Search

System Design and Arrhythmia

Why should we care about system design and arrhythmia? Because the symptoms of arrhythmia, or an abnormal heart rhythm, are incredibly varied (in manifestation and intensity – some individuals never experience symptoms). In some individuals the heart beats too fast, and in others, too slow. Thus, early diagnosis is key, because arrhythmia can lead to death.[1] This paper describes the “systematic design of an intelligent classification system for decision support in Holter ECG monitoring” that will “detect the possibility of arrhythmia in real time.”[2]


This intelligent system is fed by the Holter monitoring device, which continuously and remotely collects long-term Electrocardiographic (ECG) information, and it can be integrated with smartphones. The essence of the system follows:


1. The patient wears an ECG monitor.


2. It detects heartbeats and captures data (“We detected heartbeats and extracted features such as the QRS complex and P wave from ECG signals using the Pan–Tompkins algorithm…”[2])


3. The detected heartbeats are classified by a decision support system (“into 16 types using a decision tree.”[2]) This system (or subsystem) uses the patient’s long-term ECG information (collected in a database) to construct a decision tree with applied expert feedback and domain knowledge. The Pan–Tompkins algorithm is applied to extract heartbeat features, which in turn enables the classification.[2]


4. The results of the analyses are sent to a monitoring station or a monitoring professional via their smartphone enabling real-time diagnosis.


5. If the professional determines actions are required, they can contact the patient via their smartphone and both parties, the medical professional and the patient, begin the treatment process in parallel (concurrently).


This is an example of an Automated Decision System (ADS), described by our textbook as “a rule-based system that provides a solution, usually in one functional area, to a specific repetitive problem.”[3] The solutions can be presented to humans who make the final decision, as in the case of this intelligent ECG monitoring system.


This might be considered to be a combination of ADS with Expert System (ES), but a foundational aspect of an ES is contradicted by the larger system. Symbolic reasoning is used in ES instead of mathematical calculation, and the intelligent ECG monitoring system clearly uses the Pan–Tompkins algorithm applied to continuous data from the patient’s heart to support the classification process. However, the decision support system may be considered a subsystem in that the construct of the decision tree requires the use of a knowledge base (expert feedback and domain knowledge), and this is the foundation of an ES.


The collaboration between the medical professionals and the patient is actually interactive, albeit data initiated and a feedback loop completed with either silence (no problem) or direction (problem). The actions taken based on the data transform the system into one that includes knowledge management as well.


A popular example of an ES in the medical field is IBM’s Watson. “Watson is already capable of storing far more medical information than doctors, and unlike humans, its decisions are all evidence-based and free of cognitive biases and overconfidence. It's also capable of understanding natural language, generating hypotheses, evaluating the strength of those hypotheses, and learning — not just storing data, but finding meaning in it.”[4] Watson represents an accurate, consistent, current, and low-cost solution for medical diagnosis, but it does not remove the human doctor from the equation. Basically, the doctor meets with the patient and gains an understanding of the symptoms. The doctor enters this information into Watson and provides recommended treatment options along with corresponding confidence ratings. It is currently up and running at the MD Anderson Cancer Center and Wellpoint. The primary implementation challenge for Watson is developing the ability to leverage the vast amounts of information at its disposal. It has the “what”, but not the “so what”. Not yet, anyways.


References:

1. What is an Arrhythmia? Retrieved from: https://www.heart.org/idc/groups/heart-public/@wcm/@hcm/documents/downloadable/ucm_300290.pdf

2. Park, J and Kyungtae Kang. Intelligent Classification of Heartbeats for Automated Real-Time ECG Monitoring. Retrieved from: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4270110/

3. Sharda, R., Delen, D., and Turban, E. (2015). Business Intelligence and Analytics: Systems for Decision Support. Boston: Pearson

4. Friedman, Lauren. IBM's Watson Supercomputer May Soon Be the Best Doctor in The World. Retrieved from: http://www.businessinsider.com/ibms-watson-may-soon-be-the-best-doctor-in-the-world-2014-4


0 views

Recent Posts

See All

Abstract This research finds that ineffective or absent data collection processes can be destructive to an organization in three ways. First, the use of nonrepresentative or anecdotal sampling is a so

An agent-based simulation model supports “complex decision systems where a system or network is modeled as a set of autonomous decision-making units called agents that individually evaluate their situ

Basic and Applied Social Psychology (BASP) has banned the use of p-values and confidence intervals in articles they publish. The publication “will require strong descriptive statistics, including effe