for accurate diagnosis of human cervical diseases. One major challenge for clinical adoption, however, is the steep learning curve clinicians need to overcome to interpret OCM images. Developing an intelligent technique for computer-aided diagnosis (CADx) to accurately interpret OCM images will facilitate clinical adoption of the technology and improve patient care. METHODS: 497 high-resolution three-dimensional (3-D) OCM volumes (600 cross-sectional images each) were collected from 159 ex vivo specimens of 92 female patients. OCM image features were extracted using a convolutional neural network (CNN) model, concatenated with patient information [e.g., age and human papillomavirus (HPV) results], and classified using a support vector machine classifier. Ten-fold cross-validations were utilized to test the performance of the CADx method in a five-class classification task and a binary classification task. RESULTS: An 88.3 ± 4.9% classification accuracy was achieved for five fine-grained classes of cervical tissue, namely normal, ectropion, low-grade and high-grade squamous intraepithelial lesions (LSIL and HSIL), and cancer. In the binary classification task [low-risk (normal, ectropion, and LSIL) versus high-risk (HSIL and cancer)], the CADx method achieved an area-under-the-curve value of 0.959 with an 86.7 ± 11.4% sensitivity and 93.5 ± 3.8% specificity. CONCLUSION: The proposed deep-learning-based CADx method outperformed four human experts. It was also able to identify morphological characteristics in OCM images that were consistent with histopathological interpretations. SIGNIFICANCE: Label-free OCM imaging, combined with deep-learning-based CADx methods, holds a great promise to be used in clinical settings for the effective screening and diagnosis of cervical diseases.

Segmentation of Drosophila heart in optical coherence microscopy images using convolutional neural networks.

Convolutional neural networks (CNNs) are powerful tools for image segmentation and classification. Here, we use this method to identify and mark the heart region of Drosophila at different developmental stages in the cross-sectional images acquired by a custom optical coherence microscopy (OCM) system. With our well-trained CNN model, the heart regions through multiple heartbeat cycles can be marked with an intersection over union of ~86%. Various morphological and dynamical cardiac parameters can be quantified accurately with automatically segmented heart regions. This study demonstrates an efficient heart segmentation method to analyze OCM images of the beating heart in Drosophila.