A one-dimensional convolutional neural network based deep learning for high accuracy classification of transformation stages in esophageal squamous cell carcinoma tissue using micro-FTIR

Yang, Haijun; Li, Xian Chang; Zhang, Shiding; Li, Yuan; Zhu, Zunwei; Shen, Jingwei; Dai, Ning; Zhou, Fuyou

doi:10.1016/j.saa.2022.122210

Cited by 12 publications

(6 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…FTIR spectroscopy is a highly informative technique that allows us to study the interaction between cells, drugs, and polymers on a molecular level. In recent years, there have been numerous studies using machine learning techniques to analyze infrared spectra, which can predict the progression of cancer or inflammation [ 60 , 61 , 62 ]. The methodology for studying cell–drug interactions and delivery systems using Fourier transform infrared spectroscopy (FTIR) was validated using control methods such as confocal microscopy, fluorescence spectroscopy, and MTT assays.…”

Section: Resultsmentioning

confidence: 99%

Disulfide Cross-Linked Polymeric Redox-Responsive Nanocarrier Based on Heparin, Chitosan and Lipoic Acid Improved Drug Accumulation, Increased Cytotoxicity and Selectivity to Leukemia Cells by Tumor Targeting via “Aikido” Principle

Zlotnikov,

Ezhov,

Dobryakova

et al. 2024

Gels

View full text Add to dashboard Cite

We have developed a micellar formulation of anticancer drugs based on chitosan and heparin grafted with lipoic and oleic acids that can release the cytotoxic cargo (doxorubicin) in response to external stimuli, such as increased glutathione concentration—a hallmark of cancer. Natural polysaccharides (heparin and chitosan) provide the pH sensitivity of the nanocarrier: the release of doxorubicin (Dox) is enhanced in a slightly acidic environment (tumor microenvironment). Fatty acid residues are necessary for the formation of nanoparticles (micelles) and solubilization of cytostatics in a hydrophobic core. Lipoic acid residues provide the formation of a labile S-S cross-linking between polymer chains (the first variant) or covalently attached doxorubicin molecules through glutathione-sensitive S-S bridges (the second variant)—both determine Redox sensitivity of the anticancer drugs carriers stable in blood circulation and disintegrate after intracellular uptake in the tumor cells. The release of doxorubicin from micelles occurs slowly (20%/6 h) in an environment with a pH of 7.4 and the absence of glutathione, while in a slightly acidic environment and in the presence of 10 mM glutathione, the rate increases up to 6 times, with an increase in the effective concentration up to 5 times after 7 h. The permeability of doxorubicin in micellar formulations (covalent S-S cross-linked and not) into Raji, K562, and A875 cancer cells was studied using FTIR, fluorescence spectroscopy and confocal laser scanning microscopy (CLSM). We have shown dramatically improved accumulation, decreased efflux, and increased cytotoxicity compared to doxorubicin control with three tumor cell lines: Raji, K562, and A875. At the same time, cytotoxicity and permeability for non-tumor cells (HEK293T) are significantly lower, increasing the selectivity index against tumor cells by several times.

show abstract

Section: Resultsmentioning

confidence: 99%

Disulfide Cross-Linked Polymeric Redox-Responsive Nanocarrier Based on Heparin, Chitosan and Lipoic Acid Improved Drug Accumulation, Increased Cytotoxicity and Selectivity to Leukemia Cells by Tumor Targeting via “Aikido” Principle

Zlotnikov,

Ezhov,

Dobryakova

et al. 2024

Gels

View full text Add to dashboard Cite

show abstract

“…AlexNet is a feed-forward neural network established on Convolutional Neural Networks that mimics biology. It has convolutional computation and a deep structure (29) and is widely employed in image and natural language processing, among other areas. The network structure of AlexNet has convolutional primarily, pooling, and fully connected layers.…”

Section: Classification Model 231 Alexnetmentioning

confidence: 99%

“…Then based on MP-NN, the three-mode model MPI-RF, which uses random forest to integrate SMFs, CEA and image features, is superior to clinical diagnosis in the classification of pulmonary nodules (28). Yang et al (29) classified the tissue transformation stages of esophageal squamous cell carcinoma with high accuracy based on a one-dimensional convolutional neural network (1-CNN) combined with micro-FTIR. Chen et al (30) used an improved multi-scale fusion convolutional neural network on near-infrared spectral data to classify cumin and cumin with an accuracy of 100%.…”

Section: Introductionmentioning

confidence: 99%

Rapid screening for autoimmune diseases using Fourier transform infrared spectroscopy and deep learning algorithms

Wu,

Shuai,

Chen

et al. 2023

Front. Immunol.

View full text Add to dashboard Cite

IntroduceAnkylosing spondylitis (AS), rheumatoid arthritis (RA), and osteoarthritis (OA) are three rheumatic immune diseases with many common characteristics. If left untreated, they can lead to joint destruction and functional limitation, and in severe cases, they can cause lifelong disability and even death. Studies have shown that early diagnosis and treatment are key to improving patient outcomes. Therefore, a rapid and accurate method for rapid diagnosis of diseases has been established, which is of great clinical significance for realizing early diagnosis of diseases and improving patient prognosis.MethodsThis study was based on Fourier transform infrared spectroscopy (FTIR) combined with a deep learning model to achieve non-invasive, rapid, and accurate differentiation of AS, RA, OA, and healthy control group. In the experiment, 320 serum samples were collected, 80 in each group. AlexNet, ResNet, MSCNN, and MSResNet diagnostic models were established by using a machine learning algorithm.ResultThe range of spectral wave number measured by four sets of Fourier transform infrared spectroscopy is 700-4000 cm-1. Serum spectral characteristic peaks were mainly at 1641 cm-1(amide I), 1542 cm-1(amide II), 3280 cm-1(amide A), 1420 cm-1(proline and tryptophan), 1245 cm-1(amide III), 1078 cm-1(carbohydrate region). And 2940 cm-1 (mainly fatty acids and cholesterol). At the same time, AlexNet, ResNet, MSCNN, and MSResNet diagnostic models are established by using machine learning algorithms. The multi-scale MSResNet classification model combined with residual blocks can use convolution modules of different scales to extract different scale features and use resblocks to solve the problem of network degradation, reduce the interference of spectral measurement noise, and enhance the generalization ability of the network model. By comparing the experimental results of the other three models AlexNet, ResNet, and MSCNN, it is found that the MSResNet model has the best diagnostic performance and the accuracy rate is 0.87.ConclusionThe results prove the feasibility of serum Fourier transform infrared spectroscopy combined with a deep learning algorithm to distinguish AS, RA, OA, and healthy control group, which can be used as an effective auxiliary diagnostic method for these rheumatic immune diseases.

show abstract

Section: Introductionmentioning

confidence: 99%

“…Machine learning is increasingly being used in medicine to assist clinicians in diagnosis and treatment decisions, 51 as well as for predicting the development of diseases, including autoimmune. [52][53][54][55][56] Within the framework of the LatDiane project, the largest collection of T1D in the Baltics has been created. 5 Based on this database, a pilot study was conducted to see if the FTIR method based on the analysis of metabolite ngerprints in PBMC, in combination with machine learning algorithms: (1) detect autoimmunity; (2) whether it is possible to identify a specic autoimmune disease, in our study it was T1D.…”

mentioning

confidence: 99%

Prediction of type 1 diabetes with machine learning algorithms based on FTIR spectral data in peripheral blood mononuclear cells

Rostoka,

Shvirksts,

Salna

et al. 2023

Anal. Methods

View full text Add to dashboard Cite

show abstract

A one-dimensional convolutional neural network based deep learning for high accuracy classification of transformation stages in esophageal squamous cell carcinoma tissue using micro-FTIR

Cited by 12 publications

References 40 publications

Disulfide Cross-Linked Polymeric Redox-Responsive Nanocarrier Based on Heparin, Chitosan and Lipoic Acid Improved Drug Accumulation, Increased Cytotoxicity and Selectivity to Leukemia Cells by Tumor Targeting via “Aikido” Principle

Disulfide Cross-Linked Polymeric Redox-Responsive Nanocarrier Based on Heparin, Chitosan and Lipoic Acid Improved Drug Accumulation, Increased Cytotoxicity and Selectivity to Leukemia Cells by Tumor Targeting via “Aikido” Principle

Rapid screening for autoimmune diseases using Fourier transform infrared spectroscopy and deep learning algorithms

Prediction of type 1 diabetes with machine learning algorithms based on FTIR spectral data in peripheral blood mononuclear cells

Contact Info

Product

Resources

About