DNA nanostructures for exploring cell-cell communication.

The process of coordinating between the same or multiple types of cells to jointly execute various instructions in a controlled and carefully regulated environment is a very appealing field. In order to provide clearer insight into the role of cell-cell interactions and the cellular communication of this process in their local communities, several interdisciplinary approaches have been employed to enhance the core understanding of this phenomenon. DNA nanostructures have emerged in recent years as one of the most promising tools in exploring cell-cell communication and interactions due to their programmability and addressability. Herein, this review is dedicated to offering a new perspective on using DNA nanostructures to explore the progress of cell-cell communication. After briefly outlining the anchoring strategy of DNA nanostructures on cell membranes and the subsequent dynamic regulation of DNA nanostructures, this paper highlights the significant contribution of DNA nanostructures in monitoring cell-cell communication and regulating its interactions. Finally, we provide a quick overview of the current challenges and potential directions for the application of DNA nanostructures in cellular communication and interactions.

Building a predictive model to assist in the diagnosis of cervical cancer.

Aim: Cervical cancer is still one of the most common gynecologic cancers in the world. Since cervical cancer is a potentially preventive cancer, earlier detection is the most effective technique for decreasing the worldwide incidence of the illness. Materials and methods: This research presents a novel ensemble technique for predicting cervical cancer risk. Specifically, the authors introduce a voting classifier that aggregates prediction probabilities from multiple machine-learning models: logistic regression, K-nearest neighbor, decision tree, XGBoost and multilayer perceptron. Results: The average accuracy, precision, recall and f1-score of the voting classifier were 96.6, 97.4, 95.9 and 96.6, respectively. Furthermore, the voting algorithm gains average high values for all evaluation metrics (accuracy, precision, recall and f1-score). The f1-score of the algorithm is 96%, which demonstrates the robustness of the model. Conclusion: The findings suggest that the probability of having cervical cancer can be accurately predicted utilizing the voting technique.

Magnetic-Nanowaxberry-Based Simultaneous Detection of Exosome and Exosomal Proteins for the Intelligent Diagnosis of Cancer.

Exosome concentration and exosomal proteins are regarded as promising cancer biomarkers. Herein, a waxberry-like magnetic bead (magnetic-nanowaxberry) which has huge surface area and strong affinity was synthesized to couple with aptamer for exosome capture and recovery. Subsequently, we developed a fluorescent assay for the sensitive, accurate, and simultaneous quantification of exosome and cancer-related exosomal proteins [epidermal growth factor receptor (EGFR) and epithelial cell adhesion molecule (EpCAM)] by using triple-colored probes to recognize EGFR and EpCAM or spontaneously anchor to the lipid bilayer. In this design, the interference of soluble proteins can be avoided due to the dual recognition strategy. Moreover, the lipid-based quantification of exosome concentration can improve the accuracy. Besides, the simultaneous detection mode can save samples and simplify the operation steps. Consequently, the assay shows high sensitivity (the limits of detection are down to 0.96 pg/mL for EGFR, 0.19 pg/mL for EpCAM, and 2.4 × 104 particles/µL for exosome), high specificity, and satisfactory accuracy. More importantly, this technique is successfully used to analyze exosomes in plasma to distinguish cancer patients from healthy individuals. To improve the diagnostic efficacy, the deep learning was used to exploit the potential pattern hidden in data obtained by the proposed method. Also, the accuracy for the intelligent diagnosis of cancer can achieve 96.0%. This study provides a new avenue for developing new biosensors for exosome analysis and intelligent disease diagnosis.

A Holistic Review of the State-of-the-Art Microfluidics for Exosome Separation: An Overview of the Current Status, Existing Obstacles, and Future Outlook.

Exosomes, a class of small extracellular vesicles (30-150 nm), are secreted by almost all types of cells into virtually all body fluids. These small vesicles are attracting increasing research attention owing to their potential for disease diagnosis and therapy. However, their inherent heterogeneity and the complexity of bio-fluids pose significant challenges for their isolation. Even the "gold standard," differential centrifugation, suffers from poor yields and is time-consuming. In this context, recent developments in microfluidic technologies have provided an ideal system for exosome extraction and these devices exhibit some fascinating properties such as high speeds, good portability, and low sample volumes. In this review, the focus is on the state-of-the-art microfluidic technologies for exosome isolation and highlight potential directions for future research and development by analyzing the challenges faced by the current strategies.

In situ detection of plasma exosomal microRNA for lung cancer diagnosis using duplex-specific nuclease and MoS2 nanosheets.

MicroRNAs (miRNAs) encapsulated in tumor-derived exosomes are becoming ideal biomarkers for the early diagnosis and prognosis of lung cancer. However, the accuracy and sensitivity are often hampered by the extraction process of exosomal miRNA using traditional methods. Herein, this study developed a fluorogenic quantitative detection method for exosomal miRNA using the fluorescence quenching properties of molybdenum disulfide (MoS2) nanosheets and the enzyme-assisted signal amplification properties of duplex-specific nuclease (DSN). First, a fluorescently-labeled nucleic acid probe was used to hybridize the target miRNA to form a DNA/RNA hybrid structure. Under the action of the DSN, the DNA single strand in the DNA/RNA hybrid strand was selectively digested into smaller oligonucleotide fragments. At the same time, the released miRNA target triggers the next reaction cycle, so as to achieve signal amplification. Then, MoS2 was used to selectively quench the fluorescence of the undigested probe leaving the fluorescent signal of the fluorescently-labeled probe fragments. The fluorometric signals for miRNA-21 had a maximum excitation/emission wavelength of 488/518 nm. Most importantly, the biosensor was then applied for the accurate quantitative detection of miRNA-21 in exosome lysates extracted from human plasma and this method was able to successfully distinguish lung cancer patients from healthy people. This biosensor provides a simple, rapid, and a highly specific quantitative method for exosomal miRNA and has promising potential to be used in the early diagnosis of lung cancer.

A simple "signal off-on" fluorescence nanoplatform for the label-free quantification of exosome-derived microRNA-21 in lung cancer plasma.

A simple nanoplatform based on molybdenum disulfide (MoS2) nanosheets, a fluorescence quencher (signal off), and a hybridization chain reaction (HCR) signal amplification (signal on) used for the enzyme-free, label-free, and low-background signal quantification of microRNA-21 in plasma exosome is reported. According to the sequence of microRNA-21, carboxy-fluorescein (FAM)-labeled hybridization probe 1 (FAM-H1) and hybridization probes 2 (FAM-H2) were designed with excitation maxima at 488 nm and emission maxima at 518 nm. MoS2 nanosheets could adsorb FAM-H1 and FAM-H2 and quenched their fluorescence signals to reduce the background signal. However, HCR was triggered when microRNA-21 was present. Consequently, HCR products containing a large number of FAM fluorophores can emit a strong fluorescence at 518 nm and could realize the detection of microRNA-21 as low as 6 pmol/L and had a wide linear relation of 0.01-25 nmol/L. This assay has the ability of single-base mismatch recognition and could identify microRNA-21 with high specificity. Most importantly, this approach was successfully applied to the detection of plasma exosomal microRNA-21 in patients with lung cancer, and it is proposed that other targets can also be detected by changing the FAM-H1 and FAM-H2 corresponding to the target sequence. Thus, a novel, hands-on strategy for liquid biopsy was proposed and has a potential application value in the early diagnosis of lung cancer.

Klebsiella pneumoniae: an increasing threat to public health.

OBJECTIVES: This review fills the paucity of information on K. pneumoniae as a nosocomial pathogen by providing pooled data on epidemiological risk factors, resistant trends and profiles and resistant and virulent genes of this organism in Asia. METHODS: Exhaustive search was conducted using PubMed, Web of Science, and Google scholar for most studies addressing the prevalence, risk factors, drug resistant-mediated genes and/or virulent factors of K. pneumoniae in Asia. Data extracted for meta-analysis were analyzed using comprehensive meta-analysis version 3. Trends data for the isolation rate and resistance rates were entered into Excel spread sheet and the results were presented in graphs. RESULTS: The prevalence rate of drug resistance in K. pneumoniae were; amikacin (40.8%) [95% CI 31.9-50.4], aztreonam (73.3%) [95% CI 59.9-83.4], ceftazidime (75.7%) [95% CI 65.4-83.6], ciprofloxacin (59.8%) [95% CI 48.6-70.1], colistin (2.9%) [95% CI 1.8-4.4], cefotaxime (79.2%) [95% CI 68.0-87.2], cefepime (72.6) [95% CI 57.7-83.8] and imipenem (65.6%) [95% CI 30.8-89.0]. TEM (39.5%) [95% CI 15.4-70.1], SHV-11 (41.8%) [95% CI 16.2-72.6] and KPC-2 (14.6%) [95% CI 6.0-31.4] were some of the resistance mediated genes observed in this study. The most virulent factors utilized by K. pneumoniae are; hypermucoviscous phenotype and mucoviscosity-related genes, genes for biosynthesis of lipopolysaccharide, iron uptake and transport genes and finally, adhesive genes. CONCLUSION: It can be concluded that, antimicrobial resistant in K. pneumoniae is a clear and present danger in Asia which needs strong surveillance to curb this menace. It is very important for public healthcare departments to monitor and report changes in antimicrobial-resistant isolates.

Bacteria-derived extracellular vesicles: endogenous roles, therapeutic potentials and their biomimetics for the treatment and prevention of sepsis.

Sepsis is one of the medical conditions with a high mortality rate and lacks specific treatment despite several years of extensive research. Bacterial extracellular vesicles (bEVs) are emerging as a focal target in the pathophysiology and treatment of sepsis. Extracellular vesicles (EVs) derived from pathogenic microorganisms carry pathogenic factors such as carbohydrates, proteins, lipids, nucleic acids, and virulence factors and are regarded as "long-range weapons" to trigger an inflammatory response. In particular, the small size of bEVs can cross the blood-brain and placental barriers that are difficult for pathogens to cross, deliver pathogenic agents to host cells, activate the host immune system, and possibly accelerate the bacterial infection process and subsequent sepsis. Over the years, research into host-derived EVs has increased, leading to breakthroughs in cancer and sepsis treatments. However, related approaches to the role and use of bacterial-derived EVs are still rare in the treatment of sepsis. Herein, this review looked at the dual nature of bEVs in sepsis by highlighting their inherent functions and emphasizing their therapeutic characteristics and potential. Various biomimetics of bEVs for the treatment and prevention of sepsis have also been reviewed. Finally, the latest progress and various obstacles in the clinical application of bEVs have been highlighted.

The effect of breakfast on childhood obesity: a systematic review and meta-analysis.

Objective: Previous cohort trials have shown that skipping breakfast increases the risk of obesity or overweight in children. However, this finding remains controversial. Through a meta-analysis, this study systematically evaluated the effect of skipping breakfast on the prevalence of obesity or overweight in children. Methods: We performed a literature search for studies published until March 19, 2023. using the Cochrane, PubMed, and Embase databases. Based on the inclusion and exclusion criteria, observational studies on the relationship between skipping breakfast and overweight/obesity in children and adolescents were analyzed. Three investigators independently screened the relevant literature, extracted the data, and assessed the risk of bias. The quality of the included studies was assessed using the Newcastle-Ottawa Scale (NOS). A random-effects model was used. The odds ratio (OR) with its 95% confidence interval (CI) was used to indicate the effect size. Results: A total of 40 retrospective studies with 323,244 children ranging in age from 2 to 20 years were included in this study. The results of this meta-analysis showed that children and adolescents who skipped breakfast had a significantly higher prevalence of obesity or overweight than those who ate breakfast (OR, 1.59; 95% CI, 1.33-1.90; P < 0.001). Skipping breakfast was positively associated with overweight in children and adolescents (OR, 1.37; 95% CI, 1.23-1.54; P < 0.001). Similarly, skipping breakfast was positively associated with obesity in children and adolescents (OR, 1.51; 95% CI, 1.30-1.76; P < 0.001). The effect was also different by sex, with girls being the most affected (OR, 1.47; 95% CI, 1.23-1.76; P < 0.001). There was also a correlation between skipping breakfast and abdominal obesity in children (OR, 0.65; 95% CI, 0.55-0.77; P < 0.001). Conclusion: This meta-analysis suggested that skipping breakfast is associated with an increased risk of overweight/obesity in children and adolescents. The findings provide support for a possible protective role of breakfast against excessive weight gain in children and adolescents. However, more rigorous study designs with validated and standardized measures of relevant variables are needed.

Screening of the differentially expressed proteins in malignant transformation of BEAS-2B cells induced by coal tar pitch extract.

Coal tar pitch extract (CTPE) was carcinogenic and could cause occupational lung cancer. Hence, we explored the changes of protein molecules during CTPE-induced malignant transformation (MT) of immortalized human bronchial epithelial (BEAS-2B) cells and provided clues for screening early biomarkers of CTPE-associated occupational lung cancer. The MT model of BEAS-2B cells induced by CTPE with 15.0 µg/mL. Subsequently, the MT of the BEAS-2B cells was verified by morphological observation, cell proliferation test, plate colony formation assay, and cell cycle assay. At the end of the experiment, we explored the differentially expressed proteins (DEPs) by total protein tandem mass tags quantitative proteomics technique between DMSO40 cells and CTPE40 cells. It was found that the proliferation ability, and colony formation rate were enhanced, and the cell cycle was changed. Then, bioinformatics analysis showed that a total of 107 DEPs were screened between CTPE40 and DMSO40 cells, of which 74 were up-regulated and 33 were down-regulated. As a result, 6 hub proteins were screened by protein-protein interaction network analysis. The expression levels of COX7A2, COX7C, MT-CO2, NDUFB4, and NDUFB7 were up-regulated as well as the expression of RPS29 protein was down-regulated. In summary, we established an MT model in vitro and explored the changes in protein molecules. As a result, this study suggested that changes of protein molecules, including COX7A2, COX7C, NDUFB7, MT-CO2, NDUFB4, and RPS29, occurred at the stage of BEAS-2B cell malignancy following CTPE exposure, which provided key information for screening biomarkers for CTPE-related occupational lung cancer.

Microbial infections as potential risk factors for lung cancer: Investigating the role of human papillomavirus and chlamydia pneumoniae.

Background: Lung cancer is the leading cause of cancer morbidity and mortality worldwide. Apart from tobacco smoke and dietary factors, microbial infections have been reported as the third leading cause of cancers globally. Deciphering the association between microbiome and lung cancer will provide potential biomarkers and novel insight in lung cancer progression. In this current study, we performed a meta-analysis to decipher the possible association between C. pneumoniae and human papillomavirus (HPV) and the risk of lung cancer. Methods: Literature search was conducted in most English and Chinese databases. Data were analyzed using CMA v.3.0 and RevMan v.5.3 software (Cochrane-Mantel-Haenszel method) by random-effects (DerSimonian and Laird) model. Results: The overall pooled estimates for HPV studies revealed that HPV infections in patients with lung cancer were significantly higher than those in the control group (OR = 2.33, 95% CI = 1.57-3.37, p < 0.001). Base on subgroup analysis, HPV infection rate was significantly higher in Asians (OR = 6.38, 95% CI = 2.33-17.46, p < 0.001), in tissues (OR = 5.04, 95% CI = 2.27-11.19, p < 0.001) and blood samples (OR = 1.40, 95% CI = 1.02-1.93, p = 0.04) of lung cancer patients but non-significantly lower in males (OR = 0.84, 95% CI = 0.57-1.22, p =0.35) and among lung cancer patients at clinical stage I-II (OR = 0.95, 95% CI = 0.61-1.49, p = 0.82). The overall pooled estimates from C. pneumoniae studies revealed that C. pneumoniae infection is a risk factor among lung cancer patients who are IgA seropositive (OR = 1.88, 95% CI = 1.30-2.70, p < 0.001) and IgG seropositive (OR = 1.50, 95% CI = 1.10-2.04, p = 0.010). All seronegative IgA (OR = 0.69, 95% CI = 0.42-1.16, p = 0.16) and IgG (OR = 0.66, 95% CI = 0.42-105, p = 0.08) titers are not associative risk factors to lung cancer. Conclusions: Immunoglobulin (IgA) and IgG seropositive titers of C. pneumoniae and lungs infected with HPV types 16 and 18 are potential risk factors associated with lung cancer.

A SERS bioassay based on vancomycin-modified PEI-interlayered nanocomposite and aptamer-functionalized SERS tags for synchronous detection of Acinetobacter baumannii and Klebsiella pneumoniae.

Klebsiella pneumoniae (KP) and Acinetobacter baumannii (AB) are two important gram-negative bacteria that cause pneumonia and have been recently known to be associated with food. The rapid detection of these pathogens in food is important to minimize their colonization of the gut and stop new threats of the disease from spreading across the food chain. Herein, a double-edged sword aptasensor was developed for the synchronous detection of KP and AB in food and clinical samples. A highly sensitive, selective, specific, and synchronous detection of the target bacteria was achieved, and the limit of detection (LOD) was 10 cells/mL with a liner range of 50 to 105 cells/mL. The total assay time was 1.5 h. This study does not only provide a new tool for the detection of the target bacteria, but also serves as a promising tool for food safety and pneumonia diagnosis.

Universal DNAzyme walkers-triggered CRISPR-Cas12a/Cas13a bioassay for the synchronous detection of two exosomal proteins and its application in intelligent diagnosis of cancer.

Exosomal proteins are considered to be promising indicators of cancer. Herein, a novel DNAzyme walkers-triggered CRISPR-Cas12a/Cas13a strategy was proposed for the synchronous determination of exosomal proteins: serum amyloid A-1 protein (SAA1) and coagulation factor V (FV). In this design, the paired antibodies were used to recognize targets, thereby ensuring the specificity. DNAzyme walkers were employed to convert the contents of SAA1 and FV into activators (P1 and P2), and one target can produce abundant activators, thus achieving an initial amplification of signal. Furthermore, the P1 and P2 can activate CRISPR-Cas12a/Cas13a system, which in turn trans-cleaves the reporters, enabling a second amplification and generating two fluorescent signals. The assay is highly sensitive (limits of detection as low as 30.00 pg/mL for SAA1 and 200.00 pg/mL for FV), highly specific and ideally accurate. More importantly, it is universal and can be used to detect both non-membrane and membrane proteins in exosome. Besides, the method can be successfully applied to detect SAA1 and FV in plasma exosomes to differentiate between lung cancer patients and healthy individuals. To explore the application of the developed method in tumor diagnosis, a deep learning model based on the expressions of SAA1 and FV was developed. The accuracy of this model can achieve 86.96%, which proves that it has a promising practical application capacity. Thus, this study does not only provide a new tool for the detection of exosomal proteins and cancer diagnosis, but also propose a new strategy to detect non-nucleic acid analytes for CRISPR-Cas system.

Epigenetic compounds targeting pharmacological target lysine specific demethylase 1 and its impact on immunotherapy, chemotherapy and radiotherapy for treatment of tumor recurrence and resistance.

It has been proven that metastatic recurrence and therapeutic resistance are linked. Due to the variability of individuals and tumors, as well as the tumor's versatility in avoiding therapies, therapy resistance is more difficult to treat. Therapy resistance has significantly restricted the clinical feasibility and efficacy of tumor therapy, despite the discovery of novel compounds and therapy combinations with increasing efficacy. In several tumors, lysine specific demethylase 1 (LSD1) has been associated to metastatic recurrence and therapeutic resistance. For researchers to better comprehend how LSD1-mediated tumor therapy resistance occurs and how to overcome it in various tumors, this study focused on the role of LSD1 in tumor recurrence and therapeutic resistance. The importance of therapeutically targeted LSD1 was also discussed. Most gene pathway signatures are related to LSD1 inhibitor sensitivity. However, some gene pathway signatures, especially in AML, negatively correlate with LSD1 inhibitor sensitivity, but targeting LSD1 makes the therapy-resistant tumor sensitive to physiological doses of conventional therapy. We propose that combining LSD1 inhibitor with traditional tumor therapy can help patients attain a complete response and prevent cancer relapse.

Magnetic-nanowaxberry-based microfluidic ExoSIC for affinity and continuous separation of circulating exosomes towards cancer diagnosis.

Exosomes are seen as promising biomarkers for minimally invasive liquid biopsies and disease surveillance. However, the complexity of body fluids, inherent heterogeneity, and tiny size of exosomes impede their extraction, consequently restricting their clinical application. In this study, in order to efficiently isolate exosomes from clinical samples, an irregular serpentine channel microfluidic chip (ExoSIC) was designed to continuously separate exosomes from plasma based on a magnetic-nanowaxberry (MNWB). In the ExoSIC, irregular serpentine microchannels are utilized to increase fluid chaotic mixing, hence improving exosome capture efficiency. In comparison to commonly used spherical magnetic particles, the designed MNWB can not only enhance the capture efficiency of exosomes, but also possess a size-exclusion effect to improve exosome purity. Consequently, the ExoSIC exhibited a large yield (24 times higher than differential centrifugation), optimum purity (greater than precipitation and similar to differential centrifugation), and high specificity. Furthermore, the ExoSIC was utilized for plasma-based cancer diagnosis by multiplex monitoring of five exosomal biomarkers (exosomal concentration, EGFR, EpCAM, SAA1 and FV), and the AUC reached 0.791. This work provides a comprehensive framework for exosome-based cancer diagnostics in order to meet clinical requirements for exosome isolation and downstream analysis.

A digital immuno-PCR assay for simultaneous determination of 5-methylcytosine and 5-hydroxymethylcytosine in human serum.

This work aimed to develop an ultrasensitive and specific immunosorbent assay for simultaneous detection of double DNA methylation marks. Being considered the most important indicators in disease diagnosis, clinical treatment, and prognosis, 5-methylcytosine (5mC) and 5-hydroxymethylcytosine (5hmC) were chosen as the proof-of-concept targets. The described strategy consisted of Phos-tag Biotin anchoring at streptavidin-magnetic nanoparticles, specific immune recognition of anti-5mC antibody and anti-5hmC antibody and labeling of Barcode-antibody, signal amplification of immune PCR and digital PCR machine. Under optimal conditions, the digital immuno-PCR assay showed a board dynamic range from 2.7 × 10-13 mol/L to 2.7 × 10-9 mol/L and the detection limits were 61.7 fmol/L for 5mC, and of 0.111 pmol/L for 5hmC. A 16-fold and 186-fold improvement of LOD were obtained by the proposed approach for 5mC and 5hmC detection compared with real-time immune PCR. The approach also showed ideal specificity, repeatability and stability. The recovery test demonstrated that the digital immuno-PCR assay is a promising platform for the simultaneous determination of the two DNA methylation marks in human serum sample.

Machine learning-assisted prediction of pneumonia based on non-invasive measures.

Background: Pneumonia is an infection of the lungs that is characterized by high morbidity and mortality. The use of machine learning systems to detect respiratory diseases via non-invasive measures such as physical and laboratory parameters is gaining momentum and has been proposed to decrease diagnostic uncertainty associated with bacterial pneumonia. Herein, this study conducted several experiments using eight machine learning models to predict pneumonia based on biomarkers, laboratory parameters, and physical features. Methods: We perform machine-learning analysis on 535 different patients, each with 45 features. Data normalization to rescale all real-valued features was performed. Since it is a binary problem, we categorized each patient into one class at a time. We designed three experiments to evaluate the models: (1) feature selection techniques to select appropriate features for the models, (2) experiments on the imbalanced original dataset, and (3) experiments on the SMOTE data. We then compared eight machine learning models to evaluate their effectiveness in predicting pneumonia. Results: Biomarkers such as C-reactive protein and procalcitonin demonstrated the most significant discriminating power. Ensemble machine learning models such as RF (accuracy = 92.0%, precision = 91.3%, recall = 96.0%, f1-Score = 93.6%) and XGBoost (accuracy = 90.8%, precision = 92.6%, recall = 92.3%, f1-score = 92.4%) achieved the highest performance accuracy on the original dataset with AUCs of 0.96 and 0.97, respectively. On the SMOTE dataset, RF and XGBoost achieved the highest prediction results with f1-scores of 92.0 and 91.2%, respectively. Also, AUC of 0.97 was achieved for both RF and XGBoost models. Conclusions: Our models showed that in the diagnosis of pneumonia, individual clinical history, laboratory indicators, and symptoms do not have adequate discriminatory power. We can also conclude that the ensemble ML models performed better in this study.

Construction and bioapplications of aptamer-based dual recognition strategy.

Aptamer-based dual recognition strategy, using dual aptamers or the cooperation of aptamers with other recognition elements, can better utilize the advantages of each recognition molecule and increase the design flexibility to effectively overcome the limitations of a single molecule recognition strategy, thereby improving the sensitivity and selectivity and facilitating the regulation of biological process. Hence, this review systematically tracks the construction and application of dual aptamers recognition strategy in the versatile detection of protein biomarkers, pathogenic microorganisms, cancer cells, and the treatment of some diseases and, more importantly, in functional regulation and imaging of cell-surface protein receptors. Then, the cooperation of aptamers with other recognition elements are briefly introduced. Potential challenges facing this field have been highlighted, aiming to expand bioanalytical applications of aptamer-based dual or multiple recognition strategies and meet the growing demand for precision medicine.

Application of logistic regression and convolutional neural network in prediction and diagnosis of high-risk populations of lung cancer.

OBJECTIVES: The early detection, early diagnosis, and early treatment of lung cancer are the best strategies to improve the 5-year survival rate. Logistic regression analysis can be a helpful tool in the early detection of high-risk groups of lung cancer. Convolutional neural network (CNN) could distinguish benign from malignant pulmonary nodules, which is critical for early precise diagnosis and treatment. Here, we developed a risk assessment model of lung cancer and a high-precision classification diagnostic model using these technologies so as to provide a basis for early screening of lung cancer and for intelligent differential diagnosis. METHODS: A total of 355 lung cancer patients, 444 patients with benign lung disease and 472 healthy people from The First Affiliated Hospital of Zhengzhou University were included in this study. Moreover, the dataset of 607 lung computed tomography images was collected from the above patients. The logistic regression method was employed to screen the high-risk groups of lung cancer, and the CNN model was designed to classify pulmonary nodules into benign or malignant nodules. RESULTS: The area under the curve of the lung cancer risk assessment model in the training set and the testing set were 0.823 and 0.710, respectively. After finely optimizing the settings of the CNN, the area under the curve could reach 0.984. CONCLUSIONS: This performance demonstrated that the lung cancer risk assessment model could be used to screen for high-risk individuals with lung cancer and the CNN framework was suitable for the differential diagnosis of pulmonary nodules.