Predictors Of Health Status Research Articles

Trends in voice characteristics in patients with heart failure (VENTURE) in Switzerland: Protocol for a longitudinal observational pilot study.

Heart Failure (HF) is a major health and economic issue worldwide. HF-related expenses are largely driven by hospital admissions and re-admissions, many of which are potentially preventable. Current self-management programs, however, have failed to reduce hospital admissions. This may be explained by their low predictive power for decompensation and high adherence requirements. Slight alterations in the voice profile may allow to detect decompensation in HF patients at an earlier stage and reduce hospitalizations. This pilot study investigates the potential of voice as a digital biomarker to predict health status deterioration in HF patients. In a two-month longitudinal observational study, we collect voice samples and HF-related quality-of-life questionnaires from 35 stable HF patients. Patients use our developed study application installed on a tablet at home during the study period. From the collected data, we use signal processing to extract voice characteristics from the audio samples and associate them with the answers to the questionnaire data. The primary outcome will be the correlation between voice characteristics and HF-related quality-of-life health status. The study was reviewed and approved by the Cantonal Ethics Committee Zurich (BASEC ID:2022-00912). Results will be published in medical and technical peer-reviewed journals.

Digital Companion for Elders in Tracking Health and Intelligent Recommendation Support using Deep Learning

Ambient assisted living (AAL) facilitates the daily routines of elderly people, particularly those who have clinical difficulties or physical limitations. The latest technologies like distributed compuring,internet of things (IoT) and machine learning pave the ground for the creation of an effective automated tracker which aids elder citizens to live independently. The suggested system is attempted to design a wearable that monitors the blood glucose level through sweat. To achieve high accuracy, the proposed system uses ambient sensing and deep learning based techniques. It places a strong emphasis on calculating the health index by taking into account numerous disease-related characteristics or vitals such as heart rate, blood pressure, SpO2, blood glucose level, respiration rate, sweat rate, uric acid, and temperature. From the wearable device designed the vital signs are gathered, further environmental sensors and camera fixed around the person continually monitors the behavioral pattern along with physiological signals. This ensures the improved accuracy of health state prediction from its conventional models in place. The key advantage of this device is that it may be held and operated anyplace without interrupting their day-to-day tasks because the device is to be cheap, reliable and speedy.

Health Status Prediction of Lithium Battery Based on LSTM Model with Optimization Algorithms

Consider the current status of health (SOH) of lithium batteries, which presents challenging existing issues of accurately predicting and calculating. In this paper, an LSTM model and multi-optimization algorithm were used to estimate the battery health state. Taking advantage of the fast convergence speed and wide global optimization range of the optimization algorithm, optimized the number of layers and neurons in the LSTM model so the LSTM model was established, used to predict the health status of lithium batteries, and compared with the LSTM model prediction method without optimization. The results showed that the error of the battery health prediction model based on the proposed prediction model was less than 3%, the prediction accuracy was higher than the LSTM model without optimization, and the model had better accuracy and stability.

CPAE: Contrastive predictive autoencoder for unsupervised pre-training in health status prediction

Background and objectiveFully-supervised learning approaches have shown promising results in some health status prediction tasks using Electronic Health Records (EHRs). These traditional approaches rely on sufficient labeled data to learn from. However, in practice, acquiring large-scaled labeled medical data for various prediction tasks is often not feasible. Thus, it is of great interest to utilize contrastive pre-training to leverage the unlabeled information. MethodsIn this work, we propose a novel data-efficient framework, contrastive predictive autoencoder (CPAE), to first learn without labels from the EHR data in the pre-training process, and then fine-tune on the downstream tasks. Our framework comprises of two parts: (i) a contrastive learning process, inherited from contrastive predictive coding (CPC), which aims to extract global slow-varying features, and (ii) a reconstruction process, which forces the encoder to capture local features. We also introduce the attention mechanism in one variant of our framework to balance the above two processes. ResultsExperiments on real-world EHR dataset verify the effectiveness of our proposed framework on two downstream tasks (i.e., in-hospital mortality prediction and length-of-stay prediction), compared to their supervised counterparts, the CPC model, and other baseline models. ConclusionsBy comprising of both contrastive learning components and reconstruction components, CPAE aims to extract both global slow-varying information and local transient information. The best results on two downstream tasks are all achieved by CPAE. The variant AtCPAE is particularly superior when fine-tuned on very small training data. Further work may incorporate techniques of multi-task learning to optimize the pre-training process of CPAEs. Moreover, this work is based on the benchmark MIMIC-III dataset which only includes 17 variables. Future work may extend to a larger number of variables.

Predictive Models for Health Deterioration: Understanding Disease Pathways for Personalized Medicine.

Artificial intelligence (AI) and machine learning (ML) methods are currently widely employed in medicine and healthcare. A PubMed search returns more than 100,000 articles on these topics published between 2018 and 2022 alone. Notwithstanding several recent reviews in various subfields of AI and ML in medicine, we have yet to see a comprehensive review around the methods' use in longitudinal analysis and prediction of an individual patient's health status within a personalized disease pathway. This review seeks to fill that gap. After an overview of the AI and ML methods employed in this field and of specific medical applications of models of this type, the review discusses the strengths and limitations of current studies and looks ahead to future strands of research in this field. We aim to enable interested readers to gain a detailed impression of the research currently available and accordingly plan future work around predictive models for deterioration in health status. Expected final online publication date for the Annual Review of Biomedical Engineering, Volume 25 is June 2023. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.

A scalable and real-time system for disease prediction using big data processing

The growing chronic diseases patients and the centralization of medical resources cause significant economic impact resulting in hospital visits, hospital readmission, and other healthcare costs. This paper proposes a scalable and real-time system for disease prediction from medical data streams. This is carried out by integrating Twitter, Apache Kafka, Apache Spark and Apache Cassandra. Thus, Twitter users tweet attributes related to health, Kafka streaming receives all desired tweets attributes and ingest them to Spark streaming. Here, a machine learning algorithm is applied to predict health status and send back a response message through Kafka. The heart disease dataset, obtained from the UCI repository, was used for experiments. In order to enhance prediction accuracy, Relief algorithm is used for features selection. We compared sex types of relevant machine learning algorithms implemented by Spark MLlib such as Random Forest (RF), Naive Bayes, Support Vector Machine, Multilayer Perceptron, Decision Tree and Logistic Regression with the full features as well as selected features. The highest classification accuracy of 92.05% was reported using RF with selected features. The scalability of RF using Spark MLlib and WEKA framework for both training and application stages was measured. The results show significantly better performances of Spark in terms of scalability and computing times.

Predictors of self-reported general health status in people experiencing homelessness with serious mental illness.

The growing homeless population in the U.S.A. is disproportionately impacted by poor mental and physical health status, including a higher incidence of acute and chronic health problems, increased hospitalizations, and premature mortality compared to the general population. This study examined the association between demographic, social, and clinical factors and perceptions of general health status among the homeless population during admission to an integrated behavioral health treatment program. The study sample included 331 adults experiencing homelessness with a serious mental illness or co-occurring disorder. Participants were enrolled in services at a day program for unsheltered homeless adults, a residential substance use treatment program for males experiencing homelessness, a psychiatric step-down respite program for those experiencing homelessness following psychiatric hospitalization, permanent supportive housing for formerly chronically homeless adults, a faith-based food distribution program, and homeless encampment sites in a large urban area. Participants were interviewed using The Substance Abuse and Mental Health Services Administration's National Outcome Measures tool and a validated health-related quality of life measurement tool, SF-36. Data were examined using in elastic net regression. The study found seven factors to be particularly strong predictors of SF-36 general health scores.Male gender, "other" sexual identity, stimulant use, and Asian race were all associated with better perceptions of health status, while transgender status, inhalant use, and number of times arrested were associated with poorer perceptions. This study suggests targeted areas for health screening within the homeless population; however, more studies are necessary to demonstrate generalizability of the results.

Future Perspectives in Pediatric Dentistry: Where are We Now and where are We Heading?

Rodrigues JA, Olegario I, Assunção CM, et al. Future Perspectives in Pediatric Dentistry: Where are We Now and where are We Heading? Int J Clin Pediatr Dent 2022;15(6):793-797.

The Complexity of Online Patient Education Materials About Interventional Neuroradiology Procedures Published by Major Academic Institutions.

Introduction Health literacy is an independent predictor of population health status and is directly related to the readability of available patient education material. The National Institutes of Health (NIH) and the American Medical Association have recommended that patient education materials (PEMs) be written between a fourth- and a sixth-grade education level. The authors assessed the readability of online PEMs about neurointerventional procedures that have been published by several academic institutions across the US. Methods Online PEMs regarding five common neurointerventional procedures, including mechanical thrombectomy for large vessel occlusion, cerebral diagnostic angiography, carotid artery stenting, endovascular aneurysm embolization, and epidural steroid injection collected from the websites of 20 top institutions in Neurology and Neurosurgery. The materials were assessed via five readability scales and then were statistically analyzed and compared to non-institutional education websites (Wikipedia.com and WebMD.com). Results None of the PEMs were written at or below the NIH's recommended 6th-grade reading level. The average educational level required to comprehend the texts across all institutions, as assessed by the readability scales, was 10-11th grade level. Some materials required a college-level education or higher. Material from non-institutional websites had significantly lower readability scores compared to the 20 institutions. Conclusions Current PEMs related to neurointerventional procedures are not written at or below the NIH's recommended fourth- to sixth-grade education level. Given the complexity of those procedures, significant attention should be pointed toward an improvement in the available online materials.

Investigation of Machine Learning Methods for Predictive Maintenance of the Ultra-High-Pressure Reactor in a Polyethylene-Vinyl Acetate Production Process

Ethylene-Vinyl Acetate (EVA) copolymer was synthesized from ethylene and vinyl acetate at high temperatures and ultra-high pressures. In this condition, any reactor disturbances, such as process or mechanical faults, may trigger the run-away decomposition reaction. This paper proposes a procedure for constructing a conditional health status prediction structure that uses a virtual health index (HI) to monitor the reactor bearing’s remaining useful life (RUL). The piecewise linear remaining useful life (PL-RUL) model was constructed by machine learning regression methods trained on the vibration and distributed control system (DCS) datasets. This process consists of using Welch’s power spectrum density transformation and machine learning regression methods to fit the PL-RUL model, following a health status construction process. In this research, we search for and determine the optimum value for the remaining useful life period (TRUL), a key parameter for the PL-RUL model for the system, as 70 days. This paper uses four-fold cross-validation to evaluate seven different regression algorithms and concludes that the Extremely randomized trees (ERTs) is the best machine learning model for predicting PL-RUL, with an average relative absolute error (RAE) of 0.307 and a Linearity of 15.064. The Gini importance of the ensemble trees is used to identify the critical frequency bands and prepare them for additional dimensionality reduction. Compared to two frequency band selection techniques, the RAE and Linearity prediction results can be further improved to 0.22 and 8.38.

Environmental and Behavioral Factors in Association with Lung Function Impairment in Children Living in Wuhan, China.

Children’s lung function is a significant predictor of health status throughout their lifetime. This study aims to identify the prevalence of impaired lung function in children and the potential influencing factors in elementary school children of Wuhan, China. Children of 6−12 years old were enrolled from elementary schools in Wuhan, China, in 2018, on the basis of a cross-sectional study design. Information on personal behavior patterns and household characteristics, as well as parental factors, was collected based on a questionnaire survey. Spirometry was used to measure fifteen lung function indicators. Logistic regression models were used to estimate odds ratios of prevalence of impaired lung function with regard to socioeconomic, personal behavior patterns, household, and parental factors, respectively. Four lung function indicators such as FET and EVC in urban children had higher values than in the suburban children after adjusting for confounders (p < 0.05). A higher prevalence of impaired lung function (FEV6, FEV3, EVC, and VC) was found in the children from the rural area than in those from the urban area. A sex difference in FET impairment was observed, however, no significant difference in impairment in other lung function indicators were found between girls and boys. The elevated height and weight of the children was significantly associated with an increased and decreased prevalence of impaired lung function, respectively, particularly both for FEV6, FEV3, FIVC, and FIV1. Opening windows for a longer time in summer was significantly associated with a lower prevalence of impaired FEF25 and MVV, and an extended time of opening windows in winter was significantly associated with a lower prevalence of impaired FEV6. While, opening windows for a longer time in autumn was significantly associated with higher prevalence of impaired FEV6 and FEV1/VC. Home renovations, doing physical exercise for more than 30 min per day, air pollution exposure during commuting, sleeping in own rooms, height stunting, and preterm birth were significantly associated with an increased risk of impaired lung function. Breastfeeding, having a father with a white-collar profession and with a higher education level were positively associated with the lower prevalence of impaired lung function. Impaired lung function is commonly found in school children in Wuhan, nowadays. Breastfeeding, opening windows long-term in summer and winter, higher socioeconomics, and an urban living environment were protective factors for impaired lung function. However, opening windows long-term and using air conditioning short-term in autumn, as well as home renovations, doing physical exercise for more than 30 min per day, preterm birth, height stunting, and air pollution exposure during commuting were regarded as significant risk factors for impaired lung function. Promoting breastfeeding, lengthening window opening times in winter and summer, and controlling household renovation and air pollution exposure during commuting are recommended to reduce the risk of impaired lung function in children in Wuhan.

A Descriptive Study of Health Literacy and Social Determinants of Health as Curricula Topics in Medical School Education

The purpose of this investigation was to assess the extent to which health literacy and social determinants of health exist together in medical school curricula, and the attitudes and beliefs of medical school educators toward the relevance of these topics taught in the curriculum. The research used a descriptive cross-sectional survey design of institutions that comprised the Accelerating Change in Medical Education (ACE) Consortium of the American Medical Association. The study population was 36 ACE institutions, but only 11 ACE institutions made up the study sample. Results also showed that five health literacy items were taught as curricula topics in medical school education with 100% (n = 10) of the respondents teaching how to use plain language skills for oral communication. Respondents rated the level at which their institution prioritized instructional methods to explicitly teach social determinants of health as a topic in the medical curriculum with three (27.3%) ranking the priority level as extremely high, seven (63.6%) ranking it as high, and one (9.1%) ranking it as low. Medical educators rated five social determinants of health influencing a person’s health status, with “economic stability” and “social and community context” having the first and second highest mean rankings, respectively. Nine (81.5%) medical educators agreed that health literacy is a social determinant of health (M = 8.73) and a predictor of health status (M = 7.82).

Associations of somatic symptom disorder with pain, disability and quality of life in patients with chronic low back pain.

Literature findings have suggested that psychological factors, including anxiety, depression and somatic symptom disorder (SSD), are predictors of poor outcomes in individuals with chronic low back pain (CLBP). The aim of this study was to examine the correlations between anxiety, depression and SSD with pain, disability and health-related quality of life (HRQoL) in Greek CLBP patients. Ninety-two participants with CLBP recruited using random systematic sampling from an outpatient physiotherapy department, who completed a battery of paper-and-pencil questionnaires included items on demographic characteristics, as well the Numerical Pain Rating Scale (NPRS) for pain, the Rolland-Morris Disability Questionnaire for disability (RMDQ), the EuroQoL 5-dimension 5-level (EQ-5D-5L) for health status, the Somatic Symptom Scale-8 (SSS-8) for SSD, the Hospital Anxiety and Depression Scale (HADS) for anxiety and depression. A Mann-Whitney test and a Kruskall-Wallis test were used for the comparison of continuous variables between two groups and among more than two groups, respectively. Moreover, Spearman correlations coefficients were used to explore the association between subjects' demographics, SSS-8, HADS-Anxiety, HADS-Depression, NPS, RMDQ and EQ-5D-5L indices. Predictors of health status, pain and disability were assessed using multiple regression analyses, whereas the level of statistical significance was set at p<0.05. The response rate was 94.6% (87 participants, 55 of whom were women) and the mean age of the sample was 59.6 years (SD=15.1). A tendency of weak negative associations was noted between scores of SSD, anxiety and depression with EQ-5D-5L indices, whereas only a weak positive correlation was found between levels of SSD with pain and disability. After examining in a multiple regression analysis, only SSD emerged as prognostic factor of poor HRQoL, greater levels of pain and disability. In conclusion, the elevated scores of SSD significantly predict worse HRQoL, intense pain and severe disability in Greek CLBP patients. Further research is needed to test our findings in larger and more representative samples of the Greek general population.

A Novel Krill Herd Based Random Forest Algorithm for Monitoring Patient Health

Artificial Intelligence (AI) is finding increasing application in healthcare monitoring. Machine learning systems are utilized for monitoring patient health through the use of IoT sensor, which keep track of the physiological state by way of various health data. Thus, early detection of any disease or derangement can aid doctors in saving patients’ lives. However, there are some challenges associated with predicting health status using the common algorithms, such as time requirements, chances of errors, and improper classification. We propose an Artificial Krill Herd based on the Random Forest (AKHRF) technique for monitoring patients’ health and eliciting an optimal prescription based on their health status. To begin with, various patient datasets were collected and trained into the system using IoT sensors. As a result, the framework developed includes four processes: preprocessing, feature extraction, classification, and result visibility. Additionally, preprocessing removes errors, noise, and missing values from the dataset, whereas feature extraction extracts the relevant information. Then, in the classification layer, we updated the fitness function of the krill herd to classify the patient’s health status and also generate a prescription. We found that the results from the proposed framework are comparable to the results from other state-of-the-art techniques in terms of sensitivity, specificity, Area under the Curve (AUC), accuracy, precision, recall, and F-measure.

Real-Time Prediction of Medical Demand and Mental Health Status in Ukraine under Russian Invasion Using Tweet Analysis.

Russia's invasion of Ukraine (February 24, 2022) has begun and there are concerns about the impact on health care supply and mental health. This study analyzed tweets in the Ukrainian language to capture the medical needs and mental health conditions in wartime Ukraine by focusing on ostensibly relevant words. The number of tweets containing the keywords and their overall proportion was compared before and after the Russian invasion of Ukraine. The survey period was divided into four phases-the pre-2022 Russian invasion, acute phase (4 weeks), subacute phase (12 weeks), and the chronic phase (8 weeks) up to August 10, 2022. The analysis targeted tweets sent in Ukrainian. The tweets were screened using a set of six classes with 75 key groups and 303 Ukrainian (204 original Japanese) keywords. Overall, 98,526,440 tweets were analyzed, with a pre-invasion and post-onset average of 1,096,976 and 3,328,243 tweets/week (a 3.0-fold increase), respectively. Of these, 3,197,443 tweets contained the keywords, with a pre-invasion and invasion average of 26,241 and 114,640 tweets/week (a 4.4-fold increase), respectively. The post-onset phase witnessed a considerable increase in all classes-medical services, treatment, medical resources, medical situations, and special situations-but not in the symptom class. Keywords related to psychological distress and anxiety immediately increased during the acute phase; those related to depression and post-traumatic stress reactions continued increasing as the invasion persisted, which may have reflected the mental state of those impacted. Analyzing tweets is useful for predicting people's real-time physical and mental health needs during wartime.

Intelligent Hybrid Scheme for Health Monitoring of Degrading Rotary Machines: An Adaptive Fuzzy c-Means Coupled With 1-D CNN

The development of a health monitoring scheme for complex rotary machines using unlabeled and imbalanced multivariate data is a well-recognized challenging problem. The challenges and complexity further increase when the health of the machine degrades over time. In order to utilize unlabeled and imbalanced multivariate degradation data, this paper develops an intelligent health monitoring scheme for rotary machines based on the proposed adaptive fuzzy <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">c</i> -means clustering algorithm followed by ensemble learning, mean-variance-based samples generation method, and 1-D convolutional neural network. The adaptive fuzzy <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">c</i> -means clustering algorithm uses nondominated sorting genetic algorithm II to automatically determine the different health state labels from unlabeled degradation data. The principle of ensemble learning is used to increase the robustness of the obtained health state labels. Further, based on obtained health states, a new solution is presented to construct the true remaining useful life adaptively. The proposed framework presents a novel method for generating samples based on mean-variance to deal with imbalanced health states. Furthermore, an architecture of 1-D convolutional neural networks for multivariate data is designed to predict the health states and remaining useful life of rotary machines. The proposed approach has been validated through aero-engine and XJTU-SY bearing datasets. The health state prediction accuracy is as high as 99.97% and 100% for aero-engine and XJTU-SY datasets, respectively, using the proposed scheme. The results and comparative studies show the effectiveness of the proposed scheme.

Face Recognition Method under Adaptive Image Matching and Dictionary Learning Algorithm.

In this research, a robust face recognition method based on adaptive image matching and a dictionary learning algorithm was proposed. A Fisher discriminant constraint was introduced into the dictionary learning algorithm program so that the dictionary had certain category discrimination ability. The purpose was to use this technology to reduce the influence of pollution, absence, and other factors on face recognition and improve the recognition rate. The optimization method was used to solve the loop iteration to obtain the expected specific dictionary, and the selected specific dictionary was used as the representation dictionary in adaptive sparse representation. In addition, if a specific dictionary was placed in a seed space of the original training data, the mapping matrix can be used to represent the mapping relationship between the specific dictionary and the original training sample, and the test sample could be corrected according to the mapping matrix to remove the contamination in the test sample. Moreover, the feature face method and dimension reduction method were used to process the specific dictionary and the corrected test sample, and the dimensions were reduced to 25, 50, 75, 100, 125, and 150, respectively. In this research, the recognition rate of the algorithm in 50 dimensions was lower than that of the discriminatory low-rank representation method (DLRR), and the recognition rate in other dimensions was the highest. The adaptive image matching classifier was used for classification and recognition. The experimental results showed that the proposed algorithm had a good recognition rate and good robustness against noise, pollution, and occlusion. Health condition prediction based on face recognition technology has the advantages of being noninvasive and convenient operation.

Development of an Artificial Intelligence-Supported Hybrid Data Management Platform for Monitoring Depression and Anxiety Symptoms in the Perinatal Period: Pilot-Scale Study

One of the forces driving science and industry is machine learning, but the proliferation of Big Data necessitates paradigm shifts from conventional approaches in applying machine learning techniques to this massive amount of data with varying velocity. Computers are now capable of accurately diagnosing a variety of medical conditions thanks to the availability of immense healthcare datasets and advancements in machine learning techniques. The study’s primary aim is to identify the most compelling questions on anxiety and depression in pregnant women by extracting features through performance-optimized algorithms. In this way, it is aimed to reach the result in a shorter time with fewer questions. The next goal of this work is to create an instant remote health status prediction system for depression and anxiety in pregnant women based on the Apache Spark Big Data processing engine, which concentrates on using machine learning models on streaming Big Data. In this scalable system, the application receives data from pregnant women to forecast the patient’s health condition. It then applies the machine learning algorithm that produces the best results for this dataset. With the assistance of this big data platform, the time-consuming anxiety and depression detection procedure in a pregnant woman can be replaced with a computer-based technique that works in an instant with a respectable amount of accuracy.

A novel composed method of cleaning anomy data for improving state prediction of wind turbine

Improving the efficiency of wind turbine state prediction is an important goal of wind energy utilization. But much of abnormal data existing in supervisory control and data acquisition (SCADA) seriously affects the health state prediction of wind turbine. In this paper, a new composed method is proposed to clean SACAD data according to abnormal data type of wind turbine. In proposed composed method, a preprocessing method is first presented to get rid of outliers of power curve based on operational mechanism, and a new data cleaning method called TTLOF (Thompson tau-local outlier factor) is proposed to quantify particularly data points and eliminate outliers by setting correlation parameter thresholds. In TTLOF cleaning data, Empirical copula-based mutual information (ECMI) is used to select correlation parameters for anomaly characteristic assessments, and each parameter interval is divided for performing segmentation fine cleaning which can reduce the model complexity of identifying anomaly characteristics. Finally, a deep learning network which is long short-term memory (LSTM) is used to verify the effectiveness of the proposed data cleaning method. By analyzing the state monitoring results, it is shown the proposed composed method is more effective for cleaning anomy data than other methods.

Research on frequency band sparse optimization algorithm in health index of rolling bearing

SummaryThe construction of health indicators of rolling bearing is significant for the monitoring of its health condition and residual life prediction. The main work is based on the frequency band sparse optimization algorithm in the rolling bearing health index. On the one hand, the index construction method based on machine learning model has poor interpretability; on the other hand, there are many interference components in the original vibration signal, extracting statistical features directly based on the original signal as a poor health indicator. For the above problems, this paper proposes a rolling bearing health index construction method based on sparse frequency band characterization. First, the signal including each independent sub‐band has been obtained by wavelet packet decomposition. Second, a sparse model is constructed by combining the sub‐band amplitudes and matrices decomposed by the wavelet packet in the fault state and the normal state, so as to select the frequency band sensitive to the fault information. Thereafter, the spectral amplitude sum of sensitive frequency band (SASSFB) is calculated as a health indicator that reflects the degradation information during the complete life operation of the rolling bearing. Finally, the effectiveness of the method is verified by utilizing the published bearing accelerated life experimental data set.