Diagnosis System Research Articles

Spatial variations in health service utilization among migrant population: a perspective on health equity.

As health equity becomes a prioritized goal in global health policy, extensive research has revealed that socio-economic and geographical factors jointly exacerbate barriers to medical service access for both internal and international migrant populations, further accelerating existing health disparities. This study explores healthcare service utilization disparities among internal migrants in China, a population profoundly affected by the country's economic reforms and urbanization since the late 1970s. These transformations have led to significant migratory movements and subsequent healthcare challenges for these populations. Leveraging data from the 2017 China Migrant Dynamic Survey, comprising 169,989 samples across 28 provinces, we introduce a novel metric-the "No Treatment ratio" (NT-ratio). This ratio quantifies the proportion of migrants who, after falling ill, choose not to seek treatment relative to the total migrant population in a given province or region, serving as a critical measure of health risk. Building upon Anderson's Behavioral Model of Health Services Use, we adapted the model to better reflect the unique circumstances of migrant populations. The study employs spatial autocorrelation, hotspot analysis, and geodetector techniques to dissect the multifaceted factors influencing healthcare disparities. Our Findings reveal that the NT-ratio is significantly higher in eastern and northeastern China. Key factors influencing the NT-ratio include age, left-behind experiences, health education, and per capita medical resources. In response to these disparities, we recommend optimizing the distribution of medical resource, strengthening tiered diagnosis and treatment systems, and integrating health, education, and social security resources. These measures aim to improve healthcare utilization among migrant populations and reduce health inequities, aligning with global health objectives.

A Modified Deep Semantic Segmentation Model for Analysis of Whole Slide Skin Images

Automated segmentation of biomedical image has been recognized as an important step in computer-aided diagnosis systems for detection of abnormalities. Despite its importance, the segmentation process remains an open challenge due to variations in color, texture, shape diversity and boundaries. Semantic segmentation often requires deeper neural networks to achieve higher accuracy, making the segmentation model more complex and slower. Due to the need to process a large number of biomedical images, more efficient and cheaper image processing techniques for accurate segmentation are needed. In this article, we present a modified deep semantic segmentation model that utilizes the backbone of EfficientNet-B3 along with UNet for reliable segmentation. We trained our model on Non-melanoma skin cancer segmentation for histopathology dataset to divide the image in 12 different classes for segmentation. Our method outperforms the existing literature with an increase in average class accuracy from 79 to 83%. Our approach also shows an increase in overall accuracy from 85 to 94%.

Empowering Brain Tumor Diagnosis through Explainable Deep Learning

Brain tumors are among the most lethal diseases, and early detection is crucial for improving patient outcomes. Currently, magnetic resonance imaging (MRI) is the most effective method for early brain tumor detection due to its superior imaging quality for soft tissues. However, manual analysis of brain MRI scans is prone to errors, largely influenced by the radiologists’ experience and fatigue. To address these challenges, computer-aided diagnosis (CAD) systems are more significant. These advanced computer vision techniques such as deep learning provide accurate predictions based on medical images, enhancing diagnostic precision and reliability. This paper presents a novel CAD framework for multi-class brain tumor classification. The framework employs six pre-trained deep learning models as the base and incorporates comprehensive data preprocessing and augmentation strategies to enhance computational efficiency. To address issues related to transparency and interpretability in deep learning models, Gradient-weighted Class Activation Mapping (Grad-CAM) is utilized to visualize the decision-making processes involved in tumor classification from MRI scans. Additionally, a user-friendly Brain Tumor Detection System has been developed using Streamlit, demonstrating its practical applicability in real-world settings and providing a valuable tool for clinicians. All simulation results are derived from a public benchmark dataset, showing that the proposed framework achieves state-of-the-art performance, with accuracy approaching 99% in ResNet-50, Xception, and InceptionV3 models.

Improving the diagnostic strategy for thyroid nodules: a combination of artificial intelligence-based computer-aided diagnosis system and shear wave elastography.

Thyroid nodules are highly prevalent in the general population, posing a clinical challenge in accurately distinguishing between benign and malignant cases. This study aimed to investigate the diagnostic performance of different strategies, utilizing a combination of a computer-aided diagnosis system (AmCAD) and shear wave elastography (SWE) imaging, to effectively differentiate benign and malignant thyroid nodules in ultrasonography. A total of 126 thyroid nodules with pathological confirmation were prospectively included in this study. The AmCAD was utilized to analyze the ultrasound imaging characteristics of the nodules, while the SWE was employed to measure their stiffness in both transverse and longitudinal thyroid scans. Twelve diagnostic patterns were formed by combining AmCAD diagnosis and SWE values, including isolation, series, parallel, and integration. The diagnostic performance was assessed using the receiver operating characteristic curve and area under the curve (AUC). Sensitivity, specificity, accuracy, missed malignancy rate, and unnecessary biopsy rate were also determined. Various diagnostic schemes have shown specific advantages in terms of diagnostic performance. Overall, integrating AmCAD with SWE imaging in the transverse scan yielded the most favorable diagnostic performance, achieving an AUC of 72.2% (95% confidence interval (CI): 63.0-81.5%), outperforming other diagnostic schemes. Furthermore, in the subgroup analysis of nodules measuring <2 cm or 2-4 cm, the integrated scheme consistently exhibited promising diagnostic performance, with AUCs of 74.2% (95% CI: 61.9-86.4%) and 77.4% (95% CI: 59.4-95.3%) respectively, surpassing other diagnostic schemes. The integrated scheme also effectively addressed thyroid nodule management by reducing the missed malignancy rate to 9.5% and unnecessary biopsy rate to 22.2%. The integration of AmCAD and SWE imaging in the transverse thyroid scan significantly enhances the diagnostic performance for distinguishing benign and malignant thyroid nodules. This strategy offers clinicians the advantage of obtaining more accurate clinical diagnoses and making well-informed decisions regarding patient management.

BwMMV-pred: a novel ensemble learning approach using blood smear images for malaria prediction

The use of machine learning in healthcare has become widespread, enhancing the capabilities of doctors and clinicians. This study introduces a novel ensemble learning approach named Blending with Meta Majority Voting (BwMMV) for malaria prediction using blood smear images. The BwMMV technique combines the strengths of eight base classifiers to form an intermediate dataset, which is subsequently used to train five distinct meta-models using different machine learning algorithms. A Local Binary Pattern Histogram (LBPH) method is employed to extract texture features from blood smear images, effectively capturing the underlying patterns necessary for classification. The final classification decision is determined through a majority voting mechanism, selecting the outcome with the most votes as the final prediction. Our results indicate that the BwMMV approach significantly outperforms traditional hard voting and blending techniques, achieving superior accuracy, robustness, and resilience in performance. This innovative method demonstrates promising potential as a powerful tool for automated diagnosis systems, with the ability to be expanded to analyze various datasets efficiently.

Evaluation of the quickmic system in the rapid diagnosis of Gram-negative bacilli bacteremia.

The rapid diagnosis of antibiotic sensitivity in Gram-negative bacilli is of paramount importance in clinical microbiology, particularly in cases of bacteremia. The escalating challenge of antimicrobial resistance underscores the need for expeditious antibiotic susceptibility diagnostics to guide empirical antibiotic therapy effectively. In light of this, we present our study that evaluates the QuickMIC System, a recently developed rapid diagnostic antibiogram. QuickMIC System, offers a novel approach to phenotypic testing, providing information on the activity of 12 antibiotics against key pathogens, including Escherichia coli, Klebsiella spp., Pseudomonas aeruginosa, Acinetobacter baumannii, Enterobacter cloacae, Proteus spp., Citrobacter spp., and Serratia marcescens. Our investigation involved testing a total of 816 antibiotic/microorganism combinations. The study demonstrated an impressive 99.02% concordance between the QuickMIC System and the reference methods, with only eight discrepancies observed. The time to actionable minimum inhibitory concentration (MIC) ranged between 2 and 4 h, highlighting the system's efficiency in providing rapid results.

Multi-label semantic segmentation of magnetic resonance images of the prostate gland

Prostate segmentation is a substantial factor in the diagnostic pathway of suspicious prostate lesions. Medical doctors are assisted by computer-aided detection and diagnosis systems systems and methods derived from artificial intelligence deep learning-based systems have to be trained on existing data. Especially freely available labeled prostate magnetic resonance imaging data is rare. With regard to this problem, we show a method to combine two existing small datasets to form a bigger one. We present a data processing pipeline consisting of a cascaded network architecture that is able to perform multi-label semantic segmentation, hence, capable of classifying each pixel in a T2\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\ extrm{T}_{2}$$\\end{document}-weighted image not only to one class but to a subset of the prostate zones and classes of interest. This delivers richer information such as overlapping zones that are key to medical radiological examination. Additionally, we describe how to integrate expert knowledge in our deep learning system. To increase data variety for training and evaluation we use image augmentation on our two datasets—a freely available dataset and our new open-source dataset. To combine the datasets we denoise the contourings in our dataset by using an effective yet simple algorithm based on standard computer vision methods only. The performance of the presented methodology is compared and evaluated using the dice score metric and 5-fold cross-validation on all datasets. Although we trained on tiny datasets our method achieves excellent segmentation quality and is even able to detect prostate cancer. Our method to combine the two datasets reduces segmentation errors and increases data variety. The proposed architecture significantly improves performance by including expert knowledge via feature-map concatenation. On the initiative for collaborative computer vision benchmarking dataset we achieve on average dice scores of approximately 91% for the whole prostate gland, 67% for the peripheral zone and 75% for the prostate central gland. We find that image augmentation except contrast limited adaptive histogram equalisation did not have much influence on the segmentation quality. Derived and enhanced from existing methods we present an approach that is able to deliver multi-label semantic segmentation results for prostate magnetic resonance imaging. This approach is simple and could be applied to other applications of deep learning as well. It improves the segmentation results by a large margin. Once tweaked to the data, our denoising and combination algorithm delivers robust and accurate results even on data with segmentation errors.

Reducing neural network complexity via optimization algorithms for fault diagnosis in renewable energy systems

This article discusses the challenges involved in diagnosing faults in renewable energy systems. A significant challenge is the high computational demands of the artificial intelligence algorithms needed for grid-connected photovoltaic (GCPV) and wind energy conversion (WEC) systems in fault diagnosis. To address this issue, several methods are proposed to reduce computation time, minimize memory requirements, and improve efficiency. First, optimization algorithms such as the Salp Swarm Algorithm (SSA), Genetic Algorithm (GA), and Particle Swarm Optimization (PSO), combined with machine learning classifiers for feature selection, are suggested to reduce computational time and memory space requirements. This approach notably decreases computation time but has a limited impact on memory space. Secondly, further reduction in memory space requirements is recommended by using the variogram method for data reduction. This method can leverage the outputs of preceding algorithms to optimize renewable energy systems, making their operations cost-effective and efficient. Finally, the outputs of the variogram are used to train neural network (NN), recurrent neural network (RNN), and long short-term memory (LSTM) classifiers to differentiate between various modes of operation in GCPV and WEC systems. Experimental results demonstrate the effectiveness and robustness of the proposed methods, showing that memory space can be reduced by 1.3 to 5.6 times and CPU time by 1.1 to 11 times while maintaining accuracy and improving efficiency compared to conventional algorithms.

An Ensemble Learning Algorithm with Hyperparameter Optimization (ELAHO) Model for Early Detection of Maternal Health Risk (MHR) Level Using Machine Learning

Despite the advancement in the healthcare system, maternal health risk remains high, which is the most challenging aspect nowadays. There is a need to develop an effective model for early detection, monitoring, and prediction of maternal health risk levels during pregnancy. The machine learning intelligence model has proven its effectiveness and robustness in providing accurate and reliable prediction, analysis, and interpretation of medical data, reducing several risk factors for early diagnosis in healthcare. In this research work, we proposed an ensemble learning algorithm with a hyperparameter optimization (ELAHO) model using machine learning algorithms to improve its robustness, effectiveness, and model performance. The proposed method uses a hybrid model of logistic regression and support vector machines (LG-SVM) to predict maternal health risk levels during pregnancy. The method utilized Python software for training, testing, and validation. We evaluated the performance of our proposed model using accuracy, precision, sensitivity, f1-score, and ROC-AUC score. The proposed models outperformed the conventional models and achieved 100% predictive accuracy. The proposed approach has the potential to be adapted as an intelligence-monitoring system for early medical diagnosis during pregnancy. The proposed techniques will help medical professionals make quick decisions accurately and enhance monitoring to improve the level of care offered to pregnant mothers and their unborn children.

Intersystem and Interoperator Agreement of US Attenuation Coefficient for Quantifying Liver Steatosis.

Background The extent of liver steatosis can be assessed using US attenuation coefficient (AC) algorithms currently implemented in several US systems. However, little is known about intersystem and interoperator variability in measurements. Purpose To assess intersystem and interoperator agreement in US AC measurements for fat quantification in individuals with varying degrees of liver steatosis and to assess the correlation of each manufacturer's AC algorithm results with MRI proton density fat fraction (PDFF). Materials and Methods This prospective study was conducted at Southwoods Imaging, Youngstown, Ohio, September 30-October 1, 2023. Two operators independently obtained AC measurements using eight US systems equipped with an AC algorithm from different manufacturers. On the same day, MRI PDFF measurement was performed by a different operator. Correlation between US AC and MRI PDFF was assessed using a mixed-effects model. Agreement between systems and operators was evaluated using the intraclass correlation coefficient (ICC). Results Twenty-six individuals (mean age, 55.4 years ± 10.7 [SD]; 16 female participants) were evaluated. The correlation of US AC with MRI PDFF was high for five AC algorithms (r range, 0.70-0.86), moderate for two (r = 0.62 for both), and poor for one (r = 0.47). In pairwise comparisons, none of the pairs of systems achieved excellent agreement (overall ICC = 0.33 [95% CI: 0.15, 0.52]). One pair showed good agreement (ICC = 0.79 [95% CI: 0.66, 0.87]), eight pairs showed moderate agreement (ICC range, 0.50 [95% CI: 0.22, 0.69] to 0.73 [95% CI: 0.49, 0.85]), and 19 pairs showed poor agreement (ICC range, 0.11 [95% CI: -0.06, 0.37] to 0.48 [95% CI: 0.20, 0.67]). Interoperator agreement on AC value was excellent for the Samsung Medison algorithm (ICC = 0.90 [95% CI: 0.80, 0.96]), good for the Siemens Healthineers (ICC = 0.76 [95% CI: 0.54, 0.89]) and Canon Medical Systems (ICC = 0.76 [95% CI: 0.16, 0.92]) algorithms, and moderate for the remaining algorithms (ICC range, 0.50 [95% CI: 0.16, 0.73] to 0.74 [95% CI: 0.51, 0.88]). The mean AC value obtained by the two operators did not differ for any system except the system from Canon Medical Systems. Conclusion There was substantial variability in AC values obtained with different US systems, precluding interchangeability between systems for liver steatosis diagnosis and follow-up imaging. Interoperator agreement ranged from moderate to excellent. © RSNA, 2024 Supplemental material is available for this article. See also the editorial by Han in this issue.

FocalNeXt: A ConvNeXt augmented FocalNet architecture for lung cancer classification from CT-scan images

Early and accurate diagnosis of lung cancer, a life-threatening disease, is critical to the successful treatment of patients with the disease. On the other hand, it is well known that the integration of computer-aided diagnosis (CAD) systems into the diagnostic workflow facilitates the process and improves the diagnostic results of automated systems by reducing the difficulties associated with human observation. In this paper, we present FocalNeXt, a new ConvNeXt-augmented FocalNet architecture specifically designed for automated lung cancer detection from computed tomography (CT) scan images. By combining the strong attention mechanism of FocalNet and the feature extraction mechanism of ConvNeXt within the vision transformer paradigm, FocalNeXt aims to improve diagnostic accuracy. Our comprehensive evaluation of a publicly available Iraq-Oncology Teaching Hospital/National Center for Cancer Diseases (IQ-OTH/NCCD) CT-scan dataset includes rigorous comparisons and an ablation study. FocalNeXt achieved an accuracy of 99.81%, outperforming the state-of-the-art methods. The model also excelled in sensitivity (99.78%), recall (99.36%), and F1-score (99.56%), positioning FocalNeXt as a leading model for lung cancer detection. The ablation study further demonstrated its efficacy and underscored the robustness of FocalNeXt in different configurations. The results underline its potential to contribute to advances in medical imaging and personalized healthcare.

Improving the endoscopic recognition of early colorectal carcinoma using artificial intelligence: current evidence and future directions.

Background and study aims Artificial intelligence (AI) has great potential to improve endoscopic recognition of early stage colorectal carcinoma (CRC). This scoping review aimed to summarize current evidence on this topic, provide an overview of the methodologies currently used, and guide future research. Methods A systematic search was performed following the PRISMA-Scr guideline. PubMed (including Medline), Scopus, Embase, IEEE Xplore, and ACM Digital Library were searched up to January 2024. Studies were eligible for inclusion when using AI for distinguishing CRC from colorectal polyps on endoscopic imaging, using histopathology as gold standard, reporting sensitivity, specificity, or accuracy as outcomes. Results Of 5024 screened articles, 26 were included. Computer-aided diagnosis (CADx) system classification categories ranged from two categories, such as lesions suitable or unsuitable for endoscopic resection, to five categories, such as hyperplastic polyp, sessile serrated lesion, adenoma, cancer, and other. The number of images used in testing databases varied from 69 to 84,585. Diagnostic performances were divergent, with sensitivities varying from 55.0% to 99.2%, specificities from 67.5% to 100% and accuracies from 74.4% to 94.4%. Conclusions This review highlights that using AI to improve endoscopic recognition of early stage CRC is an upcoming research field. We introduced a suggestions list of essential subjects to report in research regarding the development of endoscopy CADx systems, aiming to facilitate more complete reporting and better comparability between studies. There is a knowledge gap regarding real-time CADx system performance during multicenter external validation. Future research should focus on development of CADx systems that can differentiate CRC from premalignant lesions, while providing an indication of invasion depth.

Comprehensive Analysis of Faults and Diagnosis Techniques in Cascaded Multi-Level Inverters

Reliability is a crucial factor to consider for multi-level inverters (MLIs) used in industrial applications. With the increasing number of power semiconductor devices, the potential for defects to significantly degrade the overall system is heightened. A highly effective fault-detection technique is required to minimize the impact of faults. This paper provides a comprehensive overview of the fundamental principles of multi-level inverters and the various sorts of faults that can occur in multi-level inverters. This study provides a comprehensive analysis of five-level cascaded H-bridge multilevel inverters (MLIs) under both normal and defective conditions. The paper outlines a fault-detection method that utilizes total harmonic distortion and a normalized output voltage factor. In addition, the paper discusses a fault-isolation strategy that relies on reducing amplitude modulation. This method leads to the development of a fault-tolerant inverter. The utilization of level-shifted pulse-width modulation (LSPWM) technology is employed for the purpose of switching operations. LSPWM is the most appropriate technique for MLIs that require a low amount of computational resources. The fault-diagnosis approach given is suitable for MLI-based drives, grid-connected operations, and other applications. This paper presents a comprehensive examination of the 5L-CMLI (5-Level Cascaded Multi-Level Inverter) under various fault scenarios in CMLI. Subsequently, various fault diagnosis approaches will be examined, including their advantages and disadvantages. The paper discusses several defects that can occur in the Insulated Gate Bipolar Transistor (IGBT) of a Current Mode Logic Inverter (CMLI), and also presents a design for a reliable fault diagnosis system. Furthermore, this analysis examines several fault detection strategies in CMLI, categorized according to open-loop and closed-loop dynamic systems fault classifications.

Fault isolation considering real mechanical coupling effects with application to liquid oxygen/kerosene staged combustion rocket engines

Liquid rocket engines are crucial for space exploration, necessitating reliable fault diagnosis systems. This paper focuses on fault isolation in liquid oxygen (LOX)/kerosene staged combustion rocket engines. Addressing deficiencies in current methods, we first propose a theorem to evaluate the adaptability between the analytical model and fault factors based on available sensors. Through the analysis of fault mechanisms, this paper defines fault modes using an innovative multi-factor coupling approach based on engine operational principles. Moreover, a fault isolation method with two steps is established, incorporating the optimal selection of sensors to accommodate robust requirements. This method features a clear process, low computational complexity, and comprehensive fault coverage, and it can also be applied to systems with similar complex operational mechanisms.

EP.04B.06 High-Resolution CT with 1024-Matrix for AI-Assisted Diagnosis System in the Evaluation of Pulmonary Nodules

EP.04B.06 High-Resolution CT with 1024-Matrix for AI-Assisted Diagnosis System in the Evaluation of Pulmonary Nodules

RFMiD: Retinal Image Analysis for multi-Disease Detection challenge

In the last decades, many publicly available large fundus image datasets have been collected for diabetic retinopathy, glaucoma, and age-related macular degeneration, and a few other frequent pathologies. These publicly available datasets were used to develop a computer-aided disease diagnosis system by training deep learning models to detect these frequent pathologies. One challenge limiting the adoption of a such system by the ophthalmologist is, computer-aided disease diagnosis system ignores sight-threatening rare pathologies such as central retinal artery occlusion or anterior ischemic optic neuropathy and others that ophthalmologists currently detect. Aiming to advance the state-of-the-art in automatic ocular disease classification of frequent diseases along with the rare pathologies, a grand challenge on “Retinal Image Analysis for multi-Disease Detection” was organized in conjunction with the IEEE International Symposium on Biomedical Imaging (ISBI - 2021). This paper, reports the challenge organization, dataset, top-performing participants solutions, evaluation measures, and results based on a new “Retinal Fundus Multi-disease Image Dataset” (RFMiD). There were two principal sub-challenges: disease screening (i.e. presence versus absence of pathology — a binary classification problem) and disease/pathology classification (a 28-class multi-label classification problem). It received a positive response from the scientific community with 74 submissions by individuals/teams that effectively entered in this challenge. The top-performing methodologies utilized a blend of data-preprocessing, data augmentation, pre-trained model, and model ensembling. This multi-disease (frequent and rare pathologies) detection will enable the development of generalizable models for screening the retina, unlike the previous efforts that focused on the detection of specific diseases.

S372 Evaluating the Performance of a Computer-Aided Diagnosis System in Implementing Diagnose-and-Leave and Resect-and-Discard Strategies for Diminutive Colorectal Polyps: A Real-World Pragmatic Study

S372 Evaluating the Performance of a Computer-Aided Diagnosis System in Implementing Diagnose-and-Leave and Resect-and-Discard Strategies for Diminutive Colorectal Polyps: A Real-World Pragmatic Study

Design of intelligent remote monitoring and diagnosis system for equipment

Design of intelligent remote monitoring and diagnosis system for equipment

Self-adaptive fault diagnosis for unseen working conditions based on digital twins and domain generalization

In recent years, intelligent fault diagnosis based on domain adaptation has been used to address domain shifts in cyber–physical systems; however, the need for acquiring target data sufficiently limits their applicability to unseen working conditions. To overcome such limitations, domain generalization techniques have been introduced to enhance the capacity of fault diagnostic models to operate under unseen working conditions. Nevertheless, existing approaches assume access to extensive labeled training data from various source domains, posing challenges in real-world engineering scenarios due to resource constraints. Moreover, the absence of a mechanism for updating diagnostic models over time calls for the exploration of self-adaptive generalized diagnosis models that are capable of autonomous reconfiguration in response to new unseen working conditions. In such a context, this paper proposes a self-adaptive fault diagnosis system that combines several paradigms, namely Monitor-Analyze-Plan-Execute over a shared Knowledge (MAPE-K), Domain Generalization Network Models (DGNMs), and Digital Twins (DT). The MAPE-K loop enables run-time adaptation to dynamic industrial environments without human intervention. To address the scarcity of labeled training data, digital twins are used to generate supplementary data and continuously tune parameters to reflect the dynamics of new unseen working conditions. DGNM incorporates adversarial learning and a domain-based discrepancy metric to enhance feature diversity and generalization. The introduction of multi-domain data augmentation enhances feature diversity and facilitates learning correlations among multiple domains, ultimately improving the generalization of feature representations. The proposed fault diagnosis system has been evaluated on three publicly available rotating machinery datasets to demonstrate its higher performance in cross-work operation and cross-machine tasks compared to other state-of-the-art methods.

Study on Fault Prediction Method of Hydropower Equipment Based on Equipment Health State Detection

Abstract With the continuous development of China’s electric power industry, the installed capacity of power plants is increasing, and hydropower units are developing in the direction of large capacity and high efficiency. As the complexity of the equipment is getting higher and higher, different equipment are more and more different, showing different characteristics when faults occur, which puts forward higher requirements for the safe and stable operation of hydropower units. In response to the above problems, this paper develops a set of human-machine closed-loop remote fault diagnosis reasoning machine system with self-learning function and a set of equipment health state monitoring and prediction system. This remote fault diagnosis system reasoning machine system can reduce the cost of unit equipment fault diagnosis, improve the diagnosis efficiency, and ensure the safe, stable and economic operation of the unit.