Complexity Technology in Medicine
Because of the immense complexity of the human body and due to the limitations of conventional mathematics, medicine lacks a systems perspective. Today, thanks to our Quantitative Complexity Science and Technology, it is possible to see a patient as a system of systems which interact and to provide real-time measures of his complexity and stability. This new approach provides unprecedented insight into the dynamics of a patient and enables us to issue early-warnings in an OR or an ICU. We offer a real-time engine which may be integrated into any IT system or device.
Our measures of a patient's complexity, stability or resilience are obtained in real-time not via a data-base look-up but by analyzing real-time streaming multi-channel data such as that available in an ICU or OR. Our system processes all channels as well as all channel interactions.
What is Complexity
Complexity is a natural and holistic property of every system. It is defined as a function of structure (topolgy of interactions between data channels) and uncertainty. Just like energy, complexity is a fundamental and intrinsic property of all dynamical systems: the economy, the society, the internet, the environment, traffic systems, etc. The human body is no exception. Today, complexity can be measured and turned into a powerful diagnostic and decision-making tool.
Complexity is a fundamental property of every system, just like energy. It can be measured, hence it can be managed. It is not a process. It is not about self-organization or swarms of birds. It is not order on the edge of chaos. Complexity is a physical dimension, an attribute of all systems, natural and man-made.
Monitoring and managing a patient's complexity requires a metric. Complexity is as a function of the structure of information flow within a system and entropy, which measures disorder as well as information content. Formally, complexity is defined as follows:
C = f(N; S; E)
where N is the number of variables used to describe a system (e.g. the number of data channels in an ICU), S reflects the structure of inter-dependencies between these variables, and E is entropy (which measures the level of 'disorder' in the data). Entropy is computed based on Shannon´s formulation and expresses in bits the amount of information. Based on our formulation, complexity measures the total amount of structured information in a system and its unit is the complexity bit or simply cbit.
The structure of interactions between multiple mdata channels - such those in an ICU or OR - is reflected by the Complexity Map. The map changes in times and indicates which channels dominate the current complexity of a patient (indicated by large squares on the map's diagonal).
How it Works
Complexity can be measured if data is available. A good example is the stream of data recorded in an ICU (Intensive Care Unit), OR (Operating Room) or during ECG or EEG tests. Even images, such as MRI or US scans can be processed to measure their complexity in order to detect structural changes. In the presence of traumas, organ malfunctions/failure or even treatment, the complexity of the human body changes. The magnitude of these changes is proportional to the intensity of trauma and provides a measure of the patient´s global stability. Based on this concept, we are able to issue instability pre-alarms in an ICU or during surgery.
An important feature of our technology is that the collected data is processed according to a model-free technique. This means that the measure of stability of the patient is not flawed by any particular mathematical approach. Once defined, a model is condemned to deliver only what has been hard-wired into it. This is why the adoption of a particular math model automatically influences the conclusions one will derive from the underlying data. In other words, models can distort the information that given data contains. Model-free means we do not build math models based on data extracted from medical equipment or real-time data. Data is precious and simplistic math models can destroy or warp the information it carries. This is why we have developed a proprietary approach which is data-centric and which does not rely on the conventional concept of a mathematical model.
Because modern science lacks a holistic perspective favoring super-specialization, a patient is rarely seen and treated as multi-organ dynamic system of systems. Due to this cultural limitation and because of the overwhelming complexity of the human body, only on very rare occasions is medical science quantitative. Our mission is to deliver a technology which will provide the medical community with quantitative and holistic information on the state of a patient, as well as on the impact of treatment.
Complexity combines into a single number the interactions of multiple data channels. This is why it is ideal when it comes to representing systems described via numerous variables with intricate inter-dependencies. With this approach we can deliver quantitative indices which combine in a rational manner multiple vital signs or markers.