Systems patientomics: the virtual in-silico patient

Abstract The integration of clinical and molecular sciences with advanced engineering sciences is moving the world towards a new generation of life science where physiological and pathological information from the living human body can be quantitatively described in silico via biocomputing across multiple scales of time and size and through diverse hierarchies of organization – from molecules to cells and organs to individuals. The development of the Virtual Patient will change conventional medicine, which has been based upon experience and expectation, into “predictive medicine” that will have the capacity to develop solutions based upon prior understanding of the dynamic mechanisms and the quantitative logic of human physiology. Drug discovery, medical and welfare apparatus, and clinical trials in silico will improve the development of products with higher efficiency, reliability and safety while reducing cost. They will also impact upon knowledge-intensive industries. Such program aims at playing a key role in this new area, by sharing and generating solutions as well as human resources contributing to establishment of “ in silico medicine” as a basis of the predictive medicine within an international framework. In the long-term, computational physiological models will be refined, linked and validated until they are capable of providing essential predictions to clinicians when healthcare decisions need to be made. As the amount of data reinforcing the models grows, predictions will become more and more patient-centred, with models migrating from statistical, average models to physiological and mechanistic models informed by the unique characteristics of the patient. Systems Patientomics will propose new ways of combining this rich patient information space in a highly visual, coherent, meaningful way and of generating new clinical information by blending and fusing existing information, ultimately creating a “Patient Avatar” capable of supporting the medical professional by producing new clinical knowledge emerging from the integration of patient- and population-specific information. Focal points • Benchside Quite a few experimental biologists, functional and statistical genomics researchers, involved in developing new measurement technology for biology, and even molecular systems biologists, feel that that computational methods are not relevant for their own research goals. For the lion’s share of those cases where these research goals are rationalized by their potential value for predictive, preventive and participatory medicine this is a misconception. • Bedside Systems computational approaches should be a routine part of the clinical arsenal for the diagnosis, planning and executing of therapeutic interventions. This must include the incorporation of relevant training in medical school curricula. • Industry Collaborations to be developed across the breadth of the stakeholder groups involved in public health, from public health providers and patient groups, to researchers, scientists, funders and industry. • Community Development of an ethical and legal framework (establish common rules and principles for data acquisition/sharing/integration/reduction to practice, define how to achieve patient consent, to respond to fears of misuse of provided data, define solutions for data protection/open innovation). • Governments Governments should faciliate European EHRs adoption for highets quality of patient care. • Regulatory agencies A bypass fast track route for mobile medical apps can be adopted by the FDA authority by working in close relation with app developers, physicians and healthcare givers promote high standards.


Governments
Governments should faciliate European EHRs adoption for highets quality of patient care.

Regulatory agencies
A bypass fast track route for mobile medical apps can be adopted by the FDA authority by working in close relation with app developers, physicians and healthcare givers promote high standards.
& 2014 European Society for Translational Medicine. Published by Elsevier Ltd. All rights reserved.
"A picture is worth a thousand words"

Arthur Brisbane 1911
Today's disease diagnoses are partially decision made by patients' medical histories and partially by symptoms (although patients are often bad at communicating what's really going on with their health). A Johns Hopkins study found that as many as 40,500 patients die in an ICU in the U.S. each year due to misdiagnosis, rivaling the number of deaths from breast cancer [1]. Yet another study found that 'system-related factors', e.g. poor processes, teamwork, and communication, were involved in 65% of studied diagnostic error cases. 'Cognitive factors' were involved in 75%, with 'premature closure' (sticking with the initial diagnosis and ignoring reasonable alternatives) as the most common cause. These types of diagnostic errors also add to rising healthcare expenditures, costing 300,000 USD per malpractice claim [2].
To avoid that healthcare should become more about data-driven deduction and less about trial-and-error. That is hard to pull off without technology, because of the increasing amount of data and research available. Next-generation medicine will utilize more complex models of physiology, and more sensor data than a human MD could comprehend, to suggest personalized diagnosis. Thousands of baseline and multi-omic data points and more integrative history will inform each diagnosis. Ever-improving dialog manager systems will help make data capture and exploration from patients more accurate and comprehensive. Data science will be the key to this. In the end, it will reduce costs, reduce physician workloads, and improve patient care.

Electronic health records (EHRs) -a starting point
Providing interoperability between different clinical systems, across national boundaries, and integration of clinical systems and research systems lies at the heart of the European Institute for Health Records (EuroRec) and several EU projects, e.g. TransFoRm (Translational Research and Patient Safety in Europe). Significant advances in international standards and in computational technology made possible support of interoperability [3]. Advances in the understanding of clinical judgment and decision making, and the possible ways of supporting them via ICT can inform the design of more 'intelligent' electronic health record systems.
The single richest source of routine healthcare data lies within the records of Europe's General Practitioners. Primary Care is responsible for first contact, continuing, and generalist care of the entire population from birth to death. Any project that aims to comprehensively support the integration of clinical and research data should begin with Primary Care [4]. In addition, even in countries where General Practitioners do not fulfill a 'gatekeeper' function, controlling access to specialist services, the quality of initial diagnosis at Primary Care level determines much of the future course for an individual patient with a health problem. In order to support patient safety in both clinical and research settings, significant ICT challenges need to be overcome in the areas of interoperability, common standards for data integration, data presentation, recording, scalability, and security.

2.
Integrating omics data with patient records to improve clinical outcomesa big data approach EHRs are an important part of a much larger puzzle. It is now widely accepted that big data can significantly transform health care delivery and improve outcomes for patients everywhere. The "Policy Forum on the Use of Big Data in Health Care" states: "Big data has the power to transform lives. In health care it can reveal the factors that influence health, help target appropriate care for individuals or populations, enable new discoveries, shape outcomes, and reduce costs" [5].
Additionally, the authors of the report, "Embracing the Complexity Of Genomic Data for Personalized Medicine," point out that "Numerous recent studies have demonstrated the use of genomic data, particularly gene expression signatures, as clinical prognostic factors in cancer and other complex diseases" [6]. Big data, in this context, refers to all of the types and varieties of data that are now available and can be integrated with patient information (both health care related and non-health care related) to provide more targeted and effective care and treatment, also known as personalized medicine. Among the data now available is data that is referred to as omic data. In health care, the more popular types of omic data currently being studied include genomics and eventually proteomics. However, several other omics disciplines need attention and they will soon enter mainstream point-to-care testing, as well as multiomics integration with imaging and biosensor data will enable full scope of the big data picture (Fig. 1).
3. Quantified Selfomics: from wearable sensors to mobile medical applications "2014 will be the year the 'quantified self' goes mainstream." Those were the words Silicon Valley prodigy Marc Andreessen used in a recent article to describe changes about to happen to American healthcare.
The 'quantified self,' also known as lifelogging [6], is a trend toward gathering all possible data about our daily life, such as the food we eat, quality of the air we inhale, our mood, oxygen levels, as well as our physical and mental performance (Fig. 2).
This movement combines smart devices and the Internet of Things with health monitoring apps to give us a better idea of how to optimize virtually every metric associated with our lifestyle, health, and physical performance.
On September 25, 2013, The FDA issued the Mobile Medical Applications Guidance for Industry [7], which explains the agency's oversight of mobile medical apps as devices. These mobile apps fall into the following categories: 1. Mobile platform such as a smartphone transformed into a device already requiring FDA approval, e.g. iPhone into an electrocardiography (ECG) machine. 2. Apps used as an accessory to a regulated device. 3. Mobile medical apps that perform patient-specific analysis and then provide a diagnosis, or treatment recommendations, such as a dosage plan for radiation therapy.
While there are about 17,000 available medical apps in the market, only a few have obtained FDA approval and certain concerns about preventing innovation have been raised.

Biosimulationroadmap to digital patient
Physiological mechanistic models range from smaller targeted models to answer specific R&D questions to whole-disease models that represent the complete pathophysiology of a specific disease. Success stories include the Virtual Physiological Human (VPH) and several other systems medicine projects. As part of VPH the DISCIPULUS project, developed an R&D roadmap for the development of the "Digital Patient". Its vision is to identify key steps towards realising the Digital Patient, which is a new paradigm in personalized medicine, across the whole healthcare system by focusing on the needs of clinical practitioners and healthcare professionals, biomedical and clinical researchers [8,9]. With the increasing size and complexity of any model, the opportunity arises for therapeutic groups to improve their understanding and insights into the variability and uncertainty characteristic of living systems and populations. Importantly, as pharmaceutical companies focus on precision medicines to deliver highly efficacious lowrisk therapeutics for subpopulations, these modeling approaches open new opportunities for improving our understanding of population uncertainty and variability for success.
Physiological modeling starts with the smaller targeted models to answer specific questions, and the generation of virtual cohorts that explore the hypothesis space around these questions [10]. Typically these models and cohorts are useful for discovery and preclinical investigations where scientists can understand the mechanism of action, understand the potential range of clinical responses, and gain insight into biomarkers, dosing, and critical measurements.
Having invested in smaller, more targeted models with a limited set of virtual patients, the next step of the investment journey is to expand the model with greater depth and breadth including patient registry, as well as to generate virtual populations, which are essential for precision medicine (Fig. 3).
These added investments provide valuable contributions for later stage development where an understanding of the trial population response ahead of time is critical to defining the right clinical trial and patient population: Identification of responders and non-responders within a population.
Identification of biomarker panels for drug effect in subpopulations.

Virtualization: a way to digitally enabled precision medicine
The use of virtualization and cloud computing is growing quickly among industries of all sizes. A roadmap initiative Avicena: A strategy for in silico Clinical Trails, began in October 2013 and is designed to cultivate consensus among a broad range of stakeholders who will come together to agree a research and technological development strategy. In order to add to scientific and medical progress, several pharmaceutical companies are engaged in a program to make clinical study data and other clinical study related documents more widely accessible for approved products or after termination of a drug development program. Researchers will be granted access via a secure analysis environment after approval of his/her study proposal by an   independent external review panel and based on a data sharing agreement. To quote the newly launched Nature Publishing Group journal Scientific Data's editorial: "The question is no longer whether research data should be shared, but how to make effective data sharing a common and well-rewarded part of research culture" [11]. Biomedicine is undergoing a revolution driven by high throughput and connective computing that is transforming medical research and practice. Using oncology as an example, the speed and capacity of genomic sequencing technologies is advancing the utility of individual genetic profiles for anticipating risk and targeting therapeutics. The goal is to enable an era of "P4" medicine that will become increasingly more predictive, personalized, preemptive, and participative over time (Fig. 4). This vision hinges on leveraging potentially innovative and disruptive technologies in medicine to accelerate discovery and to reorient clinical practice for patient-centered care.

Conclusion
The rapid growth in biological databases; models of cells, tissues, and organs; and the development of powerful computing software and algorithms have made it possible to explore functionality in a quantitative manner all the way from the level of genes to the physiological function of whole organs and regulatory systems. This review highlights the current and future possibilities, as well as pin points potential challenges, such as regulatory and open data sharing. Systems patientomics of the 21st century is set to become highly quantitative and, therefore, one of the most computer-intensive disciplines.

Executive summary
Next-generation medicine will include broad use of new technologies and computational approaches enabling virtual in silico biosimulation of physiological processes and disease conditions. Integrated analysis, which will require extensive bioinformatics and biostatistics support, named as systems medicine, has the potential to interrogate these molecular phenotypes and identify patterns associated with the disease. Implementation of Electronic Health Records across Europe is a starting point, which will need international standards and interoperability.
Emerging technologies, such as biosensors and mobile medical applications will be added within the big data analytical toolbox.
Virtual biosimulation, digital patients and populations will reduce R&D and medical practice costs and will enable precision medicine targeting each individual case.