Diagnostic performance for pulmonary adenocarcinoma on CT: comparison of radiologists with and without three-dimensional convolutional neural network.

OBJECTIVES: To compare diagnostic performance for pulmonary invasive adenocarcinoma among radiologists with and without three-dimensional convolutional neural network (3D-CNN). METHODS: Enrolled were 285 patients with adenocarcinoma in situ (AIS, n = 75), minimally invasive adenocarcinoma (MIA, n = 58), and invasive adenocarcinoma (IVA, n = 152). A 3D-CNN model was constructed with seven convolution-pooling and two max-pooling layers and fully connected layers, in which batch normalization, residual connection, and global average pooling were used. Only the flipping process was performed for augmentation. The output layer comprised two nodes for two conditions (AIS/MIA and IVA) according to prognosis. Diagnostic performance of the 3D-CNN model in 285 patients was calculated using nested 10-fold cross-validation. In 90 of 285 patients, results from each radiologist (R1, R2, and R3; with 9, 14, and 26 years of experience, respectively) with and without the 3D-CNN model were statistically compared. RESULTS: Without the 3D-CNN model, accuracy, sensitivity, and specificity of the radiologists were as follows: R1, 70.0%, 52.1%, and 90.5%; R2, 72.2%, 75%, and 69%; and R3, 74.4%, 89.6%, and 57.1%, respectively. With the 3D-CNN model, accuracy, sensitivity, and specificity of the radiologists were as follows: R1, 72.2%, 77.1%, and 66.7%; R2, 74.4%, 85.4%, and 61.9%; and R3, 74.4%, 93.8%, and 52.4%, respectively. Diagnostic performance of each radiologist with and without the 3D-CNN model had no significant difference (p > 0.88), but the accuracy of R1 and R2 was significantly higher with than without the 3D-CNN model (p < 0.01). CONCLUSIONS: The 3D-CNN model can support a less-experienced radiologist to improve diagnostic accuracy for pulmonary invasive adenocarcinoma without deteriorating any diagnostic performances. KEY POINTS: • The 3D-CNN model is a non-invasive method for predicting pulmonary invasive adenocarcinoma in CT images with high sensitivity. • Diagnostic accuracy by a less-experienced radiologist was better with the 3D-CNN model than without the model.

Activation of PKC induces leukocyte adhesion by the dephosphorylation of ERM.

Although circulating leukocytes are non-adherent cells, they also undergo adhesion in response to external stimuli. To elucidate this switch mechanism, we investigated PMA-induced cell adhesion in myelomonocytic KG-1 cells. PMA induced microvillius collapse, decrease of cell surface rigidity and exclusion of sialomucin from adhesion sites. All these adhesion-contributing events are linked to dephosphorylation of Ezrin/Radixin/Moesin (ERM) proteins. Indeed, PMA-treatment induced quick decrease of phosphorylated ERM proteins, while expression of Moesin-T558D, a phospho-mimetic mutant, inhibited PMA-induced cell adhesion. PMA-induced cell adhesion and ERM-dephophorylation were inhibited by PKC inhibitors or by a phosphatase inhibitor, indicating the involvement of PKC and protein phophatase in these processes. In peripheral T lymphocytes, ERM-dephosphorylation by adhesion-inducing stimuli was inhibited by a PKC inhibitor. Combined, these findings strongly suggest that external stimuli induce ERM-dephosphorylation via the activation of PKC in leukocytes and that ERM-dephosphorylation leads to leukocytes' adhesion.

Convolutional Neural Network Can Recognize Drug Resistance of Single Cancer Cells.

It is known that single or isolated tumor cells enter cancer patients' circulatory systems. These circulating tumor cells (CTCs) are thought to be an effective tool for diagnosing cancer malignancy. However, handling CTC samples and evaluating CTC sequence analysis results are challenging. Recently, the convolutional neural network (CNN) model, a type of deep learning model, has been increasingly adopted for medical image analyses. However, it is controversial whether cell characteristics can be identified at the single-cell level by using machine learning methods. This study intends to verify whether an AI system could classify the sensitivity of anticancer drugs, based on cell morphology during culture. We constructed a CNN based on the VGG16 model that could predict the efficiency of antitumor drugs at the single-cell level. The machine learning revealed that our model could identify the effects of antitumor drugs with ~0.80 accuracies. Our results show that, in the future, realizing precision medicine to identify effective antitumor drugs for individual patients may be possible by extracting CTCs from blood and performing classification by using an AI system.

Sialomucin and phosphorylated-ERM are inhibitors for cadherin-mediated aggregate formation.

Cell adhesion is mediated by adhesion molecules, but also regulated by adhesion inhibitory molecules. Molecules such as leukocyte sialomucin and phosphorylated-Ezrin/Radixin/Moesin (ERM) inhibit cell-substratum adhesion. Here we show that these adhesion inhibitory molecules also inhibit aggregate formation of adherent cells in suspension culture. Expression of sialomucin, CD43 or CD34, inhibited formation of packed aggregates in HEK293T cells. Deletion mutant analysis and enzymatic cleavage indicated the significance of the extracellular sialomucin domain for this inhibition. Meanwhile, phosphorylated-ERM were decreased coincidently with aggregate formation. Combined with the inhibition of aggregate formation by the expression of phospho-mimetic Moesin mutant (Moesin-T558D), phosphorylated-ERM are inhibitors for aggregate formation. Increase of phosphorylated-ERM by CD43 and sialomucin-dependence of Moesin-T558D's inhibition indicate that sialomucin and phosphorylated-ERM collaborate to inhibit aggregate formation. Because aggregate formation of HEK293T cells is mediated by N-cadherin, sialomucin and phosphorylated-ERM inhibit cadherin-mediated cell-cell adhesion. Thus, sialomucin and phosphorylated-ERM are inhibitors for both cell-cell adhesion and cell-substratum adhesion, and regulation of these inhibitory molecules is essential for cell adhesion.

Identification of kinases and regulatory proteins required for cell migration using a transfected cell-microarray system.

BACKGROUND: Cell migration plays a major role in a variety of normal biological processes, and a detailed understanding of the associated mechanisms should lead to advances in the medical sciences in areas such as cancer therapy. Previously, we developed a simple chip, based on transfected-cell microarray (TCM) technology, for the identification of genes related to cell migration. In the present study, we used the TCM chip for high-throughput screening (HTS) of a kinome siRNA library to identify genes involved in the motility of highly invasive NBT-L2b cells. RESULTS: We performed HTS using TCM coupled with a programmed image tracer to capture time-lapse fluorescence images of siRNA-transfected cells and calculated speeds of cell movement. This first screening allowed us to identify 52 genes. After quantitative PCR (qPCR) and a second screening by a conventional transfection method, we confirmed that 32 of these genes were associated with the migration of NBT-L2b cells. We investigated the subcellular localization of proteins and levels of expression of these 32 genes, and then we used our results and databases of protein-protein interactions (PPIs) to construct a hypothetic but comprehensive signal network for cell migration. CONCLUSIONS: The genes that we identified belonged to several functional categories, and our pathway analysis suggested that some of the encoded proteins functioned as the hubs of networks required for cell migration. Our signal pathways suggest that epidermal growth factor receptor (EGFR) is an upstream regulator in the network, while Src and GRB2 seem to represent nodes for control of respective the downstream proteins that are required to coordinate the many cellular events that are involved in migration. Our microarray appears to be a useful tool for the analysis of protein networks and signal pathways related to cancer metastasis.

Fabrication and characterization for the nanoconjugates of pyridyldithio-functionalized multiwalled carbon nanotubes and cytochrome c in Langmuir-Blodgett films.

Monolayers and Langmuir-Blodgett (LB) films of pyridyldithio-functionalized multiwalled carbon nanotubes (pythio-MWNTs) conjugated with cytochrome c were investigated. Quartz crystal microbalance measurements revealed that monolayers of pythio-MWNTs could be deposited on the substrate surface with a surface density of about 270 ng/cm2. The as-prepared LB films of pythio-MWNTs could act as a support to immobilize cytochrome c with the surface density of about 5.2 microg/cm2. Composition, structure and morphology of the films were characterized by using absorption and X-ray photoelectron spectra as well as atomic force microscope. A reversible redox process for the immobilized cytochrome c was revealed with the cathodic and anionic potentials at about -0.55 and -0.3 V versus Ag/AgCl, respectively.

Single-cell manipulation and DNA delivery technology using atomic force microscopy and nanoneedle.

The recent single-cell manipulation technology using atomic force microscopy (AFM) not only allows high-resolution visualization and probing of biomolecules and cells but also provides spatial and temporal access to the interior of living cells via the nanoneedle technology. Here we review the development and application of single-cell manipulations and the DNA delivery technology using a nanoneedle. We briefly describe various DNA delivery methods and discuss their advantages and disadvantages. Fabrication of the nanoneedle, visualization of nanoneedle insertion into living cells, DNA modification on the nanoneedle surface, and the invasiveness of nanoneedle insertion into living cells are described. Different methods of DNA delivery into a living cell, such as lipofection, microinjection, and nanoneedles, are then compared. Finally, single-cell diagnostics using the nanoneedle and the perspectives of the nanoneedle technology are outlined. The nanoneedle-based DNA delivery technology provides new opportunities for efficient and specific introduction of DNA and other biomolecules into precious living cells with a high spatial resolution within a desired time frame. This technology has the potential to be applied for many basic cellular studies and for clinical studies such as single-cell diagnostics.

Probing actin filament and binding protein interaction using an atomic force microscopy.

Actin filaments play essential roles in many kinds of cellular functions by interacting with hundreds of actin binding proteins. Here we probe the interaction between actin filament and a binding protein, α-actinin, using an atomic force microscopy (AFM) and dynamic force spectroscopy (DFS). The distribution of rupturing events including specific and non-specific interactions of actin filament/α- actinin and BSA/α-actinin were analyzed. The rupture force of the actin filament/α-actinin binding was significantly larger than that of the BSA/α-actinin non-specific interaction, and the peaks represent typical multiple parallel bonds. In addition, based on the rupture forces in different loading rate DFS experiments, the dissociation constant of actin filament/α-actinin binding was estimated. The value is in good agreement with a previously reported value obtained by optical tweezer measurement. We expect that the present method will be useful for interaction measurement of actin filaments and many kinds of binding protein.

Transfer printing of transfected cell microarrays from poly(ethylene glycol)-oleyl surfaces onto biological hydrogels.

We have developed a novel technique for constructing microarrays of transfected mammalian cells on or in extracellular matrix (ECM) hydrogels by transfer printing from patterned poly(ethylene glycol) (PEG)-oleyl surfaces. A mixed solution of small interfering RNA (siRNA) and a transfection reagent was spotted on PEG-oleyl-coated glass slides using an ink-jet printer, and the cells were then transiently immobilized on the patterned transfection mixtures. After overlaying an ECM hydrogel sheet onto the immobilized cells, the cells sandwiched between the glass slide and the hydrogel sheet were incubated at 37°C for simultaneous transfection of siRNA into cells and adhesion of cells to the hydrogel sheet. Transfer of the adhered, transfected cells was completed by peeling off the hydrogel sheet. The knockdown of a model gene in the transferred cell microarray by the transfected siRNA was successfully confirmed. Transfected cell microarrays were also embedded within three-dimensional ECM hydrogels. In the three-dimensional hydrogel, the inhibition effect of siRNA on cancer cell invasion was evaluated by quantifying the size of cell clusters on the microarrays. These results indicate that transfection of cell microarrays on or in a biological matrix is a promising technique for high-throughput screening of disease-related genes by direct observation of cellular phenomena in a physiologically relevant environment.

Thermostability of Rhodopseudomonas viridis and Rhodospirillum rubrum chromatophores reflecting physiological conditions.

Relationships between growth conditions and thermostability were examined for photosynthetic inner membranes (chromatophores) from Rhodopseudomonas viridis and Rhodospirillum rubrum of which morphology, lipid composition, and protein/lipid rate are rather mutually different. Signals observed by differential scanning calorimetry of the chromatophores were correlated with thermal state transitions of the membrane components by reference to temperature dependencies of circular dichroism and absorption spectra of the purified supramolecule comprising a photoreaction center and surrounding light-harvesting pigment-protein complexes that are the prominent proteins in both membranes. The differential scanning calorimetry curves of those chromatophores exhibited different dependencies on growth stages and environmental temperatures. The obtained result appeared to reflect the differences in the protein/lipid rate and protein-lipid specificity between the two chromatophores.

Rapidly serum-degradable hydrogel templating fabrication of spherical tissues and curved tubular structures.

Development of the techniques for fabricating three-dimensional tissues still poses significant challenges for tissue engineering. We used hydrogels obtained from phenol-substituted amylopectin (AP-Ph) as templates for preparing multicellular spherical tissues (MSTs) and endothelialized curved tubular structures in type I collagen gel. AP-Ph hydrogel microparticles of diameter 200 µm and fibers of diameter 500 µm disappeared within hours of soaking in a serum-containing medium. HeLa cells and human endothelial cells were enclosed in the microparticles and hydrogel fibers, respectively, and then embedded in Ca-alginate microcapsules or the collagen gel. The enclosed cells were released in cavities formed by hydrogel degradation in the serum-containing medium. The released HeLa cells in the spherical cavities grew and formed MSTs, eventually filling the cavities. The spherical tissues were easily harvested by liquefying the Ca-alginate hydrogel microcapsule membrane by chelation using sodium citrate. The released endothelial cells grew on the tubular cavity surfaces and formed tubular structures. An endothelial cell network was formed by cell migration into the collagen gel. These results demonstrate the potential of serum-degradable AP-Ph hydrogels in constructing three-dimensional tissues.

Automated diagnosis of optical coherence tomography imaging on plaque vulnerability and its relation to clinical outcomes in coronary artery disease.

This study sought to develop a deep learning-based diagnostic algorithm for plaque vulnerability by analyzing intravascular optical coherence tomography (OCT) images and to investigate the relation between AI-plaque vulnerability and clinical outcomes in patients with coronary artery disease (CAD). A total of 1791 study patients who underwent OCT examinations were recruited from a multicenter clinical database, and the OCT images were first labeled as either normal, a stable plaque, or a vulnerable plaque by expert cardiologists. A DenseNet-121-based deep learning algorithm for plaque characterization was developed by training with 44,947 prelabeled OCT images, and demonstrated excellent differentiation among normal, stable plaques, and vulnerable plaques. Patients who were diagnosed with vulnerable plaques by the algorithm had a significantly higher rate of both events from the OCT-observed segments and clinical events than the patients with normal and stable plaque (log-rank p < 0.001). On the multivariate logistic regression analyses, the OCT diagnosis of a vulnerable plaque by the algorithm was independently associated with both types of events (p = 0.047 and p < 0.001, respectively). The AI analysis of intracoronary OCT imaging can assist cardiologists in diagnosing plaque vulnerability and identifying CAD patients with a high probability of occurrence of future clinical events.

Physical properties of mesenchymal stem cells are coordinated by the perinuclear actin cap.

Mesenchymal stem cells (MSCs) have been extensively investigated for their applications in regenerative medicine. Successful use of MSCs in cell-based therapies will rely on the ability to effectively identify their properties and functions with a relatively non-destructive methodology. In this study, we measured the surface stiffness and thickness of rat MSCs with atomic force microscopy and clarified their relation at a single-cell level. The role of the perinuclear actin cap in regulating the thickness, stiffness, and proliferative activity of these cells was also determined by using several actin cytoskeleton-modifying reagents. This study has helped elucidate a possible link between the physical properties and the physiological function of the MSCs, and the corresponding regulatory role of the actin cytoskeleton.

Regulation of cysteine-rich protein 2 localization by the development of actin fibers during smooth muscle cell differentiation.

Cysteine-rich protein 2 (CRP2) is a cofactor for smooth muscle cell (SMC) differentiation. Here, we examined the mechanism of CRP2 distribution dynamics during SMC differentiation. CRP2 protein directly associated with F-actin through its N-terminal LIM domain and Gly-rich region, as determined by ELISA. In undifferentiated cells that contain few actin stress fibers, CRP2 was broadly distributed throughout the whole cell, including the nucleus. After induction of SMC differentiation, CRP2 localized to actin stress fibers as they formed. The stress fiber-localized CRP2 entered the nucleus because of induced actin depolymerization. These CRP2 dynamics were reproduced by in silico simulation. CRP2 localization dynamics, which affect CRP2 function, are regulated by the formation of actin stress fibers in conjunction with SMC differentiation.

Successive detection of insulin-like growth factor-II bound to receptors on a living cell surface using an AFM.

In this study, we have developed a method of mechanical force detection for ligands bound to receptors on a cell surface, both of which are involved in a signal transduction pathway. This pathway is an autocrine pathway, involving the production of insulin-like growth factor-II (IGF-II) and activation of the IGF-I receptor, involved in myoblast differentiation induced by MyoD in C3H10T1/2 mouse mesenchymal stem cells. Differentiation of C3H10T1/2 was induced with the DNA demethylation agent 5-azacytidine (5-aza). The etched AFM tip used in the force detection had a flat surface of which about 10 µm(2) was in contact with a cell surface. The forces required to rupture the interactions of IGF-IIs on a cell and anti mouse IGF-II polyclonal antibody immobilized on an etched AFM tip were measured within 5 days of induction of differentiation. The mean unbinding force for a single paired antibody-ligand on a cell was about 81 pN, which was measured at a force loading rate of about 440 nN/s. The percentage of unbinding forces over 100 pN increased to 32% after 2 days from the addition of 5-aza to the medium. This method could be used in non-invasive and successive evaluation of a living cell's behavior.

Hydrolyzed eggshell membrane immobilized on phosphorylcholine polymer supplies extracellular matrix environment for human dermal fibroblasts.

We have found that a water-soluble alkaline-digested form of eggshell membrane (ASESM) can provide an extracellular matrix (ECM) environment for human dermal fibroblast cells (HDF) in vitro. Avian eggshell membrane (ESM) has a fibrous-meshwork structure and has long been utilized as a Chinese medicine for recovery from burn injuries and wounds in Asian countries. Therefore, ESM is expected to provide an excellent natural material for biomedical use. However, such applications have been hampered by the insolubility of ESM proteins. We have used a recently developed artificial cell membrane biointerface, 2-methacryloyloxyethyl phosphorylcholine polymer (PMBN) to immobilize ASESM proteins. The surface shows a fibrous structure under the atomic force microscope, and adhesion of HDF to ASESM is ASESM-dose-dependent. Quantitative mRNA analysis has revealed that the expression of type III collagen, matrix metalloproteinase-2, and decorin mRNAs is more than two-fold higher when HDF come into contact with a lower dose ASESM proteins immobilized on PMBN surface. A particle-exclusion assay with fixed erythrocytes has visualized secreted water-binding molecules around the cells. Thus, HDF seems to possess an ECM environment on the newly designed PMBN-ASESM surface, and future applications of the ASESM-PMBN system for biomedical use should be of great interest.

Layer-by-layer assembly and characterization of multilayers of a manganese porphyrin linked poly(4-vinylpyridinium) derivative and poly(styrenesulfonic acid-o-maleic) acid.

Multilayers of manganese(III) porphyrin-linked poly(vinylpyridinium) (MnTMPyP-PVPMe) polyelectrolyte and poly(styrenesulfonic acid-o-maleic) acid (PSS) have been assembled on gold, quartz, and indium tin oxide surfaces by a layer-by-layer (LBL) technique. The assembly process was monitored by measuring their absorption spectra and frequency change after each assembly, both of which confirmed the formation of three-dimensional MnTMPyP-PVPMe/PSS multilayers. The Soret absorption band of porphyrin red shifted about 8 nm in the multilayer compared with that in the dilute aqueous solution. The average mass changes for each assembly of MnTMPyP-PVPMe and PSS were estimated to be about 2.9 and 0.25 µg/cm(2), respectively. X-ray photoelectron spectra revealed that the as-prepared multilayers were composed of S 2p, C 1s, N 1s, O 1s, and Mn 2p, corresponding to polymers of MnTMPyP-PVPMe and PSS. A rough surface was observed after the assembly of MnTMPyP-PVPMe on the gold surface, but it became smoother when the PSS layer was adsorbed. The significant difference in the mass change and film morphology after the assembly of MnTMPyP-PVPMe compared to those after the assembly of PSS was ascribed to the reason that the MnTMPyP-PVPMe polyelectrolyte contained large metalloporphyrin macrocycles, which were axially coordinated to the pyridyl substituents of the PVP polymeric backbones. The cyclic voltammograms revealed two couples of redox waves in the phosphate electrolyte solution at pH 11, which corresponded to the electron-transfer processes of Mn(II)/Mn(III) and Mn(III)/Mn(IV) of polymeric manganese porphyrin MnTMPyP-PVPMe. The charge-transfer process was also investigated. Finally, the present MnTMPyP-PVPMe/PSS multilayers were used as a heterogeneous catalyst for the decoloration of an azo dye.

The 6th Asia cancer forum: what should we do to place cancer on the global health agenda? Sharing information leads to human security.

This forum discussed issues relating to the inclusion of cancer on the global health agenda, with the ultimate aim of achieving human security for all people. The forum discussed what methods are available to the cancer community in attempts to create a common data system for the rapidly growing Asian region. Discussions also focused on the preparations that can be made to consider and respond to the obstacles to the creation of an Asia-wide data and information network. It was also noted that in order to create a cancer information network, support would need to be provided to low- and middle-income countries and efforts made to ensure that data are comparable.

The 7th Asia Cancer Forum: from the perspective of human security, how can we collaborate as Asians in order to place cancer on the global health agenda? How can we fill in the gaps that exist among us?

This forum has continued to discuss the inclusion of cancer on the global health agenda, and specifically the Millennium Development Goals. The seventh forum presented an overview of activities to date, supplemented by reports from Korea, local governments in Japan and representatives from the pharmaceutical industry. Discussion focused on how to engage in measures to tackle cancer prevention and achieve early detection and effective treatment, using limited resources. It was recognized that with non-communicable diseases gaining increasing attention in international dialogue, it is now of the utmost importance to share an accurate recognition of cancer research and treatment throughout Asia and the wider world. Participants concurred that cancer issues are decoupled from the development aid agenda and that cooperation should be advanced on the basis of international cooperation without recourse to governmental development aid.

Langmuir-Blodgett films of pyridyldithio-modified multiwalled carbon nanotubes as a support to immobilize hydrogenase.

Pyridylthio-modified multiwalled carbon nanotubes (pythio-MWNTs) have been prepared by a reaction of the oxidized MWNTs with S-(2-aminoethylthio)-2-thiopyridine hydrochloride. The obtained pythio-MWNTs nanocomposites formed stable floating monolayers at the air-water interface, which were transferred onto substrate surfaces by the Langmuir-Blodgett (LB) method. Compositions and morphologies of the LB films were characterized by absorption, Raman, X-ray photoelectron spectra as well as by scan electron microscopy and atomic force microscopy. These pythio-MWNTs LB films were then used as a support to immobilize hydrogenase (H(2)ase) to form bionanocomposite of pythio-MWNTs-H(2)ase. Cyclic voltammograms for indium tin oxide electrode covered with the pythio-MWNTs-H(2)ase films were investigated in both Ar and H(2) saturated 0.05 M KCl electrolyte solutions at pH from 4.0 to 9.0. A reversible redox couple of [4Fe-4S](2+/1+) clusters of H(2)ase was recorded when the pH value was 6.0 and 9.0, with reduction and oxidation potentials appearing at about -0.70 and -0.35 V vs Ag/AgCl, respectively. It was revealed that the H(2)ase was of high catalytic activity and strong stability in the LB films of pythio-MWNTs-H(2)ase. Hence, we suggested that the present bionanocomposites could be used as heterogeneous biocatalyst to catalyze reversible reaction between protons and H(2), resulting in potential applications in biohydrogen evolution and H(2) biofuel cells.