Application of Machine Learning in Predicting Frailty Syndrome in Patients with Heart Failure

Application of Machine Learning in Predicting Frailty Syndrome in Patients with Heart Failure | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Application of Machine Learning in Predicting Frailty Syndrome in Patients with Heart Failure Remigiusz Szczepanowski, Aleksandra H. Pasieczna, Janusz Sobecki, and 16 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-2139591/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 26 Mar, 2024 Read the published version in Advances in Clinical and Experimental Medicine → Version 1 posted You are reading this latest preprint version Abstract Background Prevention and diagnosis of frailty syndrome (FS) in patients with heart failure (HF) requires innovative systems supporting medical personnel to tailor and optimize their treatment and care. Classical methods of diagnosing the FS in patients are not entirely satisfactory. Healthcare personnel in clinical setting use a combination of test and self-reports to diagnose patients and those at risk of frailty, which is time-consuming and costly. Modern medicine uses artificial intelligence (AI) to study the physical and psychosocial domains of frailty in cardiac patients with HF. Methods This paper aimed to present the potential of using the AI approach, emphasizing machine learning in predicting frailty in patients with HF. Our team critically scrutinized the literature on machine learning applications for FS syndrome, as well as reviewed frailty measurements applied to modern clinical practice. Results Our approach analysis resulted in recommendations of machine learning algorithms for predicting frailty in patients. We also presented the exemplary application of machine learning for frailty syndrome in HF patients based on TFI questionnaire measurements, taking into account psychosocial variables Conclusions We recommend AI and machine learning for the holistic and personalized care of patients at risk of the consequences of FS, improve diagnostic tools examining this syndrome, and effective collaboration between psychologists and health care professionals. Frailty Syndrome (FS) Heart Failure (HF) Artificial Intelligence (AI) Machine Learning Medical Personnel Background Frailty syndrome (FS) is broadly defined as the premature or abnormal aging of elderly patients, indicated by a set of symptoms associated with a higher risk of mortality, lower quality of life, and the increased healthcare utilization [ 1 , 2 ]. Understanding the contributions of physical, social, or psychological factors in the prevalence of frailty is an important research problem in contemporary medicine. Addressing such a challenge should result in novel frailty measurements that help healthcare personnel to promptly identify, prevent, and optimally manage patients with FS. In the clinical literature, terms such as weakness and fatigue are often associated with frailty [ 3 ]. Most definitions consider frailty as a clinically recognizable condition resulting from aging that reduces the ability to deal with daily or severe stressors [ 4 ]. However, frailty is also linked to post-surgery complications and other consequences of stress associated with prolonged hospitalization and the risk of death [ 5 ]. Until recently, the frailty concept was defined as closely linked to old age, but there are indications that younger patients can also develop this syndrome [ 6 , 7 ]. Frailty is an increasingly well-recognized clinical syndrome in cardiology that extends beyond the physiological aging process and commonly co-occurs with many cardiovascular diseases as disease-related frailty [ 7 ]. Frailty is more common in patients with HF than in the general population [ 8 ]. HF is a clinical syndrome in which the heart is not able to pump enough blood to meet the demand of the body. The condition leads to symptoms (e.g., breathlessness, ankle swelling, and fatigue) that may be accompanied by signs (e.g., elevated jugular venous pressure, pulmonary crackles, and peripheral oedema) The number of patients with HF is increasing due to the aging of the population and the therapeutic advancements which improve survival of patients with heart disease [ 7 ]. The prevalence of FS in patients with HF is approximately 45%. [ 9 ]. The Cardiovascular Health Study showed that frailty is significantly associated with HF, affecting one in two adults, independent of age or New York Heart Association (NYHA) class [ 10 , 11 ]. A diagnosis of HF indicated the additional loss of biological reserves and increased vulnerability to several adverse clinical consequences [ 12 ]. Frailty increases the risk of HF and, in patients already diagnosed, contributes to increased mortality, rehospitalization, and decreased quality of life [ 13 – 16 ]. The clinical identification of frailty can play an important role in developing preventive strategies against age-related conditions. Stressors that may affect a patient with frailty that, on the one hand, may predispose the patient to adverse health consequences and, on the other hand, may lend themselves to modification or control are divided into four groups: clinical, physical-functional, psychological, and social. They can be clinical, non-clinical, acute, chronic, reversible, treatable, and irreversible and require supportive care [ 8 ]. Measurement Instruments Of Fs There are several measurements to diagnose frailty and identify the potential risk of developing this syndrome. These measures are diverse in their approach to detecting frailty. The operationalization of FS focuses on accumulation-of-deficits or embraces multidimensional dimensions of the FS. The first approach assumes that more health deficits indicate higher frailty [ 17 , 18 ]. On the contrary, the multidimensional approach describes frailty as a dynamic state affecting an individual who experiences losses in one or more domains of human functioning (physical, psychological, social) [ 18 ]. Here, we present selected frailty measures based on either deficit’s accumulation or multidimensional approaches [ 18 ]: The Tilburg Frailty Indicator (TFI) is a self-report questionnaire that consists of 15 questions related to physical, psychological, and social deficits to identify frailty [ 19 ]. The TFI measures losses caused by the influence of a range of variables, and losses which increases the risk of adverse outcomes The Electronic Frailty Index (eFI) includes the diagnosis of 36 deficits, ranging from symptoms, diseases, disabilities, and abnormal laboratory results to classify patients into four groups – no frailty, low frailty, moderate frailty, and high frailty [ 20 ]. The FI-CD index (Frailty index based on clinical deficits, alternatively, Frailty index of accumulative deficits) is based on clinical deficits, including at least 30 comorbidities, symptoms, diseases, and disabilities [ 17 ]. The frailty phenotype developed by Fried et al. includes the assessment of unintentional weight loss of over 5 kgs in the past year, fatigue, lower grip strength, slower walking gait, and lower physical activity to classify older people [ 21 ]. The FI-B (Frailty Index based on Biomarkers) is innovative but time-consuming and costly compared to the questionnaire-based approaches [ 17 ]. The FTS (Frailty Trait Scale) consists of 12 elements covering seven dimensions: energy balance and nutrition, activity, nervous system, circulatory system, weakness, endurance, and slowing down [ 22 ]. A simplified FTS5 (based on five elements) was developed from the full FTS [ 23 ]. The Heart Failure Association (HFA) of the European Society of Cardiology (ESC) advocated a holistic, multidimensional approach was more reliable than a physical approach only, in identifying those patients with HF who have coexisting FS [ 8 ]. According to these assumptions, frailty in patients with HF should be defined as a multidimensional and dynamic condition independent of age, making a person diagnosed with heart failure more vulnerable to stressors. As frailty in HF is viewed as a dynamic and partially reversible condition, recognizing those modifiable components is important to guide management and improve HF outcomes. Focusing on the reversible components of frailty in HF may reduce the risk of adverse clinical effects, such as increased morbidity, increased health care needs: hospitalization, prolonged recovery, institutionalization, and increased dependency and higher mortality risk [ 8 ]. Early recognition of frailty in older adults with HF is needed to target interventions to slow functional decline and improve patient-centred outcomes. Ai In Medical/application/cardiology Artificial intelligence (AI) is gaining popularity and recognition as a feasible tool to support clinical decisions. There is a noticeable trend in the number of publications on AI in biomedicine, including topics such as living assistance, information processing, research, and the most urgent need in medicine: disease diagnostics and prediction [ 24 ]. In terms of AI methods, Support-vector machines (SVM) are one of the most popular options in the broadly understood medical applications [ 21 ], whereas convolutional neural networks (CNN) are the most popular in the case of disease diagnosis [ 26 ]. Most cases of disease detection methods using AI are based on data in the form of diagnostic imaging [ 25 , 27 ], and the three most common disorders detected are cardiovascular disease, sensory system disease, and cancer [ 27 ]. As noted, cardiology is at the forefront of AI applications in medicine. Machine learning is used in various parts of this field, and this connection is gaining popularity which can be observed in the number of papers published on this subject [ 28 ]. Use cases include echocardiography, nuclear cardiology, electrophysiology enhanced diagnosis, prediction of treatment, and prognosis of disease development [ 29 ]. There is a great need to improve the algorithms for detecting patients at risk of hospital admission. Despite access to patient data from devices such as pacemakers or smart watches [ 29 ]. AI-based models for cardiology can also be divided by the type of task they are designed to perform, respectively: diagnosis, classification, and prediction. Although the classification was presented as more challenging than the diagnosis, better results were obtained for this purpose (over 83.70% accuracy). In contrast, the prediction task proved to be the most difficult and produced the worst outcomes, which may be due to the variety of factors that influence disease development and mortality [ 30 ]. The most frequently used models were neural networks (including deep and convolutional networks), obtaining the most accurate results. Among other valuable algorithms, we can distinguish Random Forest, Naive Bayes, Support Vector Machines, k-Nearest Neighbours, and Gradient Boosting Machines [ 30 ]. It is also worth mentioning that during the Covid19 pandemic, AI-supported cardiological research methods were developed that allowed for better medical examination, especially for patients infected with the coronavirus [ 31 ]. Since frailty is an interdisciplinary issue, there is a need for multidisciplinary frailty definitions and their corresponding measures. Recent technological advances allow for much more extensive data to be collected, integrated, and processed in a more complex way, resulting in a significant understanding of frailty, A recent approach to predict frailty is the application of AI. AI algorithms are designed to apply machine learning to extract knowledge from available data [ 32 ]. The explainable AI (XAI) approach could be considered a suitable method for dealing with frailty problems and evaluating the relations between different syndromes, which cannot be seen directly from separate questionnaires. XAI facilitates the diagnosis and treatment of frailty as we can determine the importance of input features, enabling interpretation of the results obtained, dependencies between inputs and their values, and identification of data and concept drift. An example of such methods is tree-based algorithms applied in healthcare due to their property of explainability. Among all possibilities, XGBoost implementation can be considered a reasonable choice as it naturally deals with continuous, binary/discrete, and missing data consistently. Benefits of Application Machine Learning in Management Frail Patients with Heart failure From a medical perspective, machine learning brings advantages potential in predicting FS. These potential benefits are following [ 33 , 34 ]: Employing an explainable machine learning model may help clinicians to gain new insights into the possible determinants of frailty. While some features are non-modifiable, e.g., age and height, other factors may be directly modifiable through lifestyle changes, physical exercise, or cognitive stimulation (e.g., weight, smoking, mobility). As a result, it may be possible to ensure that a patient avoids reaching critical threshold values associated with frailty for some features. Conversely, those threshold values may be set as the targets to achieve a more stable state if engaging in rehabilitation. Facilitating a patient's diagnosis in the event of incomplete data about the patient (e.g., for a patient from another country or a patient who has not previously used medical services). Possibility of indicating significant relations between the frailty variables. Many frailty measures are based on highly correlated variables (for instance, see the Frailty Index (FI) of Accumulative Deficits (FI-CD) measure). As indicated above, frailty includes dozens of measures, e.g., physical health, behavioural risks, cognitive function, and mental health being also highly correlated (e.g., age and marital status). In practice, these measures do not bring new information important from the standpoint of frailty diagnosis. ML algorithms can effectively identify the most representative associations between frailty variables. ML algorithms can capture correlations without duplicating the same information multiple times. Developing a semi-open system with sufficient data, where AI is screening patients based on specific diagnostic taxonomy with confidence intervals (frail status, pre-frail condition / endangered/healthy). Determination of diagnostic importance of frailty components and their contribution to the FS and other comorbid diseases (e.g., assessing the importance of frailty measures for frail patients with HF). For instance, one can target those who suffer from multiple diseases, e.g., frailty and hypertension. There may be synergy effects where co-existing diseases can be linked to an increased risk of the condition under study. This will probably be more specific for this subgroup than for both diseases separately (non-linearity, in addition to the non-linearity concerning age groups and gender). Subgroup-specific "diagnostic importance of variables" could be used to diagnose patients, for example, in precision and holistic medicine. Identification of the FS and its importance (see below) based on incomplete data. Health care professionals can benefit from this functionality when facing extensive historical data, as well as a shortage of resources. Panel data mining from long-term observation. Possibility to have a more advanced predictive model for prophylaxis (preventive care). Updating and expanding the AI systems by the inclusion of existing clinical data and demographics of a given region and country. Possibility of screening pre-frail patients (non-binary output). Limiting human error caused by tiredness, subjectivity, the abundance of data, or other factors. AI supports the selection of therapeutic strategies, i.e., personalization of multidisciplinary care in HF, building health literacy and patient empowerment, and personalization educational recommendations for patients with HF and FS. AI support for multimorbid patients with FS in a modern holistic way; patients can gain more insight into their disease to further enhance self-care. Application Of Ai Algorithms For Predicting Frailty There are recent reports on applying AI algorithms to facilitate the integration of existing, classic frailty diagnosis tools. For instance, one group used AI algorithms to develop predictive models for hospitalization, fracture occurrence, disability, and mortality as proxies for frailty [ 35 ]. In another work, different combinations of indicator variables were used to predict frailty with the results compared to eFI diagnoses [ 36 ]. Their results hinted that the SVM outperformed k-nearest neighbors and the decision tree algorithms. The results of these studies were promising, but the number of explanatory variables used in the most effective SVM model was 70. The accuracy of this model was 93.5%, with a very promising Cohen's Kappa index of 87%. At the same time, the models containing 10 and 11 explanatory variables turned out to be better than some of the more numerous explanatory variables of the models, which suggests significant possibilities of using various combinations of variables. However, this work only concerned patients over 75 years of age and did not focus on any specific disease? In the context of frailty prediction, research on biomarkers has also been used to identify and classify patients with no frailty, risk of frailty, and suffering from frailty [ 37 ]. These works highlight the effectiveness of AI methods to extract relevant information. The AI-based framework applies both supervised and non-supervised learning methods. Non-supervised learning is involved in the classification or grouping methods using, for example, k-means, k-nearest neighbors, decision tree algorithms, or other such as SGD classifier or naive Bayes applied depending on the size of data. Supervised learning uses the following methods: SVM, neural networks (NN), CNN, or other deep learning methods. One of the main problems in AI methods application is appropriate data preparation and feature extraction. Data that will be used to feed AI-based FS prediction system can be divided into two main categories: Data that are possible to be joined, i.e., standard identifiers, are available = > these are the data on which the analysis will be performed. Data that feed the model irrespective of patients = > there are the data that will help create models and validate hypotheses. We do not require a standard model for all patient-specific data on the level of data lakes. However, joining data from different sources should be feasible using known identifiers. However, many other AI methods have proved to provide very efficient results; however, in the case of big data marts, deep neural networks have made great improvements and are offering the best efficiency in many application areas. Most neural models, such as networks of simple non-linear, enable exchanging information via fixed connections, adapting simple parameters to learn vector mappings [ 38 ]. However, we may also apply complex neurons, microcircuits, small neural cortical ensembles with structural connections (fixed or slowly changing) that enable us to model complex network states: rich internal knowledge in modules interacting flexibly. In general, we should find the most straightforward model suitable for a given data and easy to handle: simpler models generalize better, interpretation [ 39 ]. A proper hybrid cloud approach should be considered to carry out efficient AI calculations – some personal data points are susceptible and shouldn't leave local infrastructure. Some anonymization techniques should be utilized here – or maybe only data summaries should be processed in the cloud. Predicting Importance Of Frailty Components In Heart Failure: Analysis Of Tfi Measure Our research team analyzed the diagnostic importance of individual psychosocial and physical criteria in the diagnosis of FS in elderly patients suffering from HF [ 37 ]. Based on the AI approach and the TFI questionnaire, including physical, psychological, and social components, machine learning models were constructed using decision tree, random decision forest, and AdaBoost classifier. These models were trained, validated and tested on three separated subsets of the full dataset. To find the feature importance of the explanatory variables in machine learning classifying models, it was necessary to choose an appropriate method of evaluating these variables. The permutation method compares the accuracy of the model with its accuracy when we shuffle the values of specific variables. The procedure was performed separately for each of the fifteen TFI explanatory variables, with many permutations. The calculations were made for ten thousand permutations and a single variable in our case. The greater the number of respondents (and their answers to a given TFI question) in the sample's subsets, the more permutations should be made to obtain more accurate results. Simulation Results Machine learning models were built and verified in a sample of patients with HF treated at the University Teaching Hospital in Wrocław in 2016–2019. The sample included 666 patients, most of them had been diagnosed with FS. To determine the diagnostic validity and verify the hypotheses, selected components of the physical domain were compared with all the psychological and social domain components within the TFI questionnaire. The models with the highest classification accuracy were selected from the three machine learning algorithms, i.e., the random decision forest and the AdaBoost classifier (Table 1 ). Table 1 Results from testing phase for selected machine learning – true positives and true negatives True positives True negatives Decision Tree 79.07 87.50 Random Forest 93.02 100.00 AdaBoost 100.00 93.75 The conducted ML analysis showed that none of the variables within the social domain was more diagnostically important than the physical variables (i.e., experiencing difficulties due to difficulties in walking, lack of strength in the hands, and physical fatigue). In the case of psychological criteria for the diagnosis of FS, the variable related to irritability (i.e., feeling excited or nervous in the last month) was diagnostically more important than all considered physical variables, while the variable related to depressive mood (i.e., a decrease in mood in the previous month) was diagnostically more important than the physical variables: lack of strength in the hands and physical fatigue. Conclusions The amount of information that is required to correctly identify, interpret and successfully manage a patient with FS and HF is significant, resulting in a time-consuming and costly process. AI, using machine learning can help to parse the data. Machine learning can serve to develop new diagnostic measurements of frailty and support research on improving classic measures and addressing the theoretical issue of the operational definition of this clinical syndrome. These machine learning computations can be used to apply AI in the holistic and personalized care of patients at risk of the consequences of FS, improve diagnostic tools examining this syndrome, and effective collaboration between psychologists and health care professionals. This is applicable in holistic and patient centered medicine, which requires knowledge from all kind of disciplines, allowing for the causal and symptomatic treatment while considering the patient's various domains of life and behavior. Future development should include discussion on the compatibly of clinical patient data sources and privacy. Abbreviations AI: artificial intelligence; CNN: convolutional neural networks; eFI: Electronic Frailty Index; ESC: European Society of Cardiology; FI-B: Frailty Index based on Biomarkers; FI-CD: Frailty index based on clinical deficits; FI: Frailty Index; FS: frailty syndrome; FTS: Frailty Trait Scale; HF: heart failure; HFA: Heart Failure Association; NN: neural networks; NYHA: New York Heart Association; SVM: support-vector machines; TFI: Tilburg Frailty Indicator. Declarations Ethics approval and consent to participate Not applicable. All methods were performed in accordance with the relevant guidelines and regulations. Consent for publication Not applicable. Availability of data and materials The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request Competing interests The authors declare that they have no competing interests. Funding No funding sources are associated with this article. Author contributions R.S., R.J.J.G., A.H.P., I.U., J.B., J.S., R.K., M.W., A.H., L.H., T.J., M.B., N.V. and F.C.M. conceptualized the article; R.S., R.J.J.G., A.H.P., I.U., J.S., R.K., M.W., A.H., A.D., A.K., W.L. and M.K. performed the literature search; R.S., R.J.J.G., A.H.P., I.U., R.K., M.W., L.H., T.J., A.H., M.B., N.V., F.C.M., G.K., W.L. and M.K. drafted the paper; R.S., J.S., R.J.J.G., I.U., L.H., T.J., A.H., M.B. and F.C.M. critically revised the paper. All authors read and approved the manuscript. Acknowledgements The authors would like to thank to all experts who cooperated in preparing this paper. Author details 1 Department of Computer Science and Systems Engineering, Wroclaw University of Science and Technology, Wroclaw, Poland. 2 Department of Finance and Accounting, Kozminski University, Warsaw, Poland. 3 Department of Nursing and Obstetrics, Faculty of Health Sciences, Wroclaw Medical University, Wroclaw, Poland. 4 Center for Heart Diseases, University Hospital, Wroclaw, Poland. 5 Institute of Heart Diseases, Wroclaw Medical University, Wroclaw, Poland. 6 Department Family Medicine and Population Health, Faculty of Medicine and Health Sciences, University of Antwerp, Antwerp, Belgium. 7 Faculty of Health, Sports and Social Work, Inholland University of Applied Sciences, Amsterdam, the Netherlands. 8 Department Family Medicine and Population Health, Faculty of Medicine and Health Sciences, University of Antwerp, Antwerp, Belgium. 9 School of Nursing and Midwifery, Queen's University, Belfast, UK. 10 Department of Health, Medicine and Caring Sciences, Linkoping University, Linköping, Sweden. 11 School of Computing, Edinburgh Napier University, Edinburgh, United Kingdom. 12 Geriatric Unit, Department of Internal Medicine and Geriatrics, University of Palermo, Palermo, Italy. 13 DICEAM, University Mediterranea of Reggio Calabria, Reggio Calabria, Italy. 14 Institut de Physique et de Chimie des Matériaux, CNRS-Unistra, Université de Strasbourg, Strasbourg, France. 15 Faculty of Medicine, Medical University of Gdansk, Gdansk, Poland. References Łęgosz P, Krzowski B, Płatek AE, Ryś A, Semczuk K, Szymański FM, et al. Frailty syndrome in clinical practce- what should we remember? Folia Cardiol. 2018;13:137–43. Vitale C, Jankowska E, Hill L, Piepoli M, Doehner W, Anker SD, et al. Heart Failure Association/European Society of Cardiology position paper on frailty in patients with heart failure. Eur J Heart Fail. 2019;21:1299–305. Uchmanowicz I, Lisiak M, Jankowska-Polańska B. Narzędzia badawcze stosowane w ocenie zespo\lu kruchości. Gerontologia Polska. 2014;22:1–8. Xue Q-L. The frailty syndrome: definition and natural history. Clinics in geriatric medicine. 2011;27:1–15. Morley JE, Perry III HM, Miller DK. Something about frailty. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences. 2002;57:M698–704. Ness KK, Armstrong GT, Kundu M, Wilson CL, Tchkonia T, Kirkland JL. Frailty in childhood cancer survivors. Cancer. 2015;121:1540–7. McDonagh J, Ferguson C, Newton PJ. Frailty assessment in heart failure: an overview of the multi-domain approach. Current Heart Failure Reports. 2018;15:17–23. Vitale C, Jankowska E, Hill L, Piepoli M, Doehner W, Anker SD, et al. Heart Failure Association of the European Society of Cardiology position paper on frailty in patients with heart failure. European journal of heart failure. 2019;21:1299–305. Chen MA. Frailty and cardiovascular disease: potential role of gait speed in surgical risk stratification in older adults. Journal of geriatric cardiology: JGC. 2015;12:44. Vitale C, Spoletini I, Rosano GM. Frailty in heart failure: implications for management. Card Fail Rev. 2018; 4 (2): 104. Denfeld QE, Winters-Stone K, Mudd JO, Gelow JM, Kurdi S, Lee CS. The prevalence of frailty in heart failure: A systematic review and meta-analysis. Int J Cardiol. 2017;236:283–9. Bottle A, Kim D, Hayhoe B, Majeed A, Aylin P, Clegg A, et al. Frailty and co-morbidity predict first hospitalisation after heart failure diagnosis in primary care: population-based observational study in England. Age and ageing. 2019;48:347–54. Wleklik M, Denfeld Q, Lisiak M, Czapla M, Kałużna-Oleksy M, Uchmanowicz I. Frailty Syndrome in Older Adults with Cardiovascular Diseases–What Do We Know and What Requires Further Research? Int J Environ Res Public Health. 2022;19:2234. Uchmanowicz I, Kuśnierz M, Wleklik M, Jankowska-Polańska B, Jaroch J, Łoboz-Grudzień K. Frailty syndrome and rehospitalizations in elderly heart failure patients. Aging Clin Exp Res. 2018;30:617–23. Uchmanowicz I, Lee CS, Vitale C, Manulik S, Denfeld QE, Uchmanowicz B, et al. Frailty and the risk of all‐cause mortality and hospitalization in chronic heart failure: a meta‐analysis. ESC Heart Fail. 2020;7:3427–37. Uchmanowicz I, Pasieczna AH, Wójta-Kempa M, Gobbens RJJ, Młynarska A, Faulkner KM, et al. Physical, Psychological and Social Frailty Are Predictive of Heart Failure: A Cross-Sectional Study. J Clin Med. 2022;11:565. Mitnitski A, Collerton J, Martin-Ruiz C, Jagger C, von Zglinicki T, Rockwood K, et al. Age-related frailty and its association with biological markers of ageing. BMC Medicine. 2015;13:161. Dent E, Kowal P, Hoogendijk EO. Frailty measurement in research and clinical practice: a review. European journal of internal medicine. 2016;31:3–10. Gobbens RJ, van Assen MA, Luijkx KG, Wijnen-Sponselee MT, Schols JM. The Tilburg frailty indicator: psychometric properties. Journal of the American Medical Directors Association. 2010;11:344–55. Clegg A, Bates C, Young J, Ryan R, Nichols L, Ann Teale E, et al. Development and validation of an electronic frailty index using routine primary care electronic health record data. Age and ageing. 2016;45:353–60. Fried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J, et al. Frailty in older adults: evidence for a phenotype. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences. 2001;56:M146–57. García-García FJ, Carcaillon L, Fernandez-Tresguerres J, Alfaro A, Larrion JL, Castillo C, et al. A new operational definition of frailty: the Frailty Trait Scale. Journal of the American Medical Directors Association. 2014;15:371-e7. García-García FJ, Carnicero JA, Losa-Reyna J, Alfaro-Acha A, Castillo-Gallego C, Rosado-Artalejo C, et al. Frailty Trait Scale–Short Form: A frailty instrument for clinical practice. Journal of the American Medical Directors Association. 2020;21:1260–6. Rong G, Mendez A, Assi EB, Zhao B, Sawan M. Artificial intelligence in healthcare: review and prediction case studies. Engineering. 2020;6:291–301. Jiang F, Jiang Y, Zhi H, Dong Y, Li H, Ma S, et al. Artificial intelligence in healthcare: past, present and future. Stroke and vascular neurology. 2017;2. Shen J, Zhang CJ, Jiang B, Chen J, Song J, Liu Z, et al. Artificial intelligence versus clinicians in disease diagnosis: systematic review. JMIR medical informatics. 2019;7:e10010. Sethi GK, Ahmad N, Rehman MB, Dafallaa HMEI, Rashid M. Use of Artificial Intelligence in Healthcare Systems: State-of-the-Art Survey. In: 2021 2nd International Conference on Intelligent Engineering and Management (ICIEM). IEEE; 2021. p. 243–8. Souza Filho EM de, Fernandes F de A, Soares CL de A, Seixas FL, Santos AAS, Gismondi RA, et al. Artificial intelligence in cardiology: concepts, tools and challenges-“the horse is the one who runs, you must be the jockey.” Arquivos brasileiros de cardiologia. 2020;114:718–25. Lopez-Jimenez F, Attia Z, Arruda-Olson AM, Carter R, Chareonthaitawee P, Jouni H, et al. Artificial intelligence in cardiology: present and future. In: Mayo Clinic Proceedings. Elsevier; 2020. p. 1015–39. Šećkanović A, Šehovac M, Spahić L, Ramić I, Mamatnazarova N, Pokvić LG, et al. Review of artificial intelligence application in cardiology. In: 2020 9th Mediterranean Conference on Embedded Computing (MECO). IEEE; 2020. p. 1–5. Haleem A, Javaid M, Singh RP, Suman R. Applications of Artificial Intelligence (AI) for cardiology during COVID-19 pandemic. Sustainable Operations and Computers. 2021;2:71–8. Marr B. What is the difference between artificial intelligence and machine learning. Forbes, December. 2016;6. Iniesta R, Stahl D, McGuffin P. Machine learning, statistical learning and the future of biological research in psychiatry. Psychological medicine. 2016;46:2455–65. Mijwil M. Artificial Neural Networks Advantages and Disadvantages. 2018. Tarekegn A, Ricceri F, Costa G, Ferracin E, Giacobini M. Predictive modeling for frailty conditions in elderly people: machine learning approaches. JMIR medical informatics. 2020;8:e16678. Ambagtsheer RC, Shafiabady N, Dent E, Seiboth C, Beilby J. The application of artificial intelligence (AI) techniques to identify frailty within a residential aged care administrative data set. International journal of medical informatics. 2020;136:104094. Gomez-Cabrero D, Walter S, Abugessaisa I, Miñambres-Herraiz R, Palomares LB, Butcher L, et al. A robust machine learning framework to identify signatures for frailty: a nested case-control study in four aging European cohorts. Geroscience. 2021;43:1317–29. Duch W, Maszczyk T, Grochowski M. Optimal support features for meta-learning. In: Meta-Learning in Computational Intelligence. Springer; 2011. p. 317–58. Jankowski N, Duch W, Grąbczewski K. Meta-learning in computational intelligence. Springer; 2011. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 26 Mar, 2024 Read the published version in Advances in Clinical and Experimental Medicine → Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-2139591","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":175860956,"identity":"71a8345b-7c5e-4530-a5c4-e31a9764a73a","order_by":0,"name":"Remigiusz Szczepanowski","email":"data:image/png;base64,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","orcid":"","institution":"Wrocław University of Science and Technology","correspondingAuthor":true,"submittingAuthor":false,"prefix":"","firstName":"Remigiusz","middleName":"","lastName":"Szczepanowski","suffix":""},{"id":175860958,"identity":"a98c7e3d-d229-470e-a97e-13f0c6074115","order_by":1,"name":"Aleksandra H. Pasieczna","email":"","orcid":"","institution":"Kozminski University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Aleksandra","middleName":"H.","lastName":"Pasieczna","suffix":""},{"id":175860959,"identity":"ffc2edbc-3dd4-4b22-86a9-bd6acd1ecac3","order_by":2,"name":"Janusz Sobecki","email":"","orcid":"","institution":"Wrocław University of Science and Technology","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Janusz","middleName":"","lastName":"Sobecki","suffix":""},{"id":175860960,"identity":"fd8916e2-6734-47f7-a813-96a73dd023a4","order_by":3,"name":"Radoslaw Katarzyniak","email":"","orcid":"","institution":"Wrocław University of Science and Technology","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Radoslaw","middleName":"","lastName":"Katarzyniak","suffix":""},{"id":175860962,"identity":"15c93554-4748-4bbe-b9e9-047f879fb49f","order_by":4,"name":"Grzegorz Kołaczek","email":"","orcid":"","institution":"Wrocław University of Science and Technology","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Grzegorz","middleName":"","lastName":"Kołaczek","suffix":""},{"id":175860965,"identity":"903dbcf1-0189-4f2f-9d5b-763601536b59","order_by":5,"name":"Wojciech Lorkiewicz","email":"","orcid":"","institution":"Wrocław University of Science and Technology","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Wojciech","middleName":"","lastName":"Lorkiewicz","suffix":""},{"id":175860968,"identity":"0ac03d20-b13e-42d4-bb6f-f962f504bdd7","order_by":6,"name":"Maja Kędras","email":"","orcid":"","institution":"Wrocław University of Science and Technology","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Maja","middleName":"","lastName":"Kędras","suffix":""},{"id":175860969,"identity":"07d97740-2708-4746-b104-dea9d37280bf","order_by":7,"name":"Izabella Uchmanowicz","email":"","orcid":"","institution":"Wrocław Medical University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Izabella","middleName":"","lastName":"Uchmanowicz","suffix":""},{"id":175860971,"identity":"49e8f547-c4e3-43be-96e7-04c8d2cf4ca3","order_by":8,"name":"Jan Biegus","email":"","orcid":"","institution":"University Hospital in Wroclaw","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Jan","middleName":"","lastName":"Biegus","suffix":""},{"id":175860973,"identity":"572cd3f9-6d79-4563-a0be-11e208d3dce4","order_by":9,"name":"Marta Wleklik","email":"","orcid":"","institution":"Wrocław Medical University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Marta","middleName":"","lastName":"Wleklik","suffix":""},{"id":175860974,"identity":"63d8f100-eb52-4d4b-9d7c-a360aedc9e73","order_by":10,"name":"Robbert J.J. Gobbens","email":"","orcid":"","institution":"Inholland University of Applied Sciences","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Robbert","middleName":"J.J.","lastName":"Gobbens","suffix":""},{"id":175860975,"identity":"6b570b65-42d0-48ac-b301-d058a1227853","order_by":11,"name":"Loreena Hill","email":"","orcid":"","institution":"Queen's University Belfast","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Loreena","middleName":"","lastName":"Hill","suffix":""},{"id":175860976,"identity":"b4bb07c5-7ca8-40ec-b3cc-3b4e0d2d20b9","order_by":12,"name":"Tiny Jaarsma","email":"","orcid":"","institution":"Linköping University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Tiny","middleName":"","lastName":"Jaarsma","suffix":""},{"id":175860977,"identity":"eaae9c74-3f43-4975-93c1-f52180ef2859","order_by":13,"name":"Amir Hussain","email":"","orcid":"","institution":"Edinburgh Napier University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Amir","middleName":"","lastName":"Hussain","suffix":""},{"id":175860978,"identity":"9812aef8-796e-473d-a325-dd8c107ba338","order_by":14,"name":"Mario Barbagallo","email":"","orcid":"","institution":"University of Palermo","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Mario","middleName":"","lastName":"Barbagallo","suffix":""},{"id":175860979,"identity":"25e409c1-e863-4a00-abcc-39ed7a60db51","order_by":15,"name":"Nicola Veronese","email":"","orcid":"","institution":"University of Palermo","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Nicola","middleName":"","lastName":"Veronese","suffix":""},{"id":175860980,"identity":"6f04b5b7-dac3-474c-9be3-134936948ba8","order_by":16,"name":"Francesco C. Morabito","email":"","orcid":"","institution":"University of Reggio Calabria","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Francesco","middleName":"C.","lastName":"Morabito","suffix":""},{"id":175860981,"identity":"62c5b86d-0d60-435c-9930-d4e0957641f3","order_by":17,"name":"Anant Dixit","email":"","orcid":"","institution":"University of Strasbourg","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Anant","middleName":"","lastName":"Dixit","suffix":""},{"id":175860982,"identity":"eb97faba-4f76-477c-b2a6-fc5666215f6c","order_by":18,"name":"Aleksander Kahsin","email":"","orcid":"","institution":"Medical University of Gdańsk","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Aleksander","middleName":"","lastName":"Kahsin","suffix":""}],"badges":[],"createdAt":"2022-10-06 17:14:10","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-2139591/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-2139591/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.17219/acem/184040","type":"published","date":"2024-03-26T11:54:08+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":53553341,"identity":"cbe370c3-4d40-43ee-9c40-9d4b1d207216","added_by":"auto","created_at":"2024-03-27 11:54:16","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":356532,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-2139591/v1/5c09e402-b08c-4f03-8971-ae0f00996f2a.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Application of Machine Learning in Predicting Frailty Syndrome in Patients with Heart Failure","fulltext":[{"header":"Background","content":"\u003cp\u003eFrailty syndrome (FS) is broadly defined as the premature or abnormal aging of elderly patients, indicated by a set of symptoms associated with a higher risk of mortality, lower quality of life, and the increased healthcare utilization [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. Understanding the contributions of physical, social, or psychological factors in the prevalence of frailty is an important research problem in contemporary medicine. Addressing such a challenge should result in novel frailty measurements that help healthcare personnel to promptly identify, prevent, and optimally manage patients with FS.\u003c/p\u003e \u003cp\u003eIn the clinical literature, terms such as weakness and fatigue are often associated with frailty [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. Most definitions consider frailty as a clinically recognizable condition resulting from aging that reduces the ability to deal with daily or severe stressors [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. However, frailty is also linked to post-surgery complications and other consequences of stress associated with prolonged hospitalization and the risk of death [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. Until recently, the frailty concept was defined as closely linked to old age, but there are indications that younger patients can also develop this syndrome [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eFrailty is an increasingly well-recognized clinical syndrome in cardiology that extends beyond the physiological aging process and commonly co-occurs with many cardiovascular diseases as disease-related frailty [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Frailty is more common in patients with HF than in the general population [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. HF is a clinical syndrome in which the heart is not able to pump enough blood to meet the demand of the body. The condition leads to symptoms (e.g., breathlessness, ankle swelling, and fatigue) that may be accompanied by signs (e.g., elevated jugular venous pressure, pulmonary crackles, and peripheral oedema) The number of patients with HF is increasing due to the aging of the population and the therapeutic advancements which improve survival of patients with heart disease [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThe prevalence of FS in patients with HF is approximately 45%. [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. The Cardiovascular Health Study showed that frailty is significantly associated with HF, affecting one in two adults, independent of age or New York Heart Association (NYHA) class [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. A diagnosis of HF indicated the additional loss of biological reserves and increased vulnerability to several adverse clinical consequences [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. Frailty increases the risk of HF and, in patients already diagnosed, contributes to increased mortality, rehospitalization, and decreased quality of life [\u003cspan additionalcitationids=\"CR14 CR15\" citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. The clinical identification of frailty can play an important role in developing preventive strategies against age-related conditions. Stressors that may affect a patient with frailty that, on the one hand, may predispose the patient to adverse health consequences and, on the other hand, may lend themselves to modification or control are divided into four groups: clinical, physical-functional, psychological, and social. They can be clinical, non-clinical, acute, chronic, reversible, treatable, and irreversible and require supportive care [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e].\u003c/p\u003e"},{"header":"Measurement Instruments Of Fs","content":"\u003cp\u003eThere are several measurements to diagnose frailty and identify the potential risk of developing this syndrome. These measures are diverse in their approach to detecting frailty. The operationalization of FS focuses on accumulation-of-deficits or embraces multidimensional dimensions of the FS. The first approach assumes that more health deficits indicate higher frailty [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. On the contrary, the multidimensional approach describes frailty as a dynamic state affecting an individual who experiences losses in one or more domains of human functioning (physical, psychological, social) [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. Here, we present selected frailty measures based on either deficit\u0026rsquo;s accumulation or multidimensional approaches [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]:\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eThe Tilburg Frailty Indicator (TFI) is a self-report questionnaire that consists of 15 questions related to physical, psychological, and social deficits to identify frailty [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. The TFI measures losses caused by the influence of a range of variables, and losses which increases the risk of adverse outcomes\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eThe Electronic Frailty Index (eFI) includes the diagnosis of 36 deficits, ranging from symptoms, diseases, disabilities, and abnormal laboratory results to classify patients into four groups \u0026ndash; no frailty, low frailty, moderate frailty, and high frailty [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e].\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eThe FI-CD index (Frailty index based on clinical deficits, alternatively, Frailty index of accumulative deficits) is based on clinical deficits, including at least 30 comorbidities, symptoms, diseases, and disabilities [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e].\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eThe frailty phenotype developed by Fried et al. includes the assessment of unintentional weight loss of over 5 kgs in the past year, fatigue, lower grip strength, slower walking gait, and lower physical activity to classify older people [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e].\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eThe FI-B (Frailty Index based on Biomarkers) is innovative but time-consuming and costly compared to the questionnaire-based approaches [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e].\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eThe FTS (Frailty Trait Scale) consists of 12 elements covering seven dimensions: energy balance and nutrition, activity, nervous system, circulatory system, weakness, endurance, and slowing down [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e].\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eA simplified FTS5 (based on five elements) was developed from the full FTS [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e].\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cp\u003eThe Heart Failure Association (HFA) of the European Society of Cardiology (ESC) advocated a holistic, multidimensional approach was more reliable than a physical approach only, in identifying those patients with HF who have coexisting FS [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. According to these assumptions, frailty in patients with HF should be defined as \u003cem\u003ea multidimensional and dynamic condition independent of age, making a person diagnosed with heart failure more vulnerable to stressors.\u003c/em\u003e As frailty in HF is viewed as a dynamic and partially reversible condition, recognizing those modifiable components is important to guide management and improve HF outcomes. Focusing on the reversible components of frailty in HF may reduce the risk of adverse clinical effects, such as increased morbidity, increased health care needs: hospitalization, prolonged recovery, institutionalization, and increased dependency and higher mortality risk [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. Early recognition of frailty in older adults with HF is needed to target interventions to slow functional decline and improve patient-centred outcomes.\u003c/p\u003e\n\u003ch3\u003eAi In Medical/application/cardiology\u003c/h3\u003e\n\u003cp\u003eArtificial intelligence (AI) is gaining popularity and recognition as a feasible tool to support clinical decisions. There is a noticeable trend in the number of publications on AI in biomedicine, including topics such as living assistance, information processing, research, and the most urgent need in medicine: disease diagnostics and prediction [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]. In terms of AI methods, Support-vector machines (SVM) are one of the most popular options in the broadly understood medical applications [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e], whereas convolutional neural networks (CNN) are the most popular in the case of disease diagnosis [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]. Most cases of disease detection methods using AI are based on data in the form of diagnostic imaging [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e], and the three most common disorders detected are cardiovascular disease, sensory system disease, and cancer [\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eAs noted, cardiology is at the forefront of AI applications in medicine. Machine learning is used in various parts of this field, and this connection is gaining popularity which can be observed in the number of papers published on this subject [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]. Use cases include echocardiography, nuclear cardiology, electrophysiology enhanced diagnosis, prediction of treatment, and prognosis of disease development [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. There is a great need to improve the algorithms for detecting patients at risk of hospital admission. Despite access to patient data from devices such as pacemakers or smart watches [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. AI-based models for cardiology can also be divided by the type of task they are designed to perform, respectively: diagnosis, classification, and prediction. Although the classification was presented as more challenging than the diagnosis, better results were obtained for this purpose (over 83.70% accuracy). In contrast, the prediction task proved to be the most difficult and produced the worst outcomes, which may be due to the variety of factors that influence disease development and mortality [\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]. The most frequently used models were neural networks (including deep and convolutional networks), obtaining the most accurate results. Among other valuable algorithms, we can distinguish Random Forest, Naive Bayes, Support Vector Machines, k-Nearest Neighbours, and Gradient Boosting Machines [\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]. It is also worth mentioning that during the Covid19 pandemic, AI-supported cardiological research methods were developed that allowed for better medical examination, especially for patients infected with the coronavirus [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eSince frailty is an interdisciplinary issue, there is a need for multidisciplinary frailty definitions and their corresponding measures. Recent technological advances allow for much more extensive data to be collected, integrated, and processed in a more complex way, resulting in a significant understanding of frailty, A recent approach to predict frailty is the application of AI. AI algorithms are designed to apply machine learning to extract knowledge from available data [\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThe explainable AI (XAI) approach could be considered a suitable method for dealing with frailty problems and evaluating the relations between different syndromes, which cannot be seen directly from separate questionnaires. XAI facilitates the diagnosis and treatment of frailty as we can determine the importance of input features, enabling interpretation of the results obtained, dependencies between inputs and their values, and identification of data and concept drift. An example of such methods is tree-based algorithms applied in healthcare due to their property of explainability. Among all possibilities, XGBoost implementation can be considered a reasonable choice as it naturally deals with continuous, binary/discrete, and missing data consistently.\u003c/p\u003e \u003cp\u003eBenefits of Application Machine Learning in Management Frail Patients with Heart failure From a medical perspective, machine learning brings advantages potential in predicting FS. These potential benefits are following [\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]:\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eEmploying an explainable machine learning model may help clinicians to gain new insights into the possible determinants of frailty. While some features are non-modifiable, e.g., age and height, other factors may be directly modifiable through lifestyle changes, physical exercise, or cognitive stimulation (e.g., weight, smoking, mobility). As a result, it may be possible to ensure that a patient avoids reaching critical threshold values associated with frailty for some features. Conversely, those threshold values may be set as the targets to achieve a more stable state if engaging in rehabilitation.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eFacilitating a patient's diagnosis in the event of incomplete data about the patient (e.g., for a patient from another country or a patient who has not previously used medical services).\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003ePossibility of indicating significant relations between the frailty variables. Many frailty measures are based on highly correlated variables (for instance, see the Frailty Index (FI) of Accumulative Deficits (FI-CD) measure). As indicated above, frailty includes dozens of measures, e.g., physical health, behavioural risks, cognitive function, and mental health being also highly correlated (e.g., age and marital status). In practice, these measures do not bring new information important from the standpoint of frailty diagnosis. ML algorithms can effectively identify the most representative associations between frailty variables. ML algorithms can capture correlations without duplicating the same information multiple times.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eDeveloping a semi-open system with sufficient data, where AI is screening patients based on specific diagnostic taxonomy with confidence intervals (frail status, pre-frail condition / endangered/healthy).\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eDetermination of diagnostic importance of frailty components and their contribution to the FS and other comorbid diseases (e.g., assessing the importance of frailty measures for frail patients with HF). For instance, one can target those who suffer from multiple diseases, e.g., frailty and hypertension. There may be synergy effects where co-existing diseases can be linked to an increased risk of the condition under study. This will probably be more specific for this subgroup than for both diseases separately (non-linearity, in addition to the non-linearity concerning age groups and gender). Subgroup-specific \"diagnostic importance of variables\" could be used to diagnose patients, for example, in precision and holistic medicine.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eIdentification of the FS and its importance (see below) based on incomplete data. Health care professionals can benefit from this functionality when facing extensive historical data, as well as a shortage of resources.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003ePanel data mining from long-term observation. Possibility to have a more advanced predictive model for prophylaxis (preventive care).\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eUpdating and expanding the AI systems by the inclusion of existing clinical data and demographics of a given region and country.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003ePossibility of screening pre-frail patients (non-binary output).\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eLimiting human error caused by tiredness, subjectivity, the abundance of data, or other factors.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eAI supports the selection of therapeutic strategies, i.e., personalization of multidisciplinary care in HF, building health literacy and patient empowerment, and personalization educational recommendations for patients with HF and FS.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eAI support for multimorbid patients with FS in a modern holistic way; patients can gain more insight into their disease to further enhance self-care.\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e\n\u003ch3\u003eApplication Of Ai Algorithms For Predicting Frailty\u003c/h3\u003e\n\u003cp\u003eThere are recent reports on applying AI algorithms to facilitate the integration of existing, classic frailty diagnosis tools. For instance, one group used AI algorithms to develop predictive models for hospitalization, fracture occurrence, disability, and mortality as proxies for frailty [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e]. In another work, different combinations of indicator variables were used to predict frailty with the results compared to eFI diagnoses [\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]. Their results hinted that the SVM outperformed k-nearest neighbors and the decision tree algorithms. The results of these studies were promising, but the number of explanatory variables used in the most effective SVM model was 70. The accuracy of this model was 93.5%, with a very promising Cohen's Kappa index of 87%. At the same time, the models containing 10 and 11 explanatory variables turned out to be better than some of the more numerous explanatory variables of the models, which suggests significant possibilities of using various combinations of variables. However, this work only concerned patients over 75 years of age and did not focus on any specific disease?\u003c/p\u003e \u003cp\u003eIn the context of frailty prediction, research on biomarkers has also been used to identify and classify patients with no frailty, risk of frailty, and suffering from frailty [\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e]. These works highlight the effectiveness of AI methods to extract relevant information. The AI-based framework applies both supervised and non-supervised learning methods. Non-supervised learning is involved in the classification or grouping methods using, for example, k-means, k-nearest neighbors, decision tree algorithms, or other such as SGD classifier or naive Bayes applied depending on the size of data. Supervised learning uses the following methods: SVM, neural networks (NN), CNN, or other deep learning methods.\u003c/p\u003e \u003cp\u003eOne of the main problems in AI methods application is appropriate data preparation and feature extraction. Data that will be used to feed AI-based FS prediction system can be divided into two main categories:\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eData that are possible to be joined, i.e., standard identifiers, are available\u0026thinsp;=\u0026thinsp;\u0026gt;\u0026thinsp;these are the data on which the analysis will be performed.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eData that feed the model irrespective of patients\u0026thinsp;=\u0026thinsp;\u0026gt;\u0026thinsp;there are the data that will help create models and validate hypotheses.\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cp\u003eWe do not require a standard model for all patient-specific data on the level of data lakes. However, joining data from different sources should be feasible using known identifiers. However, many other AI methods have proved to provide very efficient results; however, in the case of big data marts, deep neural networks have made great improvements and are offering the best efficiency in many application areas. Most neural models, such as networks of simple non-linear, enable exchanging information via fixed connections, adapting simple parameters to learn vector mappings [\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e]. However, we may also apply complex neurons, microcircuits, small neural cortical ensembles with structural connections (fixed or slowly changing) that enable us to model complex network states: rich internal knowledge in modules interacting flexibly. In general, we should find the most straightforward model suitable for a given data and easy to handle: simpler models generalize better, interpretation [\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e]. A proper hybrid cloud approach should be considered to carry out efficient AI calculations \u0026ndash; some personal data points are susceptible and shouldn't leave local infrastructure. Some anonymization techniques should be utilized here \u0026ndash; or maybe only data summaries should be processed in the cloud.\u003c/p\u003e\n\u003ch3\u003ePredicting Importance Of Frailty Components In Heart Failure: Analysis Of Tfi Measure\u003c/h3\u003e\n\u003cp\u003eOur research team analyzed the diagnostic importance of individual psychosocial and physical criteria in the diagnosis of FS in elderly patients suffering from HF [\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e]. Based on the AI approach and the TFI questionnaire, including physical, psychological, and social components, machine learning models were constructed using decision tree, random decision forest, and AdaBoost classifier. These models were trained, validated and tested on three separated subsets of the full dataset.\u003c/p\u003e \u003cp\u003eTo find the feature importance of the explanatory variables in machine learning classifying models, it was necessary to choose an appropriate method of evaluating these variables. The permutation method compares the accuracy of the model with its accuracy when we shuffle the values of specific variables. The procedure was performed separately for each of the fifteen TFI explanatory variables, with many permutations. The calculations were made for ten thousand permutations and a single variable in our case. The greater the number of respondents (and their answers to a given TFI question) in the sample's subsets, the more permutations should be made to obtain more accurate results.\u003c/p\u003e\n\u003ch3\u003eSimulation Results\u003c/h3\u003e\n\u003cp\u003eMachine learning models were built and verified in a sample of patients with HF treated at the University Teaching Hospital in Wrocław in 2016\u0026ndash;2019. The sample included 666 patients, most of them had been diagnosed with FS. To determine the diagnostic validity and verify the hypotheses, selected components of the physical domain were compared with all the psychological and social domain components within the TFI questionnaire.\u003c/p\u003e \u003cp\u003eThe models with the highest classification accuracy were selected from the three machine learning algorithms, i.e., the random decision forest and the AdaBoost classifier (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eResults from testing phase for selected machine learning \u0026ndash; true positives and true negatives\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTrue positives\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eTrue negatives\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDecision Tree\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e79.07\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e87.50\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRandom Forest\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e93.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e100.00\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAdaBoost\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e100.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e93.75\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eThe conducted ML analysis showed that none of the variables within the social domain was more diagnostically important than the physical variables (i.e., experiencing difficulties due to difficulties in walking, lack of strength in the hands, and physical fatigue). In the case of psychological criteria for the diagnosis of FS, the variable related to irritability (i.e., feeling excited or nervous in the last month) was diagnostically more important than all considered physical variables, while the variable related to depressive mood (i.e., a decrease in mood in the previous month) was diagnostically more important than the physical variables: lack of strength in the hands and physical fatigue.\u003c/p\u003e"},{"header":"Conclusions","content":"\u003cp\u003eThe amount of information that is required to correctly identify, interpret and successfully manage a patient with FS and HF is significant, resulting in a time-consuming and costly process. AI, using machine learning can help to parse the data. Machine learning can serve to develop new diagnostic measurements of frailty and support research on improving classic measures and addressing the theoretical issue of the operational definition of this clinical syndrome. These machine learning computations can be used to apply AI in the holistic and personalized care of patients at risk of the consequences of FS, improve diagnostic tools examining this syndrome, and effective collaboration between psychologists and health care professionals. This is applicable in holistic and patient centered medicine, which requires knowledge from all kind of disciplines, allowing for the causal and symptomatic treatment while considering the patient's various domains of life and behavior. Future development should include discussion on the compatibly of clinical patient data sources and privacy.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eAI: artificial intelligence; CNN: convolutional neural networks; eFI: Electronic Frailty Index; ESC: European Society of Cardiology; FI-B: Frailty Index based on Biomarkers; FI-CD: Frailty index based on clinical deficits; FI: Frailty Index; FS: frailty syndrome; FTS: Frailty Trait Scale; HF: heart failure; HFA: Heart Failure Association; NN: neural networks; NYHA: New York Heart Association; SVM: support-vector machines; TFI: Tilburg Frailty Indicator.\u003c/p\u003e\n"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable. All methods were performed in accordance with the relevant guidelines and regulations.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003cstrong\u003eConsent for publication\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003cstrong\u003eCompeting interests\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNo funding sources are associated with this article.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003cstrong\u003eAuthor contributions\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eR.S., R.J.J.G., A.H.P., I.U., J.B., J.S., R.K., M.W., A.H., L.H., T.J., M.B., N.V. and F.C.M. conceptualized the article; R.S., R.J.J.G., A.H.P., I.U., J.S., R.K., M.W., A.H., A.D., A.K., W.L. and M.K. performed the literature search; R.S., R.J.J.G., A.H.P., I.U., R.K., M.W., L.H., T.J., A.H., M.B., N.V., F.C.M., G.K., W.L. and M.K. drafted the paper; R.S., J.S., R.J.J.G., I.U., L.H., T.J., A.H., M.B. and F.C.M. critically revised the paper. All authors read and approved the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eAcknowledgements\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors would like to thank to all experts who cooperated in preparing this paper.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003cstrong\u003eAuthor details\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003csup\u003e1\u0026nbsp;\u003c/sup\u003eDepartment of Computer Science and Systems Engineering, Wroclaw University of Science and Technology, Wroclaw, Poland. \u003csup\u003e2\u0026nbsp;\u003c/sup\u003eDepartment of Finance and Accounting, Kozminski University, Warsaw, Poland. \u003csup\u003e3\u0026nbsp;\u003c/sup\u003eDepartment of Nursing and Obstetrics, Faculty of Health Sciences, Wroclaw Medical University, Wroclaw, Poland. \u003csup\u003e4\u0026nbsp;\u003c/sup\u003eCenter for Heart Diseases, University Hospital, Wroclaw, Poland. \u003csup\u003e5\u0026nbsp;\u003c/sup\u003eInstitute of Heart Diseases, Wroclaw Medical University, Wroclaw, Poland. \u003csup\u003e6\u003c/sup\u003e Department Family Medicine and Population Health, Faculty of Medicine and Health Sciences, University of Antwerp, Antwerp, Belgium. \u003csup\u003e7\u0026nbsp;\u003c/sup\u003eFaculty of Health, Sports and Social Work, Inholland University of Applied Sciences, Amsterdam, the Netherlands. \u003csup\u003e8\u003c/sup\u003e Department Family Medicine and Population Health, Faculty of Medicine and Health Sciences, University of Antwerp, Antwerp, Belgium. \u003csup\u003e9\u003c/sup\u003e School of Nursing and Midwifery, Queen\u0026apos;s University, Belfast, UK. \u003csup\u003e10\u003c/sup\u003e Department of Health, Medicine and Caring Sciences, Linkoping University, Link\u0026ouml;ping, Sweden. \u003csup\u003e11\u003c/sup\u003e School of Computing, Edinburgh Napier University, Edinburgh, United Kingdom. \u003csup\u003e12\u0026nbsp;\u003c/sup\u003eGeriatric Unit, Department of Internal Medicine and Geriatrics, University of Palermo, Palermo, Italy. \u003csup\u003e13\u003c/sup\u003e DICEAM, University Mediterranea of Reggio Calabria, Reggio Calabria, Italy. \u003csup\u003e14\u003c/sup\u003e Institut de Physique et de Chimie des Mat\u0026eacute;riaux, CNRS-Unistra, Universit\u0026eacute; de Strasbourg, Strasbourg, France. \u003csup\u003e15\u0026nbsp;\u003c/sup\u003eFaculty of Medicine, Medical University of Gdansk, Gdansk, Poland.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eŁęgosz P, Krzowski B, Płatek AE, Ryś A, Semczuk K, Szymański FM, et al. Frailty syndrome in clinical practce- what should we remember? Folia Cardiol. 2018;13:137\u0026ndash;43.\u003c/li\u003e\n\u003cli\u003eVitale C, Jankowska E, Hill L, Piepoli M, Doehner W, Anker SD, et al. Heart Failure Association/European Society of Cardiology position paper on frailty in patients with heart failure. Eur J Heart Fail. 2019;21:1299\u0026ndash;305.\u003c/li\u003e\n\u003cli\u003eUchmanowicz I, Lisiak M, Jankowska-Polańska B. Narzędzia badawcze stosowane w ocenie zespo\\lu kruchości. Gerontologia Polska. 2014;22:1\u0026ndash;8.\u003c/li\u003e\n\u003cli\u003eXue Q-L. The frailty syndrome: definition and natural history. Clinics in geriatric medicine. 2011;27:1\u0026ndash;15.\u003c/li\u003e\n\u003cli\u003eMorley JE, Perry III HM, Miller DK. Something about frailty. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences. 2002;57:M698\u0026ndash;704.\u003c/li\u003e\n\u003cli\u003eNess KK, Armstrong GT, Kundu M, Wilson CL, Tchkonia T, Kirkland JL. Frailty in childhood cancer survivors. Cancer. 2015;121:1540\u0026ndash;7.\u003c/li\u003e\n\u003cli\u003eMcDonagh J, Ferguson C, Newton PJ. Frailty assessment in heart failure: an overview of the multi-domain approach. Current Heart Failure Reports. 2018;15:17\u0026ndash;23.\u003c/li\u003e\n\u003cli\u003eVitale C, Jankowska E, Hill L, Piepoli M, Doehner W, Anker SD, et al. Heart Failure Association of the European Society of Cardiology position paper on frailty in patients with heart failure. European journal of heart failure. 2019;21:1299\u0026ndash;305.\u003c/li\u003e\n\u003cli\u003eChen MA. Frailty and cardiovascular disease: potential role of gait speed in surgical risk stratification in older adults. Journal of geriatric cardiology: JGC. 2015;12:44.\u003c/li\u003e\n\u003cli\u003eVitale C, Spoletini I, Rosano GM. Frailty in heart failure: implications for management. Card Fail Rev. 2018; 4 (2): 104.\u003c/li\u003e\n\u003cli\u003eDenfeld QE, Winters-Stone K, Mudd JO, Gelow JM, Kurdi S, Lee CS. The prevalence of frailty in heart failure: A systematic review and meta-analysis. Int J Cardiol. 2017;236:283\u0026ndash;9.\u003c/li\u003e\n\u003cli\u003eBottle A, Kim D, Hayhoe B, Majeed A, Aylin P, Clegg A, et al. Frailty and co-morbidity predict first hospitalisation after heart failure diagnosis in primary care: population-based observational study in England. Age and ageing. 2019;48:347\u0026ndash;54.\u003c/li\u003e\n\u003cli\u003eWleklik M, Denfeld Q, Lisiak M, Czapla M, Kałużna-Oleksy M, Uchmanowicz I. Frailty Syndrome in Older Adults with Cardiovascular Diseases\u0026ndash;What Do We Know and What Requires Further Research? Int J Environ Res Public Health. 2022;19:2234.\u003c/li\u003e\n\u003cli\u003eUchmanowicz I, Kuśnierz M, Wleklik M, Jankowska-Polańska B, Jaroch J, Łoboz-Grudzień K. Frailty syndrome and rehospitalizations in elderly heart failure patients. Aging Clin Exp Res. 2018;30:617\u0026ndash;23.\u003c/li\u003e\n\u003cli\u003eUchmanowicz I, Lee CS, Vitale C, Manulik S, Denfeld QE, Uchmanowicz B, et al. Frailty and the risk of all‐cause mortality and hospitalization in chronic heart failure: a meta‐analysis. ESC Heart Fail. 2020;7:3427\u0026ndash;37.\u003c/li\u003e\n\u003cli\u003eUchmanowicz I, Pasieczna AH, W\u0026oacute;jta-Kempa M, Gobbens RJJ, Młynarska A, Faulkner KM, et al. Physical, Psychological and Social Frailty Are Predictive of Heart Failure: A Cross-Sectional Study. J Clin Med. 2022;11:565.\u003c/li\u003e\n\u003cli\u003eMitnitski A, Collerton J, Martin-Ruiz C, Jagger C, von Zglinicki T, Rockwood K, et al. Age-related frailty and its association with biological markers of ageing. BMC Medicine. 2015;13:161.\u003c/li\u003e\n\u003cli\u003eDent E, Kowal P, Hoogendijk EO. Frailty measurement in research and clinical practice: a review. European journal of internal medicine. 2016;31:3\u0026ndash;10.\u003c/li\u003e\n\u003cli\u003eGobbens RJ, van Assen MA, Luijkx KG, Wijnen-Sponselee MT, Schols JM. The Tilburg frailty indicator: psychometric properties. Journal of the American Medical Directors Association. 2010;11:344\u0026ndash;55.\u003c/li\u003e\n\u003cli\u003eClegg A, Bates C, Young J, Ryan R, Nichols L, Ann Teale E, et al. Development and validation of an electronic frailty index using routine primary care electronic health record data. Age and ageing. 2016;45:353\u0026ndash;60.\u003c/li\u003e\n\u003cli\u003eFried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J, et al. Frailty in older adults: evidence for a phenotype. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences. 2001;56:M146\u0026ndash;57.\u003c/li\u003e\n\u003cli\u003eGarc\u0026iacute;a-Garc\u0026iacute;a FJ, Carcaillon L, Fernandez-Tresguerres J, Alfaro A, Larrion JL, Castillo C, et al. A new operational definition of frailty: the Frailty Trait Scale. Journal of the American Medical Directors Association. 2014;15:371-e7.\u003c/li\u003e\n\u003cli\u003eGarc\u0026iacute;a-Garc\u0026iacute;a FJ, Carnicero JA, Losa-Reyna J, Alfaro-Acha A, Castillo-Gallego C, Rosado-Artalejo C, et al. Frailty Trait Scale\u0026ndash;Short Form: A frailty instrument for clinical practice. Journal of the American Medical Directors Association. 2020;21:1260\u0026ndash;6.\u003c/li\u003e\n\u003cli\u003eRong G, Mendez A, Assi EB, Zhao B, Sawan M. Artificial intelligence in healthcare: review and prediction case studies. Engineering. 2020;6:291\u0026ndash;301.\u003c/li\u003e\n\u003cli\u003eJiang F, Jiang Y, Zhi H, Dong Y, Li H, Ma S, et al. Artificial intelligence in healthcare: past, present and future. Stroke and vascular neurology. 2017;2.\u003c/li\u003e\n\u003cli\u003eShen J, Zhang CJ, Jiang B, Chen J, Song J, Liu Z, et al. Artificial intelligence versus clinicians in disease diagnosis: systematic review. JMIR medical informatics. 2019;7:e10010.\u003c/li\u003e\n\u003cli\u003eSethi GK, Ahmad N, Rehman MB, Dafallaa HMEI, Rashid M. Use of Artificial Intelligence in Healthcare Systems: State-of-the-Art Survey. In: 2021 2nd International Conference on Intelligent Engineering and Management (ICIEM). IEEE; 2021. p. 243\u0026ndash;8.\u003c/li\u003e\n\u003cli\u003eSouza Filho EM de, Fernandes F de A, Soares CL de A, Seixas FL, Santos AAS, Gismondi RA, et al. Artificial intelligence in cardiology: concepts, tools and challenges-\u0026ldquo;the horse is the one who runs, you must be the jockey.\u0026rdquo; Arquivos brasileiros de cardiologia. 2020;114:718\u0026ndash;25.\u003c/li\u003e\n\u003cli\u003eLopez-Jimenez F, Attia Z, Arruda-Olson AM, Carter R, Chareonthaitawee P, Jouni H, et al. Artificial intelligence in cardiology: present and future. In: Mayo Clinic Proceedings. Elsevier; 2020. p. 1015\u0026ndash;39.\u003c/li\u003e\n\u003cli\u003e\u0026Scaron;ećkanović A, \u0026Scaron;ehovac M, Spahić L, Ramić I, Mamatnazarova N, Pokvić LG, et al. Review of artificial intelligence application in cardiology. In: 2020 9th Mediterranean Conference on Embedded Computing (MECO). IEEE; 2020. p. 1\u0026ndash;5.\u003c/li\u003e\n\u003cli\u003eHaleem A, Javaid M, Singh RP, Suman R. Applications of Artificial Intelligence (AI) for cardiology during COVID-19 pandemic. Sustainable Operations and Computers. 2021;2:71\u0026ndash;8.\u003c/li\u003e\n\u003cli\u003eMarr B. What is the difference between artificial intelligence and machine learning. Forbes, December. 2016;6.\u003c/li\u003e\n\u003cli\u003eIniesta R, Stahl D, McGuffin P. Machine learning, statistical learning and the future of biological research in psychiatry. Psychological medicine. 2016;46:2455\u0026ndash;65.\u003c/li\u003e\n\u003cli\u003eMijwil M. Artificial Neural Networks Advantages and Disadvantages. 2018.\u003c/li\u003e\n\u003cli\u003eTarekegn A, Ricceri F, Costa G, Ferracin E, Giacobini M. Predictive modeling for frailty conditions in elderly people: machine learning approaches. JMIR medical informatics. 2020;8:e16678.\u003c/li\u003e\n\u003cli\u003eAmbagtsheer RC, Shafiabady N, Dent E, Seiboth C, Beilby J. The application of artificial intelligence (AI) techniques to identify frailty within a residential aged care administrative data set. International journal of medical informatics. 2020;136:104094.\u003c/li\u003e\n\u003cli\u003eGomez-Cabrero D, Walter S, Abugessaisa I, Mi\u0026ntilde;ambres-Herraiz R, Palomares LB, Butcher L, et al. A robust machine learning framework to identify signatures for frailty: a nested case-control study in four aging European cohorts. Geroscience. 2021;43:1317\u0026ndash;29.\u003c/li\u003e\n\u003cli\u003eDuch W, Maszczyk T, Grochowski M. Optimal support features for meta-learning. In: Meta-Learning in Computational Intelligence. Springer; 2011. p. 317\u0026ndash;58.\u003c/li\u003e\n\u003cli\u003eJankowski N, Duch W, Grąbczewski K. Meta-learning in computational intelligence. Springer; 2011.\u003c/li\u003e\n\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Frailty Syndrome (FS), Heart Failure (HF), Artificial Intelligence (AI), Machine Learning, Medical Personnel","lastPublishedDoi":"10.21203/rs.3.rs-2139591/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-2139591/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003ePrevention and diagnosis of frailty syndrome (FS) in patients with heart failure (HF) requires innovative systems supporting medical personnel to tailor and optimize their treatment and care. Classical methods of diagnosing the FS in patients are not entirely satisfactory. Healthcare personnel in clinical setting use a combination of test and self-reports to diagnose patients and those at risk of frailty, which is time-consuming and costly. Modern medicine uses artificial intelligence (AI) to study the physical and psychosocial domains of frailty in cardiac patients with HF.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eThis paper aimed to present the potential of using the AI approach, emphasizing machine learning in predicting frailty in patients with HF. Our team critically scrutinized the literature on machine learning applications for FS syndrome, as well as reviewed frailty measurements applied to modern clinical practice.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eOur approach analysis resulted in recommendations of machine learning algorithms for predicting frailty in patients. We also presented the exemplary application of machine learning for frailty syndrome in HF patients based on TFI questionnaire measurements, taking into account psychosocial variables\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eWe recommend AI and machine learning for the holistic and personalized care of patients at risk of the consequences of FS, improve diagnostic tools examining this syndrome, and effective collaboration between psychologists and health care professionals.\u003c/p\u003e","manuscriptTitle":"Application of Machine Learning in Predicting Frailty Syndrome in Patients with Heart Failure","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2023-02-15 13:44:23","doi":"10.21203/rs.3.rs-2139591/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"bd3137ae-a45e-4286-914b-fa8c0b6a7618","owner":[],"postedDate":"February 15th, 2023","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2024-03-27T11:54:08+00:00","versionOfRecord":{"articleIdentity":"rs-2139591","link":"https://doi.org/10.17219/acem/184040","journal":{"identity":"advances-in-clinical-and-experimental-medicine","isVorOnly":true,"title":"Advances in Clinical and Experimental Medicine"},"publishedOn":"2024-03-26 11:54:08","publishedOnDateReadable":"March 26th, 2024"},"versionCreatedAt":"2023-02-15 13:44:23","video":"","vorDoi":"10.17219/acem/184040","vorDoiUrl":"https://doi.org/10.17219/acem/184040","workflowStages":[]},"version":"v1","identity":"rs-2139591","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-2139591","identity":"rs-2139591","version":["v1"]},"buildId":"7rjqhiLT3MXkJMwkYKINL","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}