An invisible, robust copyright protection method for DNN-generated content.

Wide deployment of deep neural networks (DNNs) based applications (e.g., style transfer, cartoonish), stimulating the need for copyright protection of such application's production. Though some traditional visible copyright techniques exist, they often introduce undesired artifacts and compromise the aesthetic quality of the images. In this paper, we propose a novel invisible, robust copyright protection method, which is composed of two networks: the copyright encoder and the copyright decoder. The former projects the copyright information to the invisible perturbation with the drive of both the input of images and copyright information, thereby adding it to the image and yielding encoded images. The copyright decoder extracts copyright information from encoded images. Moreover, a robustness module is integrated to enhance the decoder's ability to decipher images against various distortions encountered on social media platforms. Furthermore, the loss function is elaborately designed, taking into account both feature space and color space, to guarantee the quality of encoded and decoded copyright images. Extensively objective and subjective experiments validate the effectiveness of the proposed method. Additionally, the physical test is conducted by posting the encoded images to social media (e.g., Weibo and Twitter) and downloading them to verify the feasibility of the proposed method in practice.

Design of dual carbon encapsulated porous micron silicon composite with compact surface for enhanced reaction kinetics of lithium-ion battery anodes.

Developing high-performance composites with fast charging and superior cycle life is paramount for lithium-ion batteries (LIBs). Herein, we synthesized a double-shell carbon-coated porous structure composite with a compact surface (P-Si@rGO@C) using low-cost commercial micron-sized silicon (Si) instead of nanoscale silicon. Results reveal that the unique P-Si@rGO@C features high adaptability to volume expansion, accelerates electron/ion transmission rate, and forms a stable solid electrolyte interphase (SEI) film. This phenomenon arises from the synergistic effect of abundant internal voids and an external double-layer carbon shell with a dense surface. Specifically, the P-Si@rGO@C anode exhibits a high initial coulombic efficiency (ICE) (88.0 %), impressive rate-capability (612.1 mAh/g at 2C), and exceptional long-term cyclability (972.2 mAh/g over 500 cycles at 0.5C). Further kinetic studies elucidate the diffusion-capacitance hybrid energy storage mechanism and reveal an improved Li+ diffusion coefficient (from 3.47 × 10-11 to 2.85 × 10-9 cm2 s-1). Ex-situ characterization confirms the crystal phase change of micron-sized Si and the formation of a stable LiF-rich SEI. Theoretical calculations support these findings by demonstrating an enhancement in the adsorption ability of Si to Li+ (from -0.89 to -0.97 eV) and a reduction in the energy migration barrier (from 0.35 to 0.18 eV). Additionally, practical LixSi powder is shown to increase the ICE of full cells from 67.4 % to 87.9 %. Furthermore, a pouch cell utilizing the prelithiated P-Si@rGO@C anode paired with LiNi1/3Co1/3Mn1/3O2 (NCM111) cathode delivers a high initial reversible capacity of 7.2 mAh and 76.8 % capacity retention after 100 cycles. This work provides insights into the application of commercial silicon-aluminum alloy powder in the anode of high-energy LIBs.

Oil-in-water and oleogel-in-water emulsion encapsulate with hemp seed oil containing Δ9-tetrahydrocannabinol and cannabinol: Stability, degradation and in vitro simulation characteristics.

The present study focused on investigating the stability and in vitro simulation characteristics of oil-in-water (O/W) and oleogel-in-water (Og/W) emulsions. Compared with O/W emulsion, the Og/W emulsion exhibited superior stability, with a more evenly spread droplet distribution, and the Og/W emulsion containing 3 % hemp seed protein (HSP) showed better stability against environmental factors, including heat treatment, ionic strength, and changes in pH. Additionally, the stability of Δ9-tetrahydrocannabinol (Δ9-THC) and cannabinol (CBN) and the in vitro digestion of hemp seed oil (HSO) were evaluated. The half-life of CBN in the Og/W emulsion was found to be 131.82 days, with a degradation rate of 0.00527. The in vitro simulation results indicated that the Og/W emulsion effectively delayed the intestinal digestion of HSO, and the bioaccessibility of Δ9-THC and CBN reached 56.0 % and 58.0 %, respectively. The study findings demonstrated that the Og/W emulsion constructed with oleogel and HSP, exhibited excellent stability.

Ultrasonic-ethanol pretreatment assisted aqueous enzymatic extraction of hemp seed oil with low Δ9-THC.

In this study, ultrasonic-ethanol pretreatment combined with AEE was developed for oil extraction from hemp seeds. The oil yield reached a maximum of 23.32 % at 200 W ultrasonic power and 30 min ultrasonic time, at this point, the degradation rate of Δ9-THC was 83.11 %. By determining the composition of hemp seed before and after pretreatment, it was shown that ultrasonic-ethanol pretreatment reduced the protein content of the raw material. An enzyme mixture consisting of pectinase and hemicellulase (1/1/1, w/w/w) was experimentally determined to be used, and the AEE extraction conditions were optimized using the Plackett-Burman design and the Box-Behnken. The optimal conditions were determined to be pH 5, total enzyme activity of 37,800 U/g, liquid-solid ratio of 10.4 mL/g, enzyme digestion temperature of 32 °C, enzymatic time of 189 min, and oil recovery of 88.38 %. The results of confocal laser scanning microscopy (CLSM) and scanning electron microscopy (SEM) showed that the emulsion formed during ultrasonic ethanol pretreatment was not uniformly distributed, and the droplets appeared to be aggregated; and the irregular pores of hemp seed increased after pretreatment. The contents of Δ9-THC and CBN in the extracted oil samples were 9.58 mg/kg and 52.45 mg/kg, respectively. Compared with the oil extracted by Soxhlet extraction (SE), the oil extracted by this experimental method was of better quality and similar in fatty acid composition.

Improving Transferability of Universal Adversarial Perturbation With Feature Disruption.

Deep neural networks (DNNs) are shown to be vulnerable to universal adversarial perturbations (UAP), a single quasi-imperceptible perturbation that deceives the DNNs on most input images. The current UAP methods can be divided into data-dependent and data-independent methods. The former exhibits weak transferability in black-box models due to overly relying on model-specific features. The latter shows inferior attack performance in white-box models as it fails to exploit the model's response information to benign images. To address the above issues, this paper proposes a novel universal adversarial attack to generate UAP with strong transferability by disrupting the model-agnostic features (e.g., edges or simple texture), which are invariant to the models. Specifically, we first devise an objective function to weaken the significant channel-wise features and strengthen the less significant channel-wise features, which are partitioned by the designed strategy. Furthermore, the proposed objective function eliminates the dependency on labeled samples, allowing us to utilize out-of-distribution (OOD) data to train UAP. To enhance the attack performance with limited training samples, we exploit the average gradient of the mini-batch input to update the UAP iteratively, which encourages the UAP to capture the local information inside the mini-batch input. In addition, we introduce the momentum term to accumulate the gradient information at each iterative step for the purpose of perceiving the global information over the training set. Finally, extensive experimental results demonstrate that the proposed methods outperform the existing UAP approaches. Additionally, we exhaustively investigate the transferability of the UAP across models, datasets, and tasks.

Self-powered biosensor using photoactive ternary nanocomposite: Testing the phospholipid content in rhodotorula glutinis oil.

In the field of oil refining, the presence of excessive residual phosphorus in crude oil can significantly impact its quality, thereby emphasizing the necessity for compact and convenient testing equipment. This study primarily focuses on developing of self-powered biosensor (SPB) using immobilizing Choline Oxidase with a photoactive ternary nanocomposite complex (CHOx-BiOI-rGO-Fe3O4 NPs-ITO) as the anode and utilizing a Pt electrode as the cathode. The successful preparation of the ternary composite photoelectrode for the anode was confirmed through a range of characterization techniques, including X-ray diffraction (XRD), Fourier transform infrared spectroscopy (FTIR), Transmission electron microscopy (TEM), Scanning electron microscopy (SEM), X-ray photoelectron spectroscopy (XPS), N2 absorption/desorption, Dynamic light scattering (DLS), and Ultraviolet-visible diffuse reflection spectrometer (UV-vis DRS). The electrochemical and photoelectrochemical properties were assessed using an electrochemical workstation, revealing a significant enhancement photoelectrical responsiveness attributed to the formation of heterojunction structures. The SPB exhibited a remarkable linear relationship between the instantaneous photocurrent and phosphatidylcholine (PC) concentration, with a regression equation of I (µA) = 39.62071C (mM) + 3.47271. The linear range covered a concentration range of 0.01-10 mM, and the detection limit (S/N = 3) was determined to be 0.008 mM. It demonstrated excellent reproducibility and storage stability, positioning it a promising alternative to High-performance liquid chromatography (HPLC) for accurate quantification of PC content in rhodotorula glutinis oil. The standard recovery PC content ranged from 98.48% to 103.53%, with a relative standard deviation (RSD) ranging from 1.4% to 2.4%. This research presents a convenient and precise detection device that has the potential to address the issue of lagging detection in the oil refining process.

Ultrasonic treatment of rice bran protein-tannic acid stabilized oil-in-water emulsions: Focus on microstructure, rheological properties and emulsion stability.

Rice bran protein (RBP)-tannic acid (TA) complex was prepared and the RBP-TA emulsions were subjected to ultrasonic treatment with different powers. Ultrasonic treatment has a positive effect on improving the properties of RBP-TA emulsion. This study investigated the influence of different ultrasonic power levels on the physicochemical properties, microstructure, rheological properties, and stability of emulsions containing RBP-TA. Under the ultrasonic treatment of 400 W, the particle size, zeta potential, and adsorbed protein content of the RBP-TA emulsion were 146.86 nm, -20.7 eV, and 61.91%, respectively. At this time, the emulsion had the best emulsifying properties, apparent viscosity, energy storage modulus and loss modulus. In addition, the POV and TBARS values of RBP-TA emulsions were 6.12 and 7.60 mmol/kg, respectively. The thermal, salt ion, pH and oxidative stability of the emulsions were investigated, and it was shown that ultrasonic treatment was effective in improving the stability of RBP-TA emulsions.

Reversible active bridging sulfur sites grafted on Ni3S2 nanobelt arrays for efficient hydrogen evolution reaction.

Elaborate and rational design of cost-effective and high-efficiency non-noble metal electrocatalysts for pushing forward the sustainable hydrogen fuel production is of great significance. Herein, a novel VS4 nanoparticle decorated Ni3S2 nanobelt array in-situ grown on nickel foam (VS4/Ni3S2/NF NBs) was prepared by a self-templated synthesis strategy. Benefitting from the unique nanobelt array structure, abundant highly active bridge S22- sites and strong electronic interaction between VS4 and Ni3S2 on the heterointerface, the integrated VS4/Ni3S2/NF NBs exhibited excellent electrocatalytic hydrogen evolution activity and robust stability. The density functional theory (DFT) further revealed the reversible conversion catalysis mechanism of bridging S22- sites in VS4/Ni3S2/NF NBs during HER process. Notably, bidentate bridging SS bonds as the predominant catalytically active centers can spontaneously open once H adsorbed its surface, leading to the aggregation of negative charges on S atoms and thus facilitating the generation of H* intermediates, and spontaneously close when H* desorption is going to form H2. Our work provides fresh insights for developing potential polysulfides as high-performance hydrogen-evolving electrocatalysts for prospective clean energy production from water splitting.

Differential evolution based dual adversarial camouflage: Fooling human eyes and object detectors.

Deep neural network-based object detectors are vulnerable to adversarial examples. Among existing works to fool object detectors, the camouflage-based method is more often adopted due to its adaptation to multi-view scenarios and non-planar objects. However, most of them can still be easily observed by human eyes, which limits their application in the real world. To fool human eyes and object detectors simultaneously, we propose a differential evolution based dual adversarial camouflage method. Specifically, we try to obtain the camouflage texture by the two-stage training, which can be wrapped over the surface of the object. In the first stage, we optimize the global texture to minimize the discrepancy between the rendered object and the scene background, making human eyes difficult to distinguish. In the second stage, we design three loss functions to optimize the local texture, which is selected from the global texture, making object detectors ineffective. In addition, we introduce the differential evolution algorithm to search for the near-optimal areas of the object to attack, improving the adversarial performance under certain attack area limitations. Experimental results show that our proposed method can obtain a good trade-off between fooling human eyes and object detectors under multiple specific scenes and objects.

In Situ Growth of Iron Sulfide on Fast Charge Transfer V2 C-MXene for Superior Sodium Storage Anodes.

Due to the upstream pressure of lithium resources, low-cost sodium-ion batteries (SIBs) have become the most potential candidates for energy storage systems in the new era. However, anode materials of SIBs have always been a major problem in their development. To address this, V2 C/Fe7 S8 @C composites with hierarchical structures prepared via an in situ synthesis method are proposed here. The 2D V2 C-MXene as the growth substrate for Fe7 S8  greatly improves the rate capability of SIBs, and the carbon layer on the surface provides a guarantee for charge-discharge stability. Unexpectedly, the V2 C/Fe7 S8 @C anode achieves satisfactory sodium storage capacity and exceptional rate performance (389.7 mAh g-1  at 5 A g-1 ). The sodium storage mechanism and origin of composites are thoroughly studied via ex situ characterization techniques and first-principles calculations. Furthermore, the constructed sodium-ion capacitor assembled with N-doped porous carbon delivers excellent energy density (135 Wh kg-1 ) and power density (11 kW kg-1 ), showing certain practical value. This work provides an advanced system of sodium storage anode materials and broadens the possibility of MXene-based materials in the energy storage.

KCNE4 expression is correlated with the pathological characteristics of colorectal cancer patients and associated with the radioresistance of cancer cells.

BACKGROUND: Colorectal cancer (CRC) is a common malignancy, and radioresistance limits the effectiveness of radiotherapy for rectal cancer. This study is performed to investigate the role and regulatory mechanism of Potassium Voltage-Gated Channel Subfamily E Regulatory Subunit 4 (KCNE4) in the radioresistance of CRC cells. METHODS: Immunohistochemical staining results of KCNE4 in normal tissues and CRC tissues were obtained from the Human Protein Atlas (HPA) database. The UALCAN database was used for analyzing KCNE4 mRNA expression in normal tissue samples and CRC tissue samples and its relationship with tumor stage. The relationship of KCNE4 expression with prognosis was analyzed utilizing the data of GEPIA database. LinkedOmics database was searched to analyze the co-expressed gene sets of KCNE4 in CRC, and to analyze the signaling pathways related with KCNE4 in CRC. GO and KEGG enrichment analyses were carried out on the co-expressed genes of KCNE4 with DAVID database. Ionizing radiation (IR)-resistant cell lines (HCT116/IR and HT29/IR) were established; cell viability was assessed via cell counting kit-8 (CCK-8) and EdU assays, and terminal deoxynucleotidyl transferase dUTP nick-end labeling (TUNEL) assay was performed for detecting cell apoptosis. Western blotting was carried out to detect the expressions of p-p85 and p-AKT. RESULTS: KCNE4 was highly expressed in CRC tissues and linked to advanced tumor stage, lymph node metastasis and poor prognosis of CRC patients. KCNE4 overexpression promoted HCT116/IR cell proliferation and inhibited the apoptosis, while KCNE4 knockdown suppressed HT29/IR cell proliferation and facilitated the apoptosis. Furthermore, high KCNE4 expression was associated with the activation of the PI3K/AKT signal pathway. CONCLUSION: KCNE4 is associated with the clinicopathological characteristics of CRC patients, and its high expression level contributes to the radioresistance of cancer cells via activating the PI3K/AKT signal pathway.

Machine learning algorithm for estimating karst rocky desertification in a peak-cluster depression basin in southwest Guangxi, China.

Karst rocky desertification (KRD) has become one of the most serious ecological and environmental problems in karst areas. At present, mapping KRD with a high accuracy and on a large scale is still a difficult problem in the control of KRD. In this study, a random forest (RF) based on maximum information coefficient and correlation coefficient feature selection is proposed to predict KRD. Nine predictors stood out as feature factors to estimate KRD. Rock exposure was the most important predictor, followed by fractional vegetation cover for the prediction of KRD processes. The kappa and classification accuracy indexes were to evaluate the performance of the model. We recorded overall accuracy rate and kappa index values of 94.7% and 0.92 for the testing datasets respectively. The RF model was then used to predict the KRD in 2001, 2011, 2016, and 2020, and it was found that the KRD in the study area has exhibited a positive trend of improvement. Therefore, the use of multisource remote sensing data combined with the RF model can obtain better prediction results of KRD, thereby providing a new idea for large-scale estimation of the KRD in peak-cluster depression.

Injection of Carbon Nanoparticles for Lymph Node Detection After Laparoscopic Colorectal Cancer Surgery.

Colorectal cancer is the third most common cause of cancer-related death in the world because of its poor prognosis and many related complications. In clinical practice, lymph node metastasis is an important index to evaluate the recurrence and survival rate of patients with rectal cancer, it also the main method to determine the treatment plan of patients with rectal cancer. Nano carbon tracer can guide surgeons to clear regional lymph nodes accurately, improve the number of lymph nodes detected, and then improve the accuracy of pathological staging of rectal cancer. It has an important clinical significance in the removal of lymph nodes of middle and low rectal cancer, and provides an important basis for the development of postoperative adjuvant treatment plan. The purpose of this study is to explore the application value of this method in improving the detection rate of lymph nodes in laparoscopic colorectal cancer. According to the research results, the total number of lymphatic gland, the number of small lymphatic gland, the rate of lymph node metastasis, the number of black stained lymphatic gland and the number of lymphatic gland confirmed by pathology in the two groups were better than those in the traditional surgical treatment, which had better application value in the radical operation of colorectal cancer.

Diagnostic and Prognostic Value of TAT, PIC, TM, and t-PAIC in Malignant Tumor Patients With Venous Thrombosis.

BACKGROUND: Venous thromboembolism (VTE) is an important complication in patients with malignant tumors. Its exact diagnosis and treatment are still lacking. We used a high-sensitive chemiluminescence method to detect thrombin-antithrombin III complex (TAT), plasmin-α2-plasmininhibitor complex (PIC), thrombomodulin (TM), and tissue plasminogen activator-inhibitor complex(t-PAIC) in combination with D-dimer and fibrin degradation product (FDP) to analyze their diagnostic and prognostic value in patients with malignant tumors. METHODS: In total, 870 patients with confirmed malignant tumors were included, 82 of whom had diagnosed VTE; 200 healthy individuals were classified as the control group. The TAT, PIC, TM, and t-PAIC were detected using Sysmex HISCL5000 automated analyzers, whereas FDP and D-dimer were detected using Sysmex CS5100 coagulation analyzer. Receiver operating characteristic (ROC) curves were used to evaluate the diagnostic efficiency. Survival probabilities were determined using Kaplan-Meier analysis, and multivariate analyses were performed using a Cox regression model. RESULTS: Compared with healthy controls, patients with malignant tumors showed significantly elevated TAT, PIC, TM, t-PAIC, D-dimer, and FDP. Similarly, compared with patients in the non-thrombosis group, those in the thrombosis group showed significantly elevated levels of the above mentioned markers. Logistic regression analysis showed that TAT, PIC, TM, t-PAIC, D-Dimer, and FDP were all associated with VTE. ROC analysis showed that "TAT+PIC+TM+t-PAIC+D-dimer+FDP"showed the highest sensitivity and specificity. Patients with elevated TAT, PIC, TM, and t-PAIC had a significantly shorter survival. Multivariate Cox survival analysis showed that TM and t-PAIC were significantly associated with poor prognosis. In addition, the incidence of VTE was significantly lower in patients with malignant tumors who were treated with low-molecular-weight heparin (LMWH), and their survival period was significantly longer than that of patients with malignant tumors who were not treated with LMWH. CONCLUSION: TAT, PIC, TM, and t-PAIC combined with D-dimer and FDP were better than the application of a single marker in the diagnosis of VTE in patients with malignant tumors. TAT and PIC can be used as sensitive markers in the diagnosis of VTE but not as prognostic markers. TM and t-PAIC might be independent prognostic indicators in patients with malignant tumors, regardless of the state of thrombus.

Unintended pregnancy and postpartum depression: A meta-analysis of cohort and case-control studies.

OBJECTIVE: The majority of original studies showed that unintended pregnancy is related to adverse obstetric outcomes, however, up to now, the influence of unintended pregnancy on the risk of developing postpartum depression (PPD) remains unclear. This study aimed to assess the association between unintended pregnancy and the risk of developing PPD by conducting a meta-analysis of cohort and case-control studies. METHODS: PubMed, Web of Science, Embase, and Cochrane Library were searched up to December 31, 2019 to identify relevant studies evaluating the association between unintended pregnancy and PPD. Meta-analysis was performed using RevMan software and Stata software. Potential heterogeneity source was explored by subgroup and sensitivity analyses, and potential publication bias was tested using Begg's funnel plots and Begg's linear regression test. RESULTS: A total of thirty studies involving 65,454 participants were included in our meta-analysis. Overall, women who get pregnant unintendedly compared with those who are intending to be pregnant were at a significantly higher risk of developing PPD (odds ratio [OR] = 1.53; 95% confidence interval [CI]: 1.35-1.74; P < 0.00001). CONCLUSIONS: Unintended pregnancy is significantly associated with the risk of developing PPD. These findings highlight the necessity of screening for pregnancy intention and integrating family planning and personalized mental health services into primary healthcare to promote maternal mental health.

Inference of Subpathway Activity Profiles Reveals Metabolism Abnormal Subpathway Regions in Glioblastoma Multiforme.

Glioblastoma, also known as glioblastoma multiforme (GBM), is the most malignant form of glioma and represents 81% of malignant brain and central nervous system (CNS) tumors. Like most cancers, GBM causes metabolic recombination to promote cell survival, proliferation, and invasion of cancer cells. In this study, we propose a method for constructing the metabolic subpathway activity score matrix to accurately identify abnormal targets of GBM metabolism. By integrating gene expression data from different sequencing methods, our method identified 25 metabolic subpathways that were significantly abnormal in the GBM patient population, and most of these subpathways have been reported to have an effect on GBM. Through the analysis of 25 GBM-related metabolic subpathways, we found that (S)-2,3-Epoxysqualene, which was at the central region of the sterol biosynthesis subpathway, may have a greater impact on the entire pathway, suggesting a potential high association with GBM. Analysis of CCK8 cell activity indicated that (S)-2,3-Epoxysqualene can indeed inhibit the activity of U87-MG cells. By flow cytometry, we demonstrated that (S)-2,3-Epoxysqualene not only arrested the U87-MG cell cycle in the G0/G1 phase but also induced cell apoptosis. These results confirm the reliability of our proposed metabolic subpathway identification method and suggest that (S)-2,3-Epoxysqualene has potential therapeutic value for GBM. In order to make the method more broadly applicable, we have developed an R system package crmSubpathway to perform disease-related metabolic subpathway identification and it is freely available on the GitHub (https://github.com/hanjunwei-lab/crmSubpathway).

Silybin Prevents Prostate Cancer by Inhibited the ALDH1A1 Expression in the Retinol Metabolism Pathway.

BACKGROUND: Silybin was known to exert inhibition in prostate cancer, but the underlying mechanism remained largely unknown. This study was designed to find out the potential target of Silybin on prostate cancer and explore the relative mechanisms. METHODS: Firstly, we screened the possible targets of Silybin through the PubChem database and Subpathway - GM. Then DU145 cells were transferred to investigate the correction about related targets, magnetic bead sorting and flow cytometry were used to sort and identify the cells. Proliferation, migration and invasion ability of DU145 cells were detected by MTT assay, Transwell assay, plate clonality and sphere formation assay. BALB/c nude mice were constructed models with implanted sarcoma and measured the tumor volume every 5 days as wells tumor weight. The levels of proteins were detected by Western blot and immunocytochemistry. RT-PCR was selected to test the expression of protein's mRNA. RESULTS: It was screened out the ALDH1A1 was highly correlated with subpathways of the Silybin risk metabolic pathway. And ALDH1A1 expression was positively correlated RARα with Ets1 by interfering with the ALDH1A1 gene. Importantly, ALDH1A1(+) cells showed proliferation, migration and invasion ability. In addition, it showed that Silybin exerted the inhibition on prostate cells by suppressed the proliferation, migration and invasion ability of cells in vitro experiment. Silybin also reduced the tumor volume and weight. And Silybin displayed obviously reduced the proteins and mRNA of ALDH1A1, RARα, Ets1 and MMP9 expressions. CONCLUSION: Our results indicated that Silybin showed inhibition of prostate cancer and the mechanism was involving with downregulating ALDH1A1 expression, thereby inhibiting the activation of RARα and preventing the activation of Ets1 to inhibit the growth and invasion of prostate cancer.

Prediction of Anticancer Peptides Using a Low-Dimensional Feature Model.

Cancer is still a severe health problem globally. The therapy of cancer traditionally involves the use of radiotherapy or anticancer drugs to kill cancer cells, but these methods are quite expensive and have side effects, which will cause great harm to patients. With the find of anticancer peptides (ACPs), significant progress has been achieved in the therapy of tumors. Therefore, it is invaluable to accurately identify anticancer peptides. Although biochemical experiments can solve this work, this method is expensive and time-consuming. To promote the application of anticancer peptides in cancer therapy, machine learning can be used to recognize anticancer peptides by extracting the feature vectors of anticancer peptides. Nevertheless, poor performance usually be found in training the machine learning model to utilizing high-dimensional features in practice. In order to solve the above job, this paper put forward a 19-dimensional feature model based on anticancer peptide sequences, which has lower dimensionality and better performance than some existing methods. In addition, this paper also separated a model with a low number of dimensions and acceptable performance. The few features identified in this study may represent the important features of anticancer peptides.