Deep Learning Prediction of TERT Promoter Mutation Status in Thyroid Cancer Using Histologic Images.

Background and objectives: Telomerase reverse transcriptase (TERT) promoter mutation, found in a subset of patients with thyroid cancer, is strongly associated with aggressive biologic behavior. Predicting TERT promoter mutation is thus necessary for the prognostic stratification of thyroid cancer patients. Materials and Methods: In this study, we evaluate TERT promoter mutation status in thyroid cancer through the deep learning approach using histologic images. Our analysis included 13 consecutive surgically resected thyroid cancers with TERT promoter mutations (either C228T or C250T) and 12 randomly selected surgically resected thyroid cancers with a wild-type TERT promoter. Our deep learning model was created using a two-step cascade approach. First, tumor areas were identified using convolutional neural networks (CNNs), and then TERT promoter mutations within tumor areas were predicted using the CNN-recurrent neural network (CRNN) model. Results: Using the hue-saturation-value (HSV)-strong color transformation scheme, the overall experiment results show 99.9% sensitivity and 60% specificity (improvements of approximately 25% and 37%, respectively, compared to image normalization as a baseline model) in predicting TERT mutations. Conclusions: Highly sensitive screening for TERT promoter mutations is possible using histologic image analysis based on deep learning. This approach will help improve the classification of thyroid cancer patients according to the biologic behavior of tumors.

Toward Interpretable Cell Image Representation and Abnormality Scoring for Cervical Cancer Screening Using Pap Smears.

Screening is critical for prevention and early detection of cervical cancer but it is time-consuming and laborious. Supervised deep convolutional neural networks have been developed to automate pap smear screening and the results are promising. However, the interest in using only normal samples to train deep neural networks has increased owing to the class imbalance problems and high-labeling costs that are both prevalent in healthcare. In this study, we introduce a method to learn explainable deep cervical cell representations for pap smear cytology images based on one-class classification using variational autoencoders. Findings demonstrate that a score can be calculated for cell abnormality without training models with abnormal samples, and we localize abnormality to interpret our results with a novel metric based on absolute difference in cross-entropy in agglomerative clustering. The best model that discriminates squamous cell carcinoma (SCC) from normals gives 0.908±0.003 area under operating characteristic curve (AUC) and one that discriminates high-grade epithelial lesion (HSIL) 0.920±0.002 AUC. Compared to other clustering methods, our method enhances the V-measure and yields higher homogeneity scores, which more effectively isolate different abnormality regions, aiding in the interpretation of our results. Evaluation using an external dataset shows that our model can discriminate abnormality without the need for additional training of deep models.

Lipocalin-2 Is a chemokine inducer in the central nervous system: role of chemokine ligand 10 (CXCL10) in lipocalin-2-induced cell migration.

The secreted protein lipocalin-2 (LCN2) has been implicated in diverse cellular processes, including cell morphology and migration. Little is known, however, about the role of LCN2 in the CNS. Here, we show that LCN2 promotes cell migration through up-regulation of chemokines in brain. Studies using cultured glial cells, microvascular endothelial cells, and neuronal cells suggest that LCN2 may act as a chemokine inducer on the multiple cell types in the CNS. In particular, up-regulation of CXCL10 by JAK2/STAT3 and IKK/NF-κB pathways in astrocytes played a pivotal role in LCN2-induced cell migration. The cell migration-promoting activity of LCN2 in the CNS was verified in vivo using mouse models. The expression of LCN2 was notably increased in brain following LPS injection or focal injury. Mice lacking LCN2 showed the impaired migration of astrocytes to injury sites with a reduced CXCL10 expression in the neuroinflammation or injury models. Thus, the LCN2 proteins, secreted under inflammatory conditions, may amplify neuroinflammation by inducing CNS cells to secrete chemokines such as CXCL10, which recruit additional inflammatory cells.

Amyloid neurotoxicity is attenuated by metallothionein: dual mechanisms at work.

Alzheimer's disease (AD) is a neurodegenerative disorder characterized by a progressive loss of memory and cognition. One of the hallmarks of AD is the accumulation of beta-amyloid (Aß). Although endoplasmic reticulum stress, mitochondrial dysfunction, and oxidative stress have been implicated in Aß toxicity, the molecular mechanism(s) of Aß-induced neurotoxicity are not fully understood. In this study, we present evidence that the glia-derived stress protein metallothionein (MT) attenuates Aß-induced neurotoxicity by unique mechanisms. MT expression was increased in brain astrocytes of a NSE-APPsw transgenic mouse model of AD. Astrocyte-derived MT protected N2a neuroblastoma cells and primary cortical neurons against Aß toxicity with concurrent reduction of reactive oxygen species levels. MT reversed Aß-induced down-regulation of Bcl-2 and survival signaling in neuroblastoma cells. Moreover, MT inhibited Aß-induced proinflammatory cytokine production from microglia. The neurotoxicity of Aß-stimulated microglia was significantly attenuated by MT-I. The results indicate that MT released from reactive astrocytes may antagonize Aß neurotoxicity by direct inhibition of Aß neurotoxicity and indirect suppression of neurotoxic microglial activation. These findings broaden the understanding of neurotoxic mechanisms of Aß and the crosstalk between Aß and MT in AD.

Diagnostic Accuracy of Loop-Mediated Isothermal Amplification Assay for Group B Streptococcus Detection in Recto-Vaginal Swab: Comparison with Polymerase Chain Reaction Test and Conventional Culture.

A rapid method for obtaining group B streptococcus (GBS) screening results has been required in the obstetric field. We aimed to determine the diagnostic performance of the Loop-Mediated Isothermal Amplification (LAMP) assay is acceptable compared to the existing polymerase chain reaction (PCR) assay. The study involved 527 pregnant women aged 19 to 44 years. Rectovaginal swabs were collected between 35 and 37 weeks of gestation or prior to impending preterm births or term labor without GBS screening. We presented the diagnostic performance of the LAMP assay with a 95% confidence interval (CI) compared to the PCR and microbiological culture. In total, 115 (21.8%), 115 (21.8%) and 23 (4.4%) patients showed positive results using the LAMP, PCR assay and microbiological culture method, respectively. The LAMP assay showed 100% sensitivity (95% CI, 96.8-100.0), 100% specificity (95% CI, 99.1-100.0) and 100% diagnostic accuracy (95% CI, 99.3-100.0) with the reference being the PCR assay. Meanwhile, the LAMP assay showed 87.0% sensitivity (95% CI, 71.0-100.0), 81.2% specificity (95% CI, 77.6-84.7), and 81.4% diagnostic accuracy (95% CI, 78.0-84.8) with the microbiological culture as a reference. This study presented the LAMP assay as an acceptable method for GBS screening with a similar performance to the existing PCR method.

Updates on Group B Streptococcus Infection in the Field of Obstetrics and Gynecology.

Group B Streptococcus (GBS, Streptococcus agalactiae) is a Gram-positive bacterium that is commonly found in the gastrointestinal and urogenital tracts. However, its colonization during pregnancy is an important cause of maternal and neonatal morbidity and mortality worldwide. Herein, we specifically looked at GBS in relation to the field of Obstetrics (OB) along with the field of Gynecology (GY). In this review, based on the clinical significance of GBS in the field of OBGY, topics of how GBS is being detected, treated, and should be prevented are addressed.

Microglia-inhibiting activity of Parkinson's disease drug amantadine.

Amantadine is currently used as an antiviral and an antiparkinsonian drug. Although the drug is known to bind a viral proton channel protein, the mechanism of action in Parkinson's disease (PD) remains to be determined. This study investigated whether the drug has an inhibitory effect on microglial activation and neuroinflammation, which have been implicated in the progression of neurodegenerative processes. Using cultured microglial cells, it was demonstrated that the drug inhibited inflammatory activation of microglia and a signaling pathway that governs the microglial activation. The drug reduced the expression and production of proinflammatory mediators in bacterial lipopolysaccharide-stimulated microglia cells. The microglia-inhibiting activity of amantadine was also demonstrated in a microglia/neuron coculture and animal models of neuroinflammation and Parkinson's disease. Collectively, our results suggest that amantadine may act on microglia in the central nervous system to inhibit their inflammatory activation, thereby attenuating neuroinflammation. These results provide a molecular basis of the glia-targeted mechanism of action for amantadine.

Novel microencapsulation of potential drugs with low molecular weight and high hydrophilicity: hydrogen peroxide as a candidate compound.

Microencapsulation of drugs into solid biodegradable polymeric microspheres via solvent evaporation technique remains challenging especially with those having low molecular weight and high hydrophilicity nature. This paper presents an efficient encapsulation protocol for this group of drugs, demonstrated using hydrogen peroxide as a model compound that is encapsulated into poly(lactic-co-glycolic acid) microspheres. Hydrogen peroxide can be employed as antiseptic agent or its decomposed form into oxygen can be useful in various pharmaceutical applications. The new encapsulation technique was developed based on the modification of conventional double emulsion and solvent evaporation protocol with a backward concentration gradient of hydrogen peroxide. This was achieved by adding and controlling the concentration of hydrogen peroxide at the continuous phase during the solidification stage of the microspheres. Parameters involved in the production and the formulation aspect were optimized to achieve the best protocol having controlled efficiency of encapsulation that is simple, safe, practical, and economical. Evaluation on the encapsulation efficiency and the release profile has been made indirectly by monitoring the dissolved oxygen level of the solution where the microspheres were incubated. Morphology of the microspheres was investigated using scanning electron microscopy. This proposed method has successfully used to prepare batches of microspheres having different encapsulation efficiencies and its potential applications have been demonstrated accordingly.

Detection of apoptosis in a rat model of focal cerebral ischemia using a homing peptide selected from in vivo phage display.

Focal cerebral ischemia, known as stroke, is caused by a sudden interruption in the blood supply to the brain. We attempted to identify peptides that can home to ischemic stroke tissue and detect the apoptosis of cells. A phage library displaying random peptides was screened for homing peptides to ischemic stroke tissue in a rat transient middle cerebral artery (MCA) occlusion model. After three rounds of in vivo screening, a phage clone displaying the most frequently occurring CLEVSRKNC sequence was selected. The CLEVSRKNC-phage preferentially homed to ischemic stroke tissue after intravenous administration into the MCA occlusion rats. The fluorescein-labeled synthetic CLEVSRKNC peptide, but not a scrambled control peptide, homed to ischemic stroke tissue with a lack of homing to non-ischemic brain tissue. The CLEVSRKNC peptide co-localized with a portion of neuronal cells, rather than with astrocytes, undergoing apoptosis at the penumbra region of stroke lesions. In autoradiographic studies, the uptake of the (131)I-labeled CLEVSRKNC peptide into an ischemic lesion increased at the first day and peaked at the third day after the injury. These results demonstrate that the CLEVSRKNC peptide can home to ischemic stroke tissue, while detecting apoptotic neuronal cells, and suggest it has applications as a targeting moiety for molecular imaging and selective drug delivery to stroke tissue.