Process mining for healthcare: Characteristics and challenges.

Process mining techniques can be used to analyse business processes using the data logged during their execution. These techniques are leveraged in a wide range of domains, including healthcare, where it focuses mainly on the analysis of diagnostic, treatment, and organisational processes. Despite the huge amount of data generated in hospitals by staff and machinery involved in healthcare processes, there is no evidence of a systematic uptake of process mining beyond targeted case studies in a research context. When developing and using process mining in healthcare, distinguishing characteristics of healthcare processes such as their variability and patient-centred focus require targeted attention. Against this background, the Process-Oriented Data Science in Healthcare Alliance has been established to propagate the research and application of techniques targeting the data-driven improvement of healthcare processes. This paper, an initiative of the alliance, presents the distinguishing characteristics of the healthcare domain that need to be considered to successfully use process mining, as well as open challenges that need to be addressed by the community in the future.

Dynamic Models Supporting Personalised Chronic Disease Management through Healthcare Sensors with Interactive Process Mining.

Rich streams of continuous data are available through Smart Sensors representing a unique opportunity to develop and analyse risk models in healthcare and extract knowledge from data. There is a niche for developing new algorithms, and visualisation and decision support tools to assist health professionals in chronic disease management incorporating data generated through smart sensors in a more precise and personalised manner. However, current understanding of risk models relies on static snapshots of health variables or measures, rather than ongoing and dynamic feedback loops of behaviour, considering changes and different states of patients and diseases. The rationale of this work is to introduce a new method for discovering dynamic risk models for chronic diseases, based on patients' dynamic behaviour provided by health sensors, using Process Mining techniques. Results show the viability of this method, three dynamic models have been discovered for the chronic diseases hypertension, obesity, and diabetes, based on the dynamic behaviour of metabolic risk factors associated. This information would support health professionals to translate a one-fits-all current approach to treatments and care, to a personalised medicine strategy, that fits treatments built on patients' unique behaviour thanks to dynamic risk modelling taking advantage of the amount data generated by smart sensors.

Design and Evaluation of a Solo-Resident Smart Home Testbed for Mobility Pattern Monitoring and Behavioural Assessment.

Aging population increase demands for solutions to help the solo-resident elderly live independently. Unobtrusive data collection in a smart home environment can monitor and assess elderly residents' health state based on changes in their mobility patterns. In this paper, a smart home system testbed setup for a solo-resident house is discussed and evaluated. We use paired Passive infra-red (PIR) sensors at each entry of a house and capture the resident's activities to model mobility patterns. We present the required testbed implementation phases, i.e., deployment, post-deployment analysis, re-deployment, and conduct behavioural data analysis to highlight the usability of collected data from a smart home. The main contribution of this work is to apply intelligence from a post-deployment process mining technique (namely, the parallel activity log inference algorithm (PALIA)) to find the best configuration for data collection in order to minimise the errors. Based on the post-deployment analysis, a re-deployment phase is performed, and results show the improvement of collected data accuracy in re-deployment phase from 81.57% to 95.53%. To complete our analysis, we apply the well-known CASAS project dataset as a reference to conduct a comparison with our collected results which shows a similar pattern. The collected data further is processed to use the level of activity of the solo-resident for a behaviour assessment.

Analyzing of Gender Behaviors from Paths Using Process Mining: A Shopping Mall Application.

The study presents some results of customer paths' analysis in a shopping mall. Bluetooth-based technology is used to collect data. The event log containing spatiotemporal information is analyzed with process mining. Process mining is a technique that enables one to see the whole process contrary to data-centric methods. The use of process mining can provide a readily-understandable view of the customer paths. We installed iBeacon devices, a Bluetooth-based positioning system, in the shopping mall. During December 2017 and January and February 2018, close to 8000 customer data were captured. We aim to investigate customer behaviors regarding gender by using their paths. We can determine the gender of customers if they go to the men's bathroom or women's bathroom. Since the study has a comprehensive scope, we focused on male and female customers' behaviors. This study shows that male and female customers have different behaviors. Their duration and paths, in general, are not similar. In addition, the study shows that the process mining technique is a viable way to analyze customer behavior using Bluetooth-based technology.

Multidisciplinary Collaboration in the Treatment of Patients With Type 2 Diabetes in Primary Care: Analysis Using Process Mining.

BACKGROUND: Public health in several countries is characterized by a shortage of professionals and a lack of economic resources. Monitoring and redesigning processes can foster the success of health care institutions, enabling them to provide a quality service while simultaneously reducing costs. Process mining, a discipline that extracts knowledge from information system data to analyze operational processes, affords an opportunity to understand health care processes. OBJECTIVE: Health care processes are highly flexible and multidisciplinary, and health care professionals are able to coordinate in a variety of different ways to treat a diagnosis. The aim of this work was to understand whether the ways in which professionals coordinate their work affect the clinical outcome of patients. METHODS: This paper proposes a method based on the use of process mining to identify patterns of collaboration between physician, nurse, and dietitian in the treatment of patients with type 2 diabetes mellitus and to compare these patterns with the clinical evolution of the patients within the context of primary care. Clustering is used as part of the preprocessing of data to manage the variability, and then process mining is used to identify patterns that may arise. RESULTS: The method is applied in three primary health care centers in Santiago, Chile. A total of seven collaboration patterns were identified, which differed primarily in terms of the number of disciplines present, the participation intensity of each discipline, and the referrals between disciplines. The pattern in which the three disciplines participated in the most equitable and comprehensive manner had a lower proportion of highly decompensated patients compared with those patterns in which the three disciplines participated in an unbalanced manner. CONCLUSIONS: By discovering which collaboration patterns lead to improved outcomes, health care centers can promote the most successful patterns among their professionals so as to improve the treatment of patients. Process mining techniques are useful for discovering those collaborations patterns in flexible and unstructured health care processes.

Wearable Sensors Integrated with Internet of Things for Advancing eHealth Care.

Health and sociological indicators alert that life expectancy is increasing, hence so are the years that patients have to live with chronic diseases and co-morbidities. With the advancement in ICT, new tools and paradigms are been explored to provide effective and efficient health care. Telemedicine and health sensors stand as indispensable tools for promoting patient engagement, self-management of diseases and assist doctors to remotely follow up patients. In this paper, we evaluate a rapid prototyping solution for information merging based on five health sensors and two low-cost ubiquitous computing components: Arduino and Raspberry Pi. Our study, which is entirely described with the purpose of reproducibility, aimed to evaluate the extent to which portable technologies are capable of integrating wearable sensors by comparing two deployment scenarios: Raspberry Pi 3 and Personal Computer. The integration is implemented using a choreography engine to transmit data from sensors to a display unit using web services and a simple communication protocol with two modes of data retrieval. Performance of the two set-ups is compared by means of the latency in the wearable data transmission and data loss. PC has a delay of 0.051 ± 0.0035 s (max = 0.2504 s), whereas the Raspberry Pi yields a delay of 0.0175 ± 0.149 s (max = 0.294 s) for N = 300. Our analysis confirms that portable devices ( p < < 0 . 01 ) are suitable to support the transmission and analysis of biometric signals into scalable telemedicine systems.

Evaluating the Social Media Performance of Hospitals in Spain: A Longitudinal and Comparative Study.

BACKGROUND: Social media is changing the way in which citizens and health professionals communicate. Previous studies have assessed the use of Health 2.0 by hospitals, showing clear evidence of growth in recent years. In order to understand if this happens in Spain, it is necessary to assess the performance of health care institutions on the Internet social media using quantitative indicators. OBJECTIVES: The study aimed to analyze how hospitals in Spain perform on the Internet and social media networks by determining quantitative indicators in 3 different dimensions: presence, use, and impact and assess these indicators on the 3 most commonly used social media - Facebook, Twitter, YouTube. Further, we aimed to find out if there was a difference between private and public hospitals in their use of the aforementioned social networks. METHODS: The evolution of presence, use, and impact metrics is studied over the period 2011- 2015. The population studied accounts for all the hospitals listed in the National Hospitals Catalog (NHC). The percentage of hospitals having Facebook, Twitter, and YouTube profiles has been used to show the presence and evolution of hospitals on social media during this time. Usage was assessed by analyzing the content published on each social network. Impact evaluation was measured by analyzing the trend of subscribers for each social network. Statistical analysis was performed using a lognormal transformation and also using a nonparametric distribution, with the aim of comparing t student and Wilcoxon independence tests for the observed variables. RESULTS: From the 787 hospitals identified, 69.9% (550/787) had an institutional webpage and 34.2% (269/787) had at least one profile in one of the social networks (Facebook, Twitter, and YouTube) in December 2015. Hospitals' Internet presence has increased by more than 450.0% (787/172) and social media presence has increased ten times since 2011. Twitter is the preferred social network for public hospitals, whereas private hospitals showed better performance on Facebook and YouTube. The two-sided Wilcoxon test and t student test at a CI of 95% show that the use of Twitter distribution is higher (P<.001) for private and public hospitals in Spain, whereas other variables show a nonsignificant different distribution. CONCLUSIONS: The Internet presence of Spanish hospitals is high; however, their presence on the 3 main social networks is still not as high compared to that of hospitals in the United States and Western Europe. Public hospitals are found to be more active on Twitter, whereas private hospitals show better performance on Facebook and YouTube. This study suggests that hospitals, both public and private, should devote more effort to and be more aware of social media, with a clear strategy as to how they can foment new relationships with patients and citizens.

Integration of Distributed Services and Hybrid Models Based on Process Choreography to Predict and Detect Type 2 Diabetes.

Life expectancy is increasing and, so, the years that patients have to live with chronic diseases and co-morbidities. Type 2 diabetes is one of the most prevalent chronic diseases, specifically linked to being overweight and ages over sixty. Recent studies have demonstrated the effectiveness of new strategies to delay and even prevent the onset of type 2 diabetes by a combination of active and healthy lifestyle on cohorts of mid to high risk subjects. Prospective research has been driven on large groups of the population to build risk scores that aim to obtain a rule for the classification of patients according to the odds for developing the disease. Currently, there are more than two hundred models and risk scores for doing this, but a few have been properly evaluated in external groups and integrated into a clinical application for decision support. In this paper, we present a novel system architecture based on service choreography and hybrid modeling, which enables a distributed integration of clinical databases, statistical and mathematical engines and web interfaces to be deployed in a clinical setting. The system was assessed during an eight-week continuous period with eight endocrinologists of a hospital who evaluated up to 8080 patients with seven different type 2 diabetes risk models implemented in two mathematical engines. Throughput was assessed as a matter of technical key performance indicators, confirming the reliability and efficiency of the proposed architecture to integrate hybrid artificial intelligence tools into daily clinical routine to identify high risk subjects.

Evaluation of Google Glass Technical Limitations on Their Integration in Medical Systems.

Google Glass is a wearable sensor presented to facilitate access to information and assist while performing complex tasks. Despite the withdrawal of Google in supporting the product, today there are multiple applications and much research analyzing the potential impact of this technology in different fields of medicine. Google Glass satisfies the need of managing and having rapid access to real-time information in different health care scenarios. Among the most common applications are access to electronic medical records, display monitorizations, decision support and remote consultation in specialties ranging from ophthalmology to surgery and teaching. The device enables a user-friendly hands-free interaction with remote health information systems and broadcasting medical interventions and consultations from a first-person point of view. However, scientific evidence highlights important technical limitations in its use and integration, such as failure in connectivity, poor reception of images and automatic restart of the device. This article presents a technical study on the aforementioned limitations (specifically on the latency, reliability and performance) on two standard communication schemes in order to categorize and identify the sources of the problems. Results have allowed us to obtain a basis to define requirements for medical applications to prevent network, computational and processing failures associated with the use of Google Glass.

Process Mining Methodology for Health Process Tracking Using Real-Time Indoor Location Systems.

The definition of efficient and accurate health processes in hospitals is crucial for ensuring an adequate quality of service. Knowing and improving the behavior of the surgical processes in a hospital can improve the number of patients that can be operated on using the same resources. However, the measure of this process is usually made in an obtrusive way, forcing nurses to get information and time data, affecting the proper process and generating inaccurate data due to human errors during the stressful journey of health staff in the operating theater. The use of indoor location systems can take time information about the process in an unobtrusive way, freeing nurses, allowing them to engage in purely welfare work. However, it is necessary to present these data in a understandable way for health professionals, who cannot deal with large amounts of historical localization log data. The use of process mining techniques can deal with this problem, offering an easily understandable view of the process. In this paper, we present a tool and a process mining-based methodology that, using indoor location systems, enables health staff not only to represent the process, but to know precise information about the deployment of the process in an unobtrusive and transparent way. We have successfully tested this tool in a real surgical area with 3613 patients during February, March and April of 2015.

The Assessment of the Association of Proton Pump Inhibitor Usage with Chronic Kidney Disease Progression through a Process Mining Approach.

Previous studies have suggested an association between Proton Pump Inhibitors (PPIs) and the progression of chronic kidney disease (CKD). This study aims to assess the association between PPI use and CKD progression by analysing estimated glomerular filtration rate (eGFR) trajectories using a process mining approach. We conducted a retrospective cohort study from 1 January 2006 to 31 December 2011, utilising data from the Stockholm Creatinine Measurements (SCREAM). New users of PPIs and H2 blockers (H2Bs) with CKD (eGFR < 60) were identified using a new-user and active-comparator design. Process mining discovery is a technique that discovers patterns and sequences in events over time, making it suitable for studying longitudinal eGFR trajectories. We used this technique to construct eGFR trajectory models for both PPI and H2B users. Our analysis indicated that PPI users exhibited more complex and rapidly declining eGFR trajectories compared to H2B users, with a 75% increased risk (adjusted hazard ratio [HR] 1.75, 95% confidence interval [CI] 1.49 to 2.06) of transitioning from moderate eGFR stage (G3) to more severe stages (G4 or G5). These findings suggest that PPI use is associated with an increased risk of CKD progression, demonstrating the utility of process mining for longitudinal analysis in epidemiology, leading to an improved understanding of disease progression.

Process mining for individualized behavior modeling using wireless tracking in nursing homes.

The analysis of human behavior patterns is increasingly used for several research fields. The individualized modeling of behavior using classical techniques requires too much time and resources to be effective. A possible solution would be the use of pattern recognition techniques to automatically infer models to allow experts to understand individual behavior. However, traditional pattern recognition algorithms infer models that are not readily understood by human experts. This limits the capacity to benefit from the inferred models. Process mining technologies can infer models as workflows, specifically designed to be understood by experts, enabling them to detect specific behavior patterns in users. In this paper, the eMotiva process mining algorithms are presented. These algorithms filter, infer and visualize workflows. The workflows are inferred from the samples produced by an indoor location system that stores the location of a resident in a nursing home. The visualization tool is able to compare and highlight behavior patterns in order to facilitate expert understanding of human behavior. This tool was tested with nine real users that were monitored for a 25-week period. The results achieved suggest that the behavior of users is continuously evolving and changing and that this change can be measured, allowing for behavioral change detection.

Process mining and data mining applications in the domain of chronic diseases: A systematic review.

The widespread use of information technology in healthcare leads to extensive data collection, which can be utilised to enhance patient care and manage chronic illnesses. Our objective is to summarise previous studies that have used data mining or process mining methods in the context of chronic diseases in order to identify research trends and future opportunities. The review covers articles that pertain to the application of data mining or process mining methods on chronic diseases that were published between 2000 and 2022. Articles were sourced from PubMed, Web of Science, EMBASE, and Google Scholar based on predetermined inclusion and exclusion criteria. A total of 71 articles met the inclusion criteria and were included in the review. Based on the literature review results, we detected a growing trend in the application of data mining methods in diabetes research. Additionally, a distinct increase in the use of process mining methods to model clinical pathways in cancer research was observed. Frequently, this takes the form of a collaborative integration of process mining, data mining, and traditional statistical methods. In light of this collaborative approach, the meticulous selection of statistical methods based on their underlying assumptions is essential when integrating these traditional methods with process mining and data mining methods. Another notable challenge is the lack of standardised guidelines for reporting process mining studies in the medical field. Furthermore, there is a pressing need to enhance the clinical interpretation of data mining and process mining results.

How to Design Successful Participatory Design Workshops for Digital Health Solutions?

Participatory design (PD) is increasingly used to support design and development of digital health solutions. The involves representatives of future user groups and experts to collect their needs and preferences and ensure easy to use and useful solutions. However, reflections and experiences with PD in designing digital health solutions are rarely reported. The objective of this paper is to collect those experiences including lessons learnt and moderator experiences, and to identify challenges. For this purpose, we conducted a multiple case study to explore the skill development process required to successfully design a solution in the three cases. From the results, we derived good practice guidelines to support designing successful PD workshops. They include adapting the workshop activities and material to the vulnerable participant group and considering their environment and previous experiences, planning sufficient time for preparation and supporting the activities with appropriate material. We conclude that PD workshop results are perceived as useful for designing digital health solutions, but careful design is very relevant.

Assessment of distance to primary percutaneous coronary intervention centres in ST-segment elevation myocardial infarction: Overcoming inequalities with process mining tools.

Objectives: In ST-segment elevation myocardial infarction (STEMI), time delay between symptom onset and treatment is critical to improve outcome. The expected transport delay between patient location and percutaneous coronary intervention (PCI) centre is paramount for choosing the adequate reperfusion therapy. The "Centro" region of Portugal has heterogeneity in PCI assess due to geographical reasons. We aimed to explore time delays between regions using process mining tools. Methods: Retrospective observational analysis of patients with STEMI from the Portuguese Registry of Acute Coronary Syndromes. We collected information on geographical area of symptom onset, reperfusion option, and in-hospital mortality. We built a national and a regional patient's flow models by using a process mining methodology based on parallel activity-based log inference algorithm. Results: Totally, 8956 patients (75% male, 48% from 51 to 70 years) were included in the national model. Most patients (73%) had primary PCI, with the median time between admission and treatment <120 minutes in every region; "Centro" had the longest delay. In the regional model corresponding to the "Centro" region of Portugal divided by districts, only 61% had primary PCI, with "Guarda" (05:04) and "Castelo Branco" (06:50) showing longer delays between diagnosis and reperfusion than "Coimbra" (01:19). For both models, in-hospital mortality was higher for those without reperfusion therapy compared to PCI and fibrinolysis. Conclusion: Process mining tools help to understand referencing networks visually, easily highlighting its inefficiencies and potential needs for improvement. A new PCI centre in the "Centro" region is critical to offer timely first-line treatment to their population.

Sequence-oriented sensitive analysis for PM2.5 exposure and risk assessment using interactive process mining.

The World Health Organization has estimated that air pollution will be one of the most significant challenges related to the environment in the following years, and air quality monitoring and climate change mitigation actions have been promoted due to the Paris Agreement because of their impact on mortality risk. Thus, generating a methodology that supports experts in making decisions based on exposure data, identifying exposure-related activities, and proposing mitigation scenarios is essential. In this context, the emergence of Interactive Process Mining-a discipline that has progressed in the last years in healthcare-could help to develop a methodology based on human knowledge. For this reason, we propose a new methodology for a sequence-oriented sensitive analysis to identify the best activities and parameters to offer a mitigation policy. This methodology is innovative in the following points: i) we present in this paper the first application of Interactive Process Mining pollution personal exposure mitigation; ii) our solution reduces the computation cost and time of the traditional sensitive analysis; iii) the methodology is human-oriented in the sense that the process should be done with the environmental expert; and iv) our solution has been tested with synthetic data to explore the viability before the move to physical exposure measurements, taking the city of Valencia as the use case, and overcoming the difficulty of performing exposure measurements. This dataset has been generated with a model that considers the city of Valencia's demographic and epidemiological statistics. We have demonstrated that the assessments done using sequence-oriented sensitive analysis can identify target activities. The proposed scenarios can improve the initial KPIs-in the best scenario; we reduce the population exposure by 18% and the relative risk by 12%. Consequently, our proposal could be used with real data in future steps, becoming an innovative point for air pollution mitigation and environmental improvement.

Building Process-Oriented Data Science Solutions for Real-World Healthcare.

The COVID-19 pandemic has highlighted some of the opportunities, problems and barriers facing the application of Artificial Intelligence to the medical domain. It is becoming increasingly important to determine how Artificial Intelligence will help healthcare providers understand and improve the daily practice of medicine. As a part of the Artificial Intelligence research field, the Process-Oriented Data Science community has been active in the analysis of this situation and in identifying current challenges and available solutions. We have identified a need to integrate the best efforts made by the community to ensure that promised improvements to care processes can be achieved in real healthcare. In this paper, we argue that it is necessary to provide appropriate tools to support medical experts and that frequent, interactive communication between medical experts and data miners is needed to co-create solutions. Process-Oriented Data Science, and specifically concrete techniques such as Process Mining, can offer an easy to manage set of tools for developing understandable and explainable Artificial Intelligence solutions. Process Mining offers tools, methods and a data driven approach that can involve medical experts in the process of co-discovering real-world evidence in an interactive way. It is time for Process-Oriented Data scientists to collaborate more closely with healthcare professionals to provide and build useful, understandable solutions that answer practical questions in daily practice. With a shared vision, we should be better prepared to meet the complex challenges that will shape the future of healthcare.

Analytical exploratory tool for healthcare professionals to monitor cancer patients' progress.

Introduction: Cancer is a primary public concern in the European continent. Due to the large case numbers and survival rates, a significant population is living with cancer needs. Consequently, health professionals must deal with complex treatment decision-making processes. In this context, a large quantity of data is collected during cancer care delivery. Once collected, these data are complex for health professionals to access to support clinical decision-making and performance review. There is a need for innovative tools that make clinical data more accessible to support cancer health professionals in these activities. Methods: Following a co-creation, an interactive approach thanks to the Interactive Process Mining paradigm, and data from a tertiary hospital, we developed an exploratory tool to present cancer patients' progress over time. Results: This work aims to collect and report the process of developing an exploratory analytical Interactive Process Mining tool with clinical relevance for healthcare professionals for monitoring cancer patients' care processes in the context of the LifeChamps project together with a graphical and navigable Process Indicator in the context of prostate cancer patients. Discussion: The tool presented includes Process Mining techniques to infer actual processes and present understandable results visually and navigable, looking for different types of patients, trajectories, and behaviors.