The articles in this literature review were identified through systematic searches of PubMed, Medline, and Embase databases. All articles were published in the English language, between January 2000 to December 2021 and with an abstract. In this review study, we judge the evidence level of different the PaCa risk factors through the criteria of grading evidence for cancer prevention. This literature review summarizes the modifiable and non-modifiable risk factors of PaCa with strong evidence, which could be used to further establish PaCa predictive model as an application of PaCa risk stratification, raise public awareness, and educate the public as a prevention program. Further studies are needed to investigate other potential risk factors.

The post Risk Factors of Pancreatic Cancer: A Literature Review appeared first on iHELP Project.

In this study, we have developed an integrated PaCa risk prediction model for PaCa using data from the UK Biobank, incorporating lifestyle-related, genetic-related, and medical history-related variables for application in healthcare settings. We used a machine learning-based random forest approach and a traditional multivariable logistic regression method to develop a PaCa predictive model for different purposes. Additionally, we employed dynamic nomograms to visualize the probability of PaCa risk in the prediction model. The top five influential features in the random forest model were age, PRS, pancreatitis, DM, and smoking.

The significant risk variables in the logistic regression model included male gender (OR = 1.17), age (OR = 1.10), non-O blood type (OR = 1.29), higher polygenic score (PRS) (Q5 vs. Q1, OR = 2.03), smoking (OR = 1.82), alcohol consumption (OR = 1.27), pancreatitis (OR = 3.99), diabetes (DM) (OR = 2.57), and gallbladder-related disease (OR = 2.07). The area under the receiver operating curve (AUC) of the logistic regression model is 0.78. Internal validation and calibration performed well in both models. Our integrative PaCa risk prediction model with the PRS effectively stratifies individuals at future risk of PaCa, aiding targeted prevention efforts and supporting community-based cancer prevention initiatives.

The post An Integrative Pancreatic Cancer Risk Prediction Model in the UK Biobank appeared first on iHELP Project.

This study investigated the PaCa risk factors and the population-attributable fraction (PAF) of modifiable risk factors in the UK Biobank cohort. The UK Biobank is a prospective cohort consisting of 502,413 participants with a mean follow-up time of 8.2 years. A binomial generalized linear regression model was used to calculate relative risks for PaCa risk factors. PAF was calculated to estimate the proportional reduction in PaCa if modifiable risk factors were to be eliminated. A total of 728 (0.14%) PaCa incident cases and 412,922 (82.19%) non-PaCa controls were analyzed from the UK Biobank.

The non-modifiable risk factors included age and gender. The modifiable risk factors were cigarette smoking, overweight and obesity, increased waist circumstance, abdominal obesity, Diabetic Mellitus (DM), and pancreatitis history. The PAF suggested that eliminating smoking and obesity can contribute around a 16% reduction in PaCa cases while avoiding abdominal obesity can eliminate PaCa cases by 22%. Preventing pancreatitis and DM could potentially reduce PaCa cases by 1% and 6%, respectively. This study has identified modifiable and non-modifiable PaCa risk factors in the UK population. The PAF of modifiable risk factors can be applied to inform PaCa prevention programs.

The post Risk Factors Associated with Pancreatic Cancer in the UK Biobank Cohort appeared first on iHELP Project.

Our study aims to further develop and evaluate the i-Help risk assessment and support platform, which utilizes big data management and artificial intelligence (AI) approaches to establish person-centric early risk prediction, prevention, and intervention measures. We will present the design of the platform along with feedback from end-users on key modules. Ultimately, we propose that the outcomes of the iHelp components will not only integrate data from multiple sources but will also provide informative and motivational individualized pancreatic cancer risk reduction recommendations.

The post Development and application of the iHELP platform to facilitate the establishment of healthy habits for the prevention of pancreatic cancer appeared first on iHELP Project.

Our study aims to establish a pancreatic cancer risk prediction model by integrating the current risk factors with established genomic biomarkers. We will present the performance of the novel integrated model which will be useful for use in primary care to identify the at-risk population. This will assist in the promotion of prevention recommendations and for future health check follow-up.

The post A novel integrated predictive model for pancreatic cancer appeared first on iHELP Project.

This data is organized into the standardized Holistic Health Records (HHRs) format which is a novelty introduced by the iHelp project. Moreover, innovative, and AI-based decision support mechanisms for personalized risk detection, and prevention, as well as recommendation and intervention models, are provided. These tools are coupled with extensive research and SotA approaches from various domains such as clinical studies, Pancreatic Cancer research, Artificial Intelligence (AI), Big Data analytics, IoT, software engineering, social media analytics, and cloud technologies. In that respect, the iHelp solution provides enhanced personalized experiences and recommendations based on patient data through different devices and systems. The solution has been also validated and evaluated for its generalization and wider applicability in other types of cancer, and more specifically on the liver, anal, and prostate. The latter showcases its potential for wider adoption from healthcare stakeholders and is designed to be highly exploitable, by emphasizing its high usability and potential integration with different HIS and platforms, leading to the introduction of a European platform for enhanced personalized healthcare.

The primary exploitable result of iHelp is personalized healthcare provisioning through integrated data, user-oriented interfaces, prediction models, prevention and intervention strategies, engagement methodologies, and innovative business models. In that context, the project aims to deliver multi-level and user-centric solutions in research, healthcare, and business communities. These solutions include different AI Models for Personalized Health Prediction that are also integrated with an Analytic Workbench that serves as a repository for the deployment and execution of the implemented AI models. Moreover, a scalable Big Data Platform offers the mechanisms for the storage of integrated healthcare data into the common HHR format.

The implementation of an advanced Decision Support System (DSS) that incorporates the utilization of visual analytics and integrates with the Monitoring, Alerting, and Feedback component provides an improved solution to the HCPs for the overall monitoring and assessment of individuals’ health. Finally, the integration with the iHelp’s patient companion application and a Tailored Conversational Virtual Coaching System provides a near real-time integration between the HCP and a patient. By leveraging these mechanisms, iHelp aims to acquire and exploit domain knowledge, patient information, and insights into PC risks from diverse data sources through a holistic data management platform. This approach aims to advance scientific understanding and communicate project results to relevant stakeholders.

The target customers for the iHelp framework can be identified as HCPs, individuals, data scientists/developers, and healthcare policymakers. More specifically, HCPs benefit from personalized healthcare and risk assessment, leading to time and cost savings. The research, design, and implementation of innovative solutions for personalized healthcare make diagnostics smarter and more targeted, such as in the case of Pancreatic Cancer, where early identification and personalized treatments can help in the design of improved screening programs. The latter will lead to improvements in the monitoring, care, and QoL of the patients. The ability to identify which preventative measure and intervention is delivering the desired impact can massively help in the development of new diagnostic and treatment regimes. Individuals gain increased awareness, early risk assessment, and personalized interventions. Health is a very sensitive issue for any individual and the interrelationships between various health conditions make it paramount to raise awareness of risks and relevant interventions. Increased awareness, early risk assessment and personalized interventions and recommendations can reduce the risks of any type of cancer, improve health literacy and QoL by empowering individuals to take an active role in the management of their own life.  Moreover, direct and (near) real-time communication with HCPs is facilitated through the delivery of user-centric mobile and wearable applications, as well as the DSS.

Data Scientists/developers are another group of stakeholders that can benefit from the innovative approaches and standardized data modeling introduced in the context of the project. Innovative containerized and federated learning approaches are followed and introduced as SotA approaches to also overcome the ethical and legal issues that may arise from different stakeholders. Moreover, advanced AI-based models for early risk prediction and assessment are introduced and developed, while a standardized data modeling approach through the realization of HHRs is followed and introduced. The latter leads to the provision of solutions and mechanisms for enhanced information acquisition, interoperability, and aggregation. Finally, healthcare policymakers can utilize AI-based analytics for informed policy and decision-making related to healthcare interventions. The analysis of personalized risks and interventions can also inform the policy-making related to the design of new and effective screening programs. Moreover, the evaluation and optimization of targeted policies through AI-infused analytics/simulations and visualizations can inform evidence-based policy and decision-making. iHelp seeks to enhance patient care, improve the QoL, and contribute to the advancement of personalized healthcare for PC, as well as for other types of cancer. In that point, it would be useful to also present and state the feedback from the HCPs from four of our pilots who also act as early adopters of the iHelp solution to further highlight its wider impact and added value in the healthcare domain.

“The UK-based “UniMan” iHelp study uses advanced technologies, particularly in the field of genomics and epigenomics. Employing large-scale data, AI models have been developed by analyzing 100’s of thousands of genetic and epigenetic markers that play a role in cancer and other diseases. Using such analyses, our participant’s health span and future health risks can be predicted by combining genetic predisposition, lifestyle, and social factors.

The iHelp platform integrates these assessments and allows for an effective way of communicating personalized risk assessments to change behaviors. By tailoring risk estimation and communication at the individual level the iHelp approach is thereby demonstrating a significant impact on changing health-related behaviors. In conclusion, the iHelp approach lies at the rapidly evolving intersection of AI and personalized medicine and is showing great potential for improving the future of disease prevention and health promotion.”

Kenneth Muir – Professor of Epidemiology, University of Manchester

Artitaya (Li) Lophatananon – Senior Research Fellow, University of Manchester

Te-Min Ken – UniMan iHELP Medical lead, University of Manchester

---

“The iHelp platform impacts hospitals by offering AI-based tools for personalized health monitoring and decision support, focusing on pancreatic cancer. It enables early risk identification and management, integrates real-time clinical data for assessment, and supports the development of predictive models for patient care. This approach can lead to improved outcomes through tailored treatment plans and early detection strategies. The early identification and detection of pancreatic cancer through the iHelp platform is seen as potentially life-saving by our hospital’s clinicians, who hold a very positive outlook on this system.”

Prof. Dr. Shabbir Syed Abdul MD, MSc, PhD. – Professor of Artificial Intelligence & Digital Health

“iHelp platform can improve the healthcare providers’ knowledge by the comprehensive medical checklist that is guiding the medic through the required activities for early disease diagnosis and provides powerful tools for fast and easy triage of the patients at risk to develop one extremely lethal malignancy.

The early and efficient triage is a prerequisite for adequate, prompt, and focused prevention.

The platform could become an irreplaceable and valuable tool for constant, daily monitoring of the patient at risk health status, behavior, and results of the preventive program implementation, saving medical specialists time and efforts for numerous medical control examinations and advice that was proved during the Medical University of Plovdiv pilot.”

Prof Rostislav Kostadinov, MD, PhD, DSc – Professor at Department for Epidemiology and Disaster Medicine – Medical University of Plovdiv, Plovdiv, Bulgaria

Prof Dilyana Vicheva, MD, PhD – Professor ENT Clinic, University Hospital for Active Treatment “Kaspela”, Plovdiv, Bulgaria

Yancho Madzharov, MD – Outpatient clinic for primary medical care – individual practice „Petia Madzharova” Ltd, Parvomai, Bulgaria

---

“I have a very positive view on the utilization of iHelp, as its tools enable better clinical monitoring and effective assessment of patient progression, as well as facilitating rapid and direct contact with study participants. In our pilot, we have utilized the tools developed by iHelp with the biomedical data provided by our hospital to ensure their utility for us.”

Juan Ramón Berenguer Marí, Biotechnology faculty at the Hospital Dénia-Marina Salud

[1] C. Lee, Z. Luo, K.Y. Ngiam, M. Zhang, K. Zheng, G. Chen, B.C. Ooi, and W.L.J Yip, “Big Healthcare data analytics: Challenges and applications,” Handbook of large-scale distributed computing in smart healthcare, pp. 11-41, 2017.

[2] A. Dubey, and A. S.Verma, “Effective Remote Healthcare and Telemedicine Approaches for Improving Digital Healthcare Systems,” in Digital Health Transformation with Blockchain and Artificial Intelligence. CRC Press, 2022, pp. 273-297.

[3] I. Mlakar, V. Šafran, D. Hari, M. Rojc, G. Alankuş, R. Pérez Luna, and U. Ariöz, “Multilingual conversational systems to drive the collection of patient-reported outcomes and integration into clinical workflows”, Symmetry, vol. 13, no. 7, p. 1187, 2021

The post The iHelp platform and its impact in the modern healthcare domain appeared first on iHELP Project.

The 10th General Assembly (GA) meeting of the iHelp project was conducted during previous October (17-18/10/2023) in the beautiful city of Bucharest, Romania, thanks to the efforts and great organization from SIE partner. This meeting is highlighted as a crucial milestone in the ongoing implementation of the iHelp project as it introduced several software prototypes and set the roadmap for the further exploitation of its innovative items.

The project, dedicated to implementing an integrated AI-based platform for the early identification, prevention, and monitoring of pancreatic cancer, showcased its innovative tools through parallel workshops and demonstrations. The Decision Support System (DSS), the Monitoring and Alerting System, the Explainable Dashboard Hub (EDH), and the Virtual Coach tools played a leading role during the presentations and demonstrations, offering a glimpse into the potential future of pancreatic cancer care. Their overall impact, usability, and added-value in the modern healthcare care were demonstrated in the context of real-world scenarios presented by the pilots and healthcare professionals of the project.

In the Decision Support System (DSS) workshop, partners verified the platform’s ability to empower the healthcare professionals in their everyday activities, as well as to improve their knowledge through the visualization of individuals integrated primary and secondary data. The DSS, powered by AI algorithms, demonstrated its proficiency in near real-time data analysis for early cancer risk identification. Its potential to provide valuable insights for personalized patient care concentrated partners attention, positioning it as a critical innovation and exploitable tool of the project. Coupled with its integration with the Explainable Dashboard Hub (EDH) it also offers enhanced explainability and interpretability on the results and outcomes derived from the execution of advanced AI models. The latter positions the integrated DSS iHelp platform as a trusted and user-friendly system that can be offered in the healthcare community and fosters its future exploitation.

In addition, the DSS is integrated with the Monitoring and Alerting System that supports healthcare professionals in the crucial aspect of continuous monitoring and follow-up of their patients. Utilizing AI to analyze patient primary and secondary data, the system showcased its capability to detect changes in the lifestyle behaviors and habits of the patients and issue timely alerts based on specific rules and targets that can be set by the healthcare professionals. The enthusiasm surrounding this tool stems from its potential to enhance preventive measures, potentially shifting the paradigm from late-stage diagnoses to proactive intervention.

Finally, the Virtual Coach is another innovative tool of the iHelp platform and showcases its integration with mobile applications that can be used by the patients themselves, offering a near real-time communication and interaction between patients and healthcare professionals. The latter is further supported through the integration of this tool with the project’s Monitoring and Alerting System based on which dedicated messages are triggered based on the goals that have been set by the healthcare professionals. The utilization of this tool empowers citizens with valuable knowledge and tips with the goal to take an active role in their own health management by also enhancing their digital and health literacy. Partners highlighted its potential to encourage healthier lifestyles and adherence to preventive measures, marking a significant step toward personalized, empowering healthcare experiences.

As final outcomes from this GA meeting, the project outlined key directions for the future. These include the continuous refinement of AI algorithms to improve the accuracy of the DSS, fostering stronger collaborations with healthcare providers, integrating user feedback for an optimized experience, and ensuring global outreach and accessibility. The collaborative efforts showcased during the GA meeting are set to shape the iHelp project’s added-value in the healthcare domain, offering a promising holistic approach for the transformation of pancreatic cancer care through innovative AI solutions.

The post Insights from the 10th GA Meeting in Bucharest appeared first on iHELP Project.

Agenda & Partners’ Preparation

The meeting agenda outlined specific topics that were covered by WP Leaders, Pilots, and Task Leaders. Unfolding across three days, each day had a designated focus. The first day emphasized the detailed rehearsal of presentations, while the second day was dedicated to refining demonstration techniques. The concluding day consolidated all discussions into a comprehensive sum-up session, ultimately leading to the final wrap-up of the entire event.

The WP & Tasks Leaders presented the technical achievements within their Work Package (WP) and addressed any outstanding points leading up to the 2nd Project Review and the final semester of the project. Additionally, they shared insights on the upcoming steps until the project’s conclusion. A crucial focus of this gathering was to execute successful rehearsals and demonstrations, and deliberate on decisions regarding the update and revision of the implementation, integration, and demonstration of functionalities slated for February 2024.

Social dinner

At the conclusion of our initial day, we orchestrated a convivial social dinner, strategically designed to cultivate, and deepen the bonds within our team.

It was not just a good time; it highlighted the positive teamwork among the project partners. The relaxed atmosphere and shared moments reinforced the collaboration that is essential for our joint efforts.

Meeting Objectives

1. Elaborated on Work Items and Address Open Points: As the project advanced, the meeting aimed to delve into the specifics of work items and scenarios, clarifying any open points and addressing identified drawbacks. The focus was on revising components and scenarios for enhanced clarity.
2. Advanced Development and Integration Efforts: The primary objective was to propel the project forward through active development and integration work. Initial Rehearsals were conducted to prepare for the 2nd Project Review, with a keen eye on identifying and promptly addressing any gaps or challenges.
3. Refined Scenario Storylines: Progressing through the project’s timeline, a critical review of each scenario’s storyline was essential. Utilizing the iHelp platform in hospital settings was mandatory to assess the viability of pilot scenarios and understand the status of recruitment processes and overall pilot activities.
4. Updated Pilot Plans and Scenarios: Building on integration progress and RP2 goals, technical partners and pilots collaboratively determined the next steps and implementations. This collaborative effort aimed to support the revised scenarios and guide pilot activities toward the successful completion of the project’s objectives.

In pursuit of these refined objectives, the project sought a well-orchestrated conclusion marked by detailed planning, dynamic development, and strategic adaptations to ensure the realization of project goals.

The post Athens hosts 11th iHelp GA Meeting: Unveiling Progress and Preparing for Project’s Culmination appeared first on iHELP Project.

Addressing Patients’ and Healthcare Institutions’ Biggest Challenges

To implement prevention or treatment campaigns for Pancreatic Cancer diseases hospitals and patients are facing major issues that limit the impact and the effectiveness of healthcare support.

So far, it has been nearly impossible for all patients who are or must be subjects of health checks and treatment to get the attention and time needed for comprehensive analysis and personalized care. Thanks to the new platform, clinicians can now leverage a rule engine that is part of the platform and create personalized risk mitigation or prevention plans considering the patient’s specific condition, lifestyle habits, and health parameters.

On the other hand, patients are often reluctant to undergo specific treatment or demonstrate unexpected behavior during consultations. Some even have concerns about visiting a specialist and taking timely measures against dangerous diseases. The tool helps clinicians set up automated data monitoring, evaluation, and online personalized risk communication with patients so that they stay informed about their condition at any time.

There is another common issue addressed by the platform and related to common digital healthcare applications that offer a generic approach and user experience to everyone. Namely, the tool lets healthcare specialists tailor specific personalized messages to patients based on behavior, and thus keep them engaged with time.

An Innovation Breakthrough in Healthcare

The product is one of a few addressing early detection and prevention of Pancreatic Cancer diseases but stands out with its features and benefits such as:

• Remote monitoring and evaluation of targeted patient’s improvement and quality of life.
• Real-time patients’ engagement which is essential for the high adoption and usage rates of the platform.
• A big step towards following a personalized pancreatic cancer risk mitigation and prevention plan.

The Monitoring & Alerting component developed is part of a comprehensive web-based Decision Support System for clinicians, which can be utilized as a Single point of Information during pancreatic cancer treatment.

Figure 1.

Figure 2.

Figure 3.

Also, the automation of health-related data processing and evaluation enables clinicians to make decisions faster in critical situations during the treatment of patients. It is important to note that the component meets specific requirements in the context of ethical regulations where communication on the progress towards personalized targets is being reviewed by healthcare professionals before being sent to a patient’s mobile phone. Overall, the features of the tool also contribute not only to driving strong engagement but also to increasing the level of trust by patients in the innovative platform and the treatment they get.

Feedback from Healthcare Professionals

The tool has already been reviewed and reflected on by healthcare professionals in Bulgaria, part of the network of the Medical University of Plovdiv. Together with iHelp team representatives they have tested and discussed the following processes of using the application:

• iHelp platform login and identification
• Patient enrollment
• Editing patient information
• Patient information visualization
• Setting personalized targets for risk identification and risk mitigation
• Setting personalized recommendations and feedback based on targets
•  Personalized recommendation review and approval by healthcare professionals

Figure 4.

Overall, the healthcare professionals suggested potential new features to be planned and offered their views on some components that might further improve the work of their colleagues.

In addition, they acknowledged the huge effort put into the system and what has been achieved so far. In their opinion, the system has the potential to be extended to serve other types of cancer treatment and not be restricted to pancreatic cancer only. They look forward to seeing how the project develops further to cover additional processes and steps planned as part of the iHelp project.

“Innovations like the iHelp Monitoring & Alerting component for remote patient monitoring systems have the potential to revolutionize the healthcare industry. By leveraging the latest technological advancements, these systems can improve both the accessibility and quality of healthcare.

We are proud of our contribution to a solution for reducing healthcare costs and improving patient outcomes, which have a long-term positive impact on society and business.”, shares Pencho Stefanov, Project Lead at Kodar, part of Strypes Group.

The post iHelp Real-Time Monitoring & Alerting Tool: First Insights and Impact appeared first on iHELP Project.

Data collection can be performed through several techniques – such as data extraction from structured sources or text mining in the case of unstructured and textual sources – at least for clinical and primary data. Secondary data collection is performed in the prospective study and through a mobile app – Healthentia – that is downloaded and installed on the participant’s smartphone. Each patient can voluntarily answer questions and report other details on their daily symptoms or other parameters, such as food intake. Additionally, patients are asked to wear a smartwatch and to connect it to the app, to retrieve data on their heart rate, sleep duration, and physical activity.

Patients can be enrolled, monitored, and managed via a dedicated dashboard for HCPs, which is connected to the patients’ app developed by Innovation Sprint. Clinicians can check whether each participant is answering questionnaires and whether the wearable is correctly synced and connected. Questionnaires’ responses, as well as data deriving from the smartwatch, are also retrieved, and reported by the iHelp Decision Support System (DSS). This framework, along with its dependencies, can be used through a central node in case the hospital cannot host a dedicated server or, as in Fondazione Policlinico Universitario Agostino Gemelli IRCCS (FPG), it is deployed in a dedicated edge node or cluster. AI techniques to draw adaptive learning models for early risk predictions as well as personalized prevention and intervention measures, are made available to the consortium partners.

Other than the tools above mentioned, FPG has developed a dashboard for monitoring the number of patients who decided to take part in the study at each pilot site. Details on the overall number of patients that joined the project, as well as the distribution of the enrolments, give insights e.g. on the interest of end-users in using the iHelp tools.

At FPG, HCPs are currently using the Healthentia dashboard and are working with technical partners to finalize the deployment of all the dependencies of the iHelp framework. Clinicians can make use of the iHelp DSS to view patients’ data and check whether any answer or parameter is higher or lower than set thresholds, and can eventually decide to analyze a situation following the usual care guidelines and best practices – in the FPG pilot, no automatic suggestions or interventions are sent to patients. Such functionalities were discussed with the technical partners and several personalized indicators and indicators were added to the DSS that should be installed in the FPG premises. Additionally, the presence of a cluster of virtual machines hosted in the hospitals’ premises guarantees that patients’ clinical data do not leave the clinic and can be shown to HCPs via the locally installed dedicated tools.

Thanks to the flexibility and generalizability of the decision support suite developed in iHelp, FPG’s clinicians proposed to extend the use of the tools to other use cases and asked to include two additional pathologies to the study and the project: anal canal and prostate cancer. The chance of extending the iHelp study to three different pathologies made clear that the tools developed in the scope of the project can be easily adapted to other use cases and, other than being accepted by clinicians, can be effectively and actively used in clinical practice. Such endorsement and interest are also proof that the framework, the methodology, and the services developed in this project, as well as results, can be offered in the future to a wider audience of stakeholders, e.g. hospitals, clinics, specialists, or other care facilities and patients. 

The post The iHelp framework in the FPG pilot – Personalization and deployment of the iHelp tools for three different pathologies appeared first on iHELP Project.