CT medical image segmentation algorithm based on deep learning technology.

For the problems of blurred edges, uneven background distribution, and many noise interferences in medical image segmentation, we proposed a medical image segmentation algorithm based on deep neural network technology, which adopts a similar U-Net backbone structure and includes two parts: encoding and decoding. Firstly, the images are passed through the encoder path with residual and convolutional structures for image feature information extraction. We added the attention mechanism module to the network jump connection to address the problems of redundant network channel dimensions and low spatial perception of complex lesions. Finally, the medical image segmentation results are obtained using the decoder path with residual and convolutional structures. To verify the validity of the model in this paper, we conducted the corresponding comparative experimental analysis, and the experimental results show that the DICE and IOU of the proposed model are 0.7826, 0.9683, 0.8904, 0.8069, and 0.9462, 0.9537 for DRIVE, ISIC2018 and COVID-19 CT datasets, respectively. The segmentation accuracy is effectively improved for medical images with complex shapes and adhesions between lesions and normal tissues.

SOC Estimation of Lithium-Ion Battery for Electric Vehicle Based on Deep Multilayer Perceptron.

The state of charge (SOC) is one of the main indexes of the lithium-ion battery, which affects the practice range of new energy vehicles and the safety of the battery. Nevertheless, the value of SOC cannot be measured directly. At present, the algorithm for estimating the state of charge is not very satisfactory. The multilayer perceptron algorithm designed during this paper encompasses a sensible impact on state estimation. During this paper, the multilayer network is designed to estimate the charged state of lithium batteries from the three-layer artificial neural network to the eleven-layer artificial neural network. After preprocessing the dataset and comparing several activation functions, the ten-layer fully connected neural network is the most efficient to estimate the SOC. In order to prevent over-fitting of the multilayer perceptron algorithm, the two techniques of the BatchNormalization layer and Dropout layer work together to inhibit over-fitting. At the same time, the accuracy of extended Kalman filter, long and short memory network, and recurrent neural network are compared. The multilayer perceptron network designed during this paper has the highest accuracy. Finally, in the open dataset, both the training and test errors achieve good results. The algorithm developed in this paper has made some progress in SOC estimation.

Discontinued drugs in 2010: cardiovascular drugs.

This perspective is a paper discussing drugs dropped from clinical development in the previous years. Specifically, this paper focuses on 16 cardiovascular drugs discontinued in 2010 after reaching Phase I - III clinical trials. Information for this perspective is mainly derived from a search of Pharmaprojects.

Discontinued drugs in 2008: cardiovascular drugs.

This perspective is part of an annual series of papers discussing drugs dropped from clinical development in the previous year. Specifically, this paper focuses on the 16 cardiovascular drugs discontinued in 2008. Information for this perspective was derived from a search of the Pharmaprojects database for drugs discontinued after reaching Phase I-III clinical trials.

Discontinued drugs in 2007: cardiovascular drugs.

This perspective is part of an annual series of papers discussing drugs dropped from clinical development in the previous year. Specifically, this paper focuses on the 16 cardiovascular drugs discontinued in 2007. Information for this perspective was derived from a search of the Pharmaprojects database for drugs discontinued after reaching Phase I - III clinical trials.

Effect of rhIL-2 in inhalant form on experimental respiratory tract infection with Klebsiella pneumoniae in mice.

AIM: To determine the protective effect of recombinant human interleukin-2 (rhIL-2) in inhalant form on experimental respiratory tract infection with Klebsiella pneumoniae in mice. METHODS: Mice were infected with the method of nasal intubation drip. During infection, mice were given rhIL-2 by sc injection and the method of nasal intubation drip. There were normal group, vehicle group, model group, rhIL-2 groups and gentamicin group. In the end, the pathological changes in the lung were observed. The survival time and the mortality within a week of each group were recorded. The total protein content, the albumin content, the activity of alkaline phosphatase and the activity of lactic dehydrogenase of broncho-alveolar lavage fluid (BALF) were dertermined and compared. RESULTS: Symptoms of Klebsiella pneumoniae were remarkably relieved because of rhIL-2 administration. The total protein content, the albumin content, the activity of alkaline phosphatase and the activity of lactic dehydrogenase of BALF were less than those in the vehicle group and the model group. CONCLUSION: Inhalation of rhIL-2 can alleviate the pathological changes in the lung after infection. At the same dose, it could be seen that the effect of rhIL-2 in inhalant form was better than that of the injection.

Deposition of insulin powders for inhalation in vitro and pharmacodynamic evaluation of absorption promoters in rats.

AIM: To prepare insulin powder for inhalation by spray-drying technology, determine the deposition of the insulin powder formulation in vitro and preliminarily investigate hypoglycemic response of the dry powder with/without absorption promoters. METHODS: The depositions of the insulin powder for inhalation were determined by the China Pharmacopoeia 2000 version addenda XH and hypoglycemic effects were evaluated by testing serum glucose with glucose oxidase-peroxidase (GOD-PAP) method. RESULTS: The depositions of the spray-dried insulin powder for inhalation were more than 40% under various humidity and their changes were not significant when air flow was no less than 18 L x min(-1). The coadministration of insulin with 8 mmol x L(-1)/dose sodium taurocholate [PA = 59.91%, Cnadir = (33 +/- 6) %] and 10 mmol x L(-1)/dose sodium deoxycholate [PA = 47.46% , Cnadir = (32 +/- 7)%] induced a significantly greater decline in blood glucose levels, while coadministration with 1% sodium caprylate, 1% sodium dodecyl sulfate, 250 microg/dose lecithin, 10 mmol x L(-1)/dose EDTA appeared to have no significant effect (P > 0.05). CONCLUSION: Insulin powder for inhalation was relatively stable under various humidity conditions and different flow current. The use of 8 mmol x L(-1)/dose sodium taurocholate and 10 mmol x L(-1)/dose sodium deoxycholate could be able to potentially improve the bioavailability of insulin by pulmonary route.

[Determination of gamma-aminobutyric acid and glutamate in human gastric mucosa by high performance liquid chromatography].

A method for analysis of gamma-aminobutyric acid (GABA) and glutamate (Glu) in human gastric mucosa by high performance liquid chromatography (HPLC) was developed. A gradient elution was used for the separation and quantification of GABA and Glu after pre-column derivatization with phenylisothiocyanate (PITC). The column was Pico x Tag for free amino acids. GABA and Glu were determined with UV detector at 254 nm. Good linearities were observed within the ranges from 0.125 to 6.25 micromol/L for GABA and from 0.025 to 2.5 mmol/L for Glu. The average recoveries were 95.4% for GABA and 93.5% for Glu. The intra- and inter-precision values were within 3.56% and 7.47% for GABA, and 1.12% and 5.98% for Glu, respectively. The method is sensitive, specific and accurate. It can be used in the determination of GABA and Glu in human gastric mucosa tissue. The concentrations of GABA and Glu in cancer tissues are significantly higher than those in normal tissue.