Neural Network Learning Algorithm Research Articles

STRUCTURAL INSIGHTS AND ANTICANCER POTENTIAL OF MELITTIN IN CD147 INTERACTION

Objective: This study investigates the interaction between melittin (PDB ID: 2MLT), a bioactive peptide from honeybee venom, and CD147 (PDB ID: 5XF0), a glycosylated transmembrane protein implicated in tumor progression. Material and Method: Employing molecular docking and bioinformatics tools, our structural analysis reveals diverse binding features, including hydrogen bonds, salt bridges, and non-bonded contacts, between the CD147 complex and melittin. Result and Discussion: Non-bonded interactions between 2MLT and specific amino acids (Gly181 and Arg201) of CD147 are highlighted, resembling aspects of the CypA/CD147 binding mechanism (Pro180-Gly181 and Arg201). The elevated anticancer potential of 2MLT was substantiated by utilizing the AntiCP 2.0 server and the ENNAACT server, employing machine learning and artificial neural network algorithms. Additionally, hydrophobicity analysis aligns with characteristics associated with anticancer peptides. Notably, thermodynamic stability variations with temperature underscore the robust binding affinity of 2MLT to the 5XF0 receptor. While our study comprehensively explores molecular interactions and predictive analyses, further in vitro and in vivo investigations are crucial to validate these findings for potential therapeutic applications.

Research on the characteristics of electro-hydraulic position servo system of RBF neural network under fuzzy rules

A radial basis function neural network PID controller under fuzzy rules (FUZZY-RBF-PID) was designed for the electro-hydraulic position servo system under the influence of uncertain factors such as load mutation, and load stiffness change. Firstly, the mathematical model of the system is established, and the frequency domain and time domain analysis of the system are carried out. Secondly, based on the analysis results, a radial basis function (RBF) neural network PID controller is designed, and fuzzy rules are innovatively used to adjust the learning rate of PID parameters in the RBF neural network learning algorithm in real time. Thirdly, the simulation results show that under the action of the FUZZY-RBF-PID controller, the unit step response of the system has high steady-state accuracy, fast response speed, and under the condition of large load stiffness, the system can recover to the steady-state value faster after being disturbed. At the same time, when the input signal is the sinusoidal signal of 10 HZ, the system under the action of the FUZZY-RBF-PID controller has no obvious phase lag phenomenon, and the tracking error is minimal. The proposed method can effectively improve the comprehensive performance of the electro-hydraulic position servo system under the influence of uncertain factors.

Unveiling surgical expertise through machine learning in a novel VR/AR spinal simulator: A multilayered approach using transfer learning and connection weights analysis

BackgroundVirtual and augmented reality surgical simulators, integrated with machine learning, are becoming essential for training psychomotor skills, and analyzing surgical performance. Despite the promise of methods like the Connection Weights Algorithm, the small sample sizes (small number of participants (N)) typical of these trials challenge the generalizability and robustness of models. Approaches like data augmentation and transfer learning from models trained on similar surgical tasks address these limitations. ObjectiveTo demonstrate the efficacy of artificial neural network and transfer learning algorithms in evaluating virtual surgical performances, applied to a simulated oblique lateral lumbar interbody fusion technique in an augmented and virtual reality simulator. DesignThe study developed and integrated artificial neural network algorithms within a novel simulator platform, using data from the simulated tasks to generate 276 performance metrics across motion, safety, and efficiency. Innovatively, it applies transfer learning from a pre-trained ANN model developed for a similar spinal simulator, enhancing the training process, and addressing the challenge of small datasets. SettingMusculoskeletal Biomechanics Research Lab; Neurosurgical Simulation and Artificial Intelligence Learning Centre, McGill University, Montreal, Canada. ParticipantsTwenty-seven participants divided into 3 groups: 9 post-residents, 6 senior and 12 junior residents. ResultsTwo models, a stand-alone model trained from scratch and another leveraging transfer learning, were trained on nine selected surgical metrics achieving 75 % and 87.5 % testing accuracy respectively. ConclusionsThis study presents a novel blueprint for addressing limited datasets in surgical simulations through the strategic use of transfer learning and data augmentation. It also evaluates and reinforces the application of the Connection Weights Algorithm from our previous publication. Together, these methodologies not only enhance the precision of performance classification but also advance the validation of surgical training platforms.

Интеллектуальная система контроля бдительности и действий машиниста

Objective: the article considers an approach to the development of a system for monitoring the vigilance and actions of the driver using technical vision and neural network models. The main purpose of the research is to improve the safety of railway transport. Methods: data collection from various sources, annotation, data purification and normalization, neural network training based on video recordings of drivers’ faces in various states and data on their behavior. Neural network learning algorithms based on the architecture of convolutional neural networks, teaching methods with a teacher, mask segmentation methods and proportional resizing of the area of interest, mask segmentation methods for determining the contour of an object in an image, deep learning algorithms such as stochastic gradient descent and error back propagation. Results: a system has been developed that determines the emotional state of the driver based on a real-time video stream, detecting signs of fatigue or distraction, warning of possible dangerous situations. This approach will allow you to quickly respond to the risks that arise in the process of train control, which allows you to increase the level of train safety. Practical significance: a system for monitoring the vigilance and actions of the driver has been developed and can be implemented on locomotives or motor-car rolling stock for real monitoring and prevention of emergency situations.

OPTIMIZATION OF CYBER-PHYSICAL SYSTEM PARAMETERS BASED ON INTELLIGENT IOT SENSORS DATA

The optimization of cyber-physical systems (CPS) parameters is researched with focusing on the integration of computation, networking, and physical processes with the Internet of Things (IoT). The use of intelligent IoT sensors is crucial for collecting real-time data, which is necessary for enhancing the efficiency, reliability, and performance of CPS. Various methods of CPS parameters optimization are analyzed and categorized into model-based approaches, data-driven approaches, and hybrid approaches. The model-based approaches rely on mathematical models to describe CPS behavior and use optimization algorithms like linear programming and evolutionary algorithms to predict system responses and optimize parameters. But, the limitations of model-based approaches are related to complex systems with uncertain or dynamic behavior. The data-driven approaches are more suitable for complex cyber-physical systems. These approaches utilize machine learning and data analytics techniques to extract patterns from sensor data, which are then used to adjust system parameters. The hybrid approaches combine elements of both model-based and data-driven methods. The method of cyber-physical system parameters optimization based on intelligent IoT sensors data processing is developed with using of distributed neural network. The optimization problem is formulated with constraints for the system parameters. The neural network mathematical model and learning algorithm are proposed. The performed research shows the importance of developing optimization methods for CPS parameters based on intelligent IoT sensor data, considering the evolving nature of IoT technology. The integrating intelligent sensors into CPS offers new opportunities for optimizing system performance but also presents challenges in data management and security that should be addressed in future.

Artificial neural network based on strong track and square root UKF for INS/GNSS intelligence integrated system during GPS outage

When INS/GNSS (inertial navigation system/global navigation satellite system) integrated system is applied, it will be affected by the insufficient number of visible satellites, and even the satellite signal will be lost completely. At this time, the positioning error of INS accumulates with time, and the navigation accuracy decreases rapidly. Therefore, in order to improve the performance of INS/GNSS integration during the satellite signals interruption, a novel learning algorithm for neural network has been presented and used for intelligence integrated system in this article. First of all, determine the input and output of neural network for intelligent integrated system and a nonlinear model for weighs updating during neural network learning has been established. Then, the neural network learning based on strong tracking and square root UKF (unscented Kalman filter) is proposed for iterations of the nonlinear model. In this algorithm, the square root of the state covariance matrix is used to replace the covariance matrix in the classical UKF to avoid the filter divergence caused by the negative definite state covariance matrix. Meanwhile, the strong tracking coefficient is introduced to adjust the filter gain in real-time and improve the tracking capability to mutation state. Finally, an improved calculation method of strong tracking coefficient is presented to reduce the computational complexity in this algorithm. The results of the simulation test and the field-positioning data show that the proposed learning algorithm could improve the calculation stability and robustness of neural network. Therefore, the error accumulation of INS/GNSS integration is effectively compensated, and then the positioning accuracy of INS/GNSS intelligence integrated system has been improved.

Multi‐model deep learning system for screening human monkeypox using skin images

AbstractPurposeHuman monkeypox (MPX) is a viral infection that transmits between individuals via direct contact with animals, bodily fluids, respiratory droplets, and contaminated objects like bedding. Traditional manual screening for the MPX infection is a time‐consuming process prone to human error. Therefore, a computer‐aided MPX screening approach utilizing skin lesion images to enhance clinical performance and alleviate the workload of healthcare providers is needed. The primary objective of this work is to devise an expert system that accurately classifies MPX images for the automatic detection of MPX subjects.MethodsThis work presents a multi‐modal deep learning system through the fusion of convolutional neural network (CNN) and machine learning algorithms, which effectively and autonomously detect MPX‐infected subjects using skin lesion images. The proposed framework, termed MPXCN‐Net is developed by fusing deep features of three pre‐trained CNNs: MobileNetV2, DarkNet19, and ResNet18. Three classifiers—K‐nearest neighbour, support vector machine (SVM), and ensemble classifier—with various kernel functions, are used to identify infected patients. To validate the efficacy of our proposed system, we employ a publicly accessible MPX skin lesion dataset.ResultsBy amalgamating features extracted from all three CNNs and utilizing the medium Gaussian kernel of the SVM classifier, our proposed system achieves an outstanding average classification accuracy of 90.4%.ConclusionsDeveloped MPXCN‐Net is suitable for testing with a large diversified dataset before being used in clinical settings.

Frequency tunable dual band graphene-silicon ceramic-based two-port THz radiator with circular polarization and high isolation features

In this work, a two-port graphene-ceramic-based antenna is designed and discussed. Some important features for the proposed two-port radiator are (a) the feed layout produces dual radiating modes, i.e., HEM11δ and HEM12δ inside the ceramic at 2.6 and 5.2 THz, respectively; (b) the asymmetrical swastik-shaped aperture creates the circularly polarized waves within dual operating bands, i.e., 2.71–2.84 and 5.15–5.42 THz; (c) a coating of graphene over ceramic makes the radiator frequency tunable by varying its chemical potential; and (d) deep neural network (DNN) and XGBoost-based machine learning (ML) algorithms are used to predict the |S11| parameter of the designed antenna in order to reduce the computational complexity. By comparing the optimized outcome obtained from HFSS with the CST EM simulator and predicted values from ML algorithms, it is verified that the designed two-port radiator works in dual frequency bands, i.e., 2.4–3.2 and 5.0–5.43 THz. Stable values of MIMO, as well as the far-field parameter, confirm the applicability of the designed antenna for THz wireless applications.

Efficiency of Federated Learning and Blockchain in Preserving Privacy and Enhancing the Performance of Credit Card Fraud Detection (CCFD) Systems

Increasing global credit card usage has elevated it to a preferred payment method for daily transactions, underscoring its significance in global financial cybersecurity. This paper introduces a credit card fraud detection (CCFD) system that integrates federated learning (FL) with blockchain technology. The experiment employs FL to establish a global learning model on the cloud server, which transmits initial parameters to individual local learning models on fog nodes. With three banks (fog nodes) involved, each bank trains its learning model locally, ensuring data privacy, and subsequently sends back updated parameters to the global learning model. Through the integration of FL and blockchain, our system ensures privacy preservation and data protection. We utilize three machine learning and deep neural network learning algorithms, RF, CNN, and LSTM, alongside deep optimization techniques such as ADAM, SGD, and MSGD. The SMOTE oversampling technique is also employed to balance the dataset before model training. Our proposed framework has demonstrated efficiency and effectiveness in enhancing classification performance and prediction accuracy.

RECURRENT NEURAL NETWORK LEARNING ALGORITHM-BASED EVENT-TRIGGERED OBSERVER OF THE PERMANENT MAGNET SYNCHRONOUS MOTOR

In this paper, we propose a novel method to estimate the electrical angular velocity, the electrical angle, and the currents of the permanent magnet synchronous motor. A recurrent neural network learning algorithm is first developed to estimate the states of the permanent magnet synchronous motor. Then, an event-triggered state observer is designed for the recurrent neural network. This state observer robustly estimates state variables of the permanent magnet synchronous motor. A sufficient condition in terms of a convex optimization problem for the existence of the event-triggered state observer is established. In contrast with the abundance of state estimation methods based on time-triggered state observers where the measurements are always continuously available, the ones in this paper are updated when an event-triggered condition holds. Therefore, it lessens the stress on communication resources while can still maintain an estimation performance. Simulation results are provided to demonstrate the merit of the proposed method.

Quantized non-volatile nanomagnetic domain wall synapse based autoencoder for efficient unsupervised network anomaly detection

Anomaly detection in real-time using autoencoders implemented on edge devices is exceedingly challenging due to limited hardware, energy, and computational resources. We show that these limitations can be addressed by designing an autoencoder with low-resolution non-volatile memory-based synapses and employing an effective quantized neural network learning algorithm. We further propose nanoscale ferromagnetic racetracks with engineered notches hosting magnetic domain walls (DW) as exemplary non-volatile memory-based autoencoder synapses, where limited state (5-state) synaptic weights are manipulated by spin orbit torque (SOT) current pulses to write different magnetoresistance states. The performance of anomaly detection of the proposed autoencoder model is evaluated on the NSL-KDD dataset. Limited resolution and DW device stochasticity aware training of the autoencoder is performed, which yields comparable anomaly detection performance to the autoencoder having floating-point precision weights. While the limited number of quantized states and the inherent stochastic nature of DW synaptic weights in nanoscale devices are typically known to negatively impact the performance, our hardware-aware training algorithm is shown to leverage these imperfect device characteristics to generate an improvement in anomaly detection accuracy (90.98%) compared to accuracy obtained with floating-point synaptic weights that are extremely memory intensive. Furthermore, our DW-based approach demonstrates a remarkable reduction of at least three orders of magnitude in weight updates during training compared to the floating-point approach, implying significant reduction in operation energy for our method. This work could stimulate the development of extremely energy efficient non-volatile multi-state synapse-based processors that can perform real-time training and inference on the edge with unsupervised data.

Digital versus manual pathology scoring of PD-L1 in patients with esophageal squamous cell carcinoma (ESCC).

e16023 Background: Immune checkpoint inhibitors (ICIs) are standard-of-care for patients with ESCC, but the association between PD-L1 and clinical outcomes of patients with ESCC receiving ICIs treatment is unclear. Currently, different PD-L1 antibody clones, scoring, and cutoffs were used for different ICIs development in ESCC. Thus, we develop a digital pathology (DP) algorithm to generate multiple scores at the same time with high reproducibility. We evaluate DP algorithm by: (1) comparing the digital and manual scores of SP263 and 22C3 assays; (2) evaluating the association with OS. Methods: Tumor tissue samples were collected from ESCC patients treated at National Taiwan University Hospital between 2017 and 2020. Slides were tested for PD-L1 using SP263 and 22C3 assays at a College of American Pathologists accredited central laboratory. H&E and PD-L1 slides were scanned for whole slide images (WSIs) using DP200 scanner at 20x. HALO (v3.5) by indica labs was used for DP development by deep learning neural network algorithms. WSIs of SP263 were annotated for region of interest (ROI), tumor/stroma area, and tumor, immune and fibroblast cells by board certified pathologists. After the algorithm is developed and run on SP263 WSIs, a board-certified pathologist reviewed the images and provided an agreement score of DP markups. The images were excluded if an agreement was less than 80%. This algorithm was then applied to 22C3 WSIs. Data of areas and cells were exported for subsequent statistical analysis using R. We calculated the DP scores based on the definition of manual readouts of TAP, CPS and s/iTILs in ROI and tumor and stroma regions in both clones’ WSIs. These DP scores were compared with manual scores using Spearman’s rank-order correlation. Association with OS were evaluated using cox proportional hazard regression. Results: Of 235 evaluable images, twelve (5.1%) were excluded due to the low agreement with the pathologist's judgment. DP generated scores showed a strong correlation with pathologists scoring (rho = 0.80 for SP263 and rho = 0.74 for 22C3). DP generated CPS-like and TAP-like scores in ROI, and tumor and stroma regions from SP263 and 22C3 also demonstrated a strong correlation between each other (rho: 0.77 - 0.86). However, s/iTILs correlation between DP and pathologist score is negligible (rho: 0.002 - 0.34). While examining the association of PD-L1 scores and s/iTILs with OS, no clear associations were found with either DP or pathologists’ scores, after adjusting for stage, ECOG, age and gender. Conclusions: DP algorithms developed in this study demonstrated the efficiency of generating multiple DP scores of CPS, TAP, and s/iTILs from one image at the same time, which performed comparable to manual scores. DP algorithm developed based on SP263 is generalizable to 22C3 and showed a good performance. Further algorithm adjustment will be made to overcome the misclassification in the 12 failed cases.

Neural Networks in Forensic Expertology and Expert Practice: Problems and Prospects

The article, from the perspective of the theory the forensic activity digitalization as a particular theory of forensic expertise, examines the prospects for the introduction of neural networks in forensic examination and the current problems arising in this case. The author notes changes in the methodology and technologies for developing expert techniques in connection with the introduction of artificial intelligence algorithms — neural networks. The areas neural networks’ application for solving various problems of science and practical activity are outlined. Specific examples demonstrate the possibilities of using supervised learning algorithms for neural networks in forensic practice. A detailed analysis of the reasons why the use of neural networks in forensic science can lead to erroneous conclusions is given. Particular attention is paid to hallucinations of deep learning neural networks on large language models. There is a danger that an expert, relying entirely on a neural network, may give the wrong conclusion, since self-learning generative neural networks do not provide an explanation for why they made a particular decision. To develop expert methods for solving typical expert problems based on neural networks, it is proposed to create databases (Dataset) for various forensic objects for analysis and machine learning. To store the Dataset, it is necessary to organize repositories that can contain data sets on types (kinds) of forensic examinations. Dataset and repositories will provide data quality control and model verification. The article substantiates the need for new competencies of a Data Scientist — a specialist who develops tools for solving forensic problems when introducing neural networks and other artificial intelligence algorithms into forensic science, as well as a machine learning engineer working in contact with him.

Syngas composition analysis for waste to methanol production: Techno-economic assessment using machine learning and Aspen plus

The yield of methanol is heavily influenced by the composition of the syngas which varies from waste type and gasification conditions. This study uniquely explores machine learning (ML) and artificial neural network (ANN) algorithms with shapley additive values (SHAP) analysis to predict syngas compositions and interpret the complexities of waste gasification. Aspen Plus is then utilized to investigate the impact of syngas compositions on the economics of methanol production. Among the different ML models, gradient boosting regression achieved remarkable accuracy in predicting carbon monoxide and carbon dioxide, while ANN excelled in predicting nitrogen and hydrogen based on the coefficient of determination (>0.95). The feature importance analysis demonstrated the gasification agent type as the most important feature. Steam gasification of waste sawdust, straw, and wood chips showed the highest methanol yield. Economic analysis indicated the lowest levelized cost of methanol (LCMeOH) to be 1195.79 USD/MT which further decreased to 933.97 USD/MT upon hydrogen addition. Uncertainty analysis using Monte Carlo method showed the possible minimum LCMeOH to be 254.05 USD/MT. This study provides a comprehensive picture of the economic viability of waste-to-methanol production and provides an important reference for stakeholders on a global scale to quickly assess the feasibility of waste-to-energy projects.

Threshold learning algorithm for memristive neural network with binary switching behavior

On-chip learning is an effective method for adjusting artificial neural networks in neuromorphic computing systems by considering hardware intrinsic properties. However, it faces challenges due to hardware nonidealities, such as the nonlinearity of potentiation and depression and limitations on fine weight adjustment. In this study, we propose a threshold learning algorithm for a variation-tolerant ternary neural network in a memristor crossbar array. This algorithm utilizes two tightly separated resistance states in memristive devices to represent weight values. The high-resistance state (HRS) and low-resistance state (LRS) defined as read current of < 0.1 μA and > 1 μA, respectively, were successfully programmed in a 32 × 32 crossbar array, and exhibited half-normal distributions due to the programming method. To validate our approach experimentally, a 64 × 10 single-layer fully connected network were trained in the fabricated crossbar for an 8 × 8 MNIST dataset using the threshold learning algorithm, where the weight value is updated when a gradient determined by backpropagation exceeds a threshold value. Thanks to the large margin between the two states of the memristor, we observed only a 0.42 % drop in classification accuracy compared to the baseline network results. The threshold learning algorithm is expected to alleviate the programming burden and be utilized in variation-tolerant neuromorphic architectures.

Intelligent Lighting Control Systems for Energy Savings in Hospital Buildings Using Artificial Neural Networks

Lighting control systems are essential in modern building automation and smart homes, efficiently managing illumination to enhance energy conservation and user comfort. This project tackles energy consumption challenges in hospital buildings by introducing Intelligent Lighting Control Systems (ILCS) that take natural light and occupancy into account, driven by Artificial Neural Networks (ANN) and diverse machine learning algorithms. In our study, we collected sensor data, processed it, and designed a lighting control system employing a feedforward neural network and various machine learning algorithms. Surprisingly, our research found that a linear regression algorithm surpassed the ANN-based system in this context. We implemented a prototype, tested it on hardware, and obtained the expected results. This research marks progress towards optimizing energy use in hospital buildings and contributing to sustainability endeavors. By combining ILCS and machine learning, it offers a promising approach for more efficient and eco-friendly lighting systems

Boosting Clear Cell Renal Carcinoma-Specific Drug Discovery Using a Deep Learning Algorithm and Single-Cell Analysis.

Clear cell renal carcinoma (ccRCC), the most common subtype of renal cell carcinoma, has the high heterogeneity of a highly complex tumor microenvironment. Existing clinical intervention strategies, such as target therapy and immunotherapy, have failed to achieve good therapeutic effects. In this article, single-cell transcriptome sequencing (scRNA-seq) data from six patients downloaded from the GEO database were adopted to describe the tumor microenvironment (TME) of ccRCC, including its T cells, tumor-associated macrophages (TAMs), endothelial cells (ECs), and cancer-associated fibroblasts (CAFs). Based on the differential typing of the TME, we identified tumor cell-specific regulatory programs that are mediated by three key transcription factors (TFs), whilst the TF EPAS1/HIF-2α was identified via drug virtual screening through our analysis of ccRCC's protein structure. Then, a combined deep graph neural network and machine learning algorithm were used to select anti-ccRCC compounds from bioactive compound libraries, including the FDA-approved drug library, natural product library, and human endogenous metabolite compound library. Finally, five compounds were obtained, including two FDA-approved drugs (flufenamic acid and fludarabine), one endogenous metabolite, one immunology/inflammation-related compound, and one inhibitor of DNA methyltransferase (N4-methylcytidine, a cytosine nucleoside analogue that, like zebularine, has the mechanism of inhibiting DNA methyltransferase). Based on the tumor microenvironment characteristics of ccRCC, five ccRCC-specific compounds were identified, which would give direction of the clinical treatment for ccRCC patients.

Hybrid hyperplane gradient learning algorithm for RBF neural network

Gradient-based algorithms are still popularly used for training radial basis function neural network (RBFNN). However, these algorithms may lead to the vanishing gradient (saddle point or local minimum) problem, which is one of the common problems that limit learning performance. To cope with this, a hybrid hyperplane gradient learning algorithm (HHGLA) is proposed to improve the learning performance of the RBFNN in this paper. First, a hyperplane gradient (HPG), based on a hyperspace constructed by a solution population (SP), is introduced. Thus, the search process can cross the surface of the cost function to avoid the appearance of vanishing gradient. Second, an adaptive learning rate is designed to restrict the search process into this hyperspace. Then, during the learning process, the hyperspace is constantly contracted to approximate the global optimal solution. Third, the convergence property of HHGLA-based RBFNN is thoroughly analyzed to ensure its application success. Finally, empirical comparisons of the proposed HHGLA-RBFNN are given to illustrate its superiority. The simulation results show the feasibility of HHGLA-RBFNN of accuracy.

Understanding and Improving Optimization in Predictive Coding Networks

Backpropagation (BP), the standard learning algorithm for artificial neural networks, is often considered biologically implausible. In contrast, the standard learning algorithm for predictive coding (PC) models in neuroscience, known as the inference learning algorithm (IL), is a promising, bio-plausible alternative. However, several challenges and questions hinder IL's application to real-world problems. For example, IL is computationally demanding, and without memory-intensive optimizers like Adam, IL may converge to poor local minima. Moreover, although IL can reduce loss more quickly than BP, the reasons for these speedups or their robustness remains unclear. In this paper, we tackle these challenges by 1) altering the standard implementation of PC circuits to substantially reduce computation, 2) developing a novel optimizer that improves the convergence of IL without increasing memory usage, and 3) establishing theoretical results that help elucidate the conditions under which IL is sensitive to second and higher-order information.

Risk prediction and analysis of gallbladder polyps with deep neural network

The aim of this study is to analyze the risk factors associated with the development of adenomatous and malignant polyps in the gallbladder. Adenomatous polyps of the gallbladder are considered precancerous and have a high likelihood of progressing into malignancy. Preoperatively, distinguishing between benign gallbladder polyps, adenomatous polyps, and malignant polyps is challenging. Therefore, the objective is to develop a neural network model that utilizes these risk factors to accurately predict the nature of polyps. This predictive model can be employed to differentiate the nature of polyps before surgery, enhancing diagnostic accuracy. A retrospective study was done on patients who had cholecystectomy surgeries at the Department of Hepatobiliary Surgery of the Second People’s Hospital of Shenzhen between January 2017 and December 2022. The patients’ clinical characteristics, lab results, and ultrasonographic indices were examined. Using risk variables for the growth of adenomatous and malignant polyps in the gallbladder, a neural network model for predicting the kind of polyps will be created. A normalized confusion matrix, PR, and ROC curve were used to evaluate the performance of the model. In this comprehensive study, we meticulously analyzed a total of 287 cases of benign gallbladder polyps, 15 cases of adenomatous polyps, and 27 cases of malignant polyps. The data analysis revealed several significant findings. Specifically, hepatitis B core antibody (95% CI −0.237 to 0.061, p < 0.001), number of polyps (95% CI −0.214 to −0.052, p = 0.001), polyp size (95% CI 0.038 to 0.051, p < 0.001), wall thickness (95% CI 0.042 to 0.081, p < 0.001), and gallbladder size (95% CI 0.185 to 0.367, p < 0.001) emerged as independent predictors for gallbladder adenomatous polyps and malignant polyps. Based on these significant findings, we developed a predictive classification model for gallbladder polyps, represented as follows, Predictive classification model for GBPs = −0.149 * core antibody − 0.033 * number of polyps + 0.045 * polyp size + 0.061 * wall thickness + 0.276 * gallbladder size − 4.313. To assess the predictive efficiency of the model, we employed precision-recall (PR) and receiver operating characteristic (ROC) curves. The area under the curve (AUC) for the prediction model was 0.945 and 0.930, respectively, indicating excellent predictive capability. We determined that a polyp size of 10 mm served as the optimal cutoff value for diagnosing gallbladder adenoma, with a sensitivity of 81.5% and specificity of 60.0%. For the diagnosis of gallbladder cancer, the sensitivity and specificity were 81.5% and 92.5%, respectively. These findings highlight the potential of our predictive model and provide valuable insights into accurate diagnosis and risk assessment for gallbladder polyps. We identified several risk factors associated with the development of adenomatous and malignant polyps in the gallbladder, including hepatitis B core antibodies, polyp number, polyp size, wall thickness, and gallbladder size. To address the need for accurate prediction, we introduced a novel neural network learning algorithm. This algorithm utilizes the aforementioned risk factors to predict the nature of gallbladder polyps. By accurately identifying the nature of these polyps, our model can assist patients in making informed decisions regarding their treatment and management strategies. This innovative approach aims to improve patient outcomes and enhance the overall effectiveness of care.